DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF GHANA LEGON PSYCHOLOGICAL EFFECTS OF YOUTH UNEMPLOYMENT IN GHANA: A CASE STUDY OF THE GREATER ACCRA REGION CHRISTOPHER MARTIN AMISSAH (10242324) THIS THESIS IS SUBMITTED TO THE DEPARTMENT OF PSYCHOLOGY OF UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILMENT FOR THE AWARD OF MASTER OF PHILOSOPHY (MPHIL) DEGREE IN SOCIAL PSYCHOLOGY MAY, 2016 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I, Christopher Martin Amissah, do hereby declare that this thesis is the result of my research carried out in the Department of Psychology, University of Ghana, Legon under the effective supervision of Dr. Maxwell Asumeng and Dr. Kingsley Nyarko who are both my lecturers and supervisors. I further declare that all research works cited in this study have been duly acknowledged. __________________________________ CHRISTOPHER MARTIN AMISSAH (STUDENT) DATE ______________________________ This thesis has been submitted for examination with the approval of: __________________________________ DR. MAXWELL ASUMENG (PRINCIPAL SUPERVISOR) DATE______________________________ __________________________________ DR. KINGSLEY NYARKO (CO-SUPERVISOR) DATE______________________________ University of Ghana http://ugspace.ug.edu.gh ii DEDICATION I dedicate this thesis to my family. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENTS I am highly grateful to God Almighty for bringing me this far in my university education. I thank Him for His countless favors and blessings which have been very instrumental in my academic accomplishments. I am also highly indebted to Dr. Maxwell Asumeng and Dr. Kingsley Nyarko for their sacrifice of time and effort in guiding me through this thesis. I am really appreciative to them for their constructive criticisms and directions which have tremendously improved my knowledge and understanding in research. May God bless you abundantly for your good works. I also wish to acknowledge my good friend Mr. Prince Addae for his constructive suggestions that helped to shape my thoughts on this research. In addition, I would like to thank my research assistants Mr. Joshua Adjei, Nelly Betty Fosu, Benedicta Addae, and Shadrach Addae for their crucial role in the data collection and entry. Their help to me is very much appreciated. Moreover, I am grateful to all the respondents who spent time to respond to my research questionnaires. Their cooperation has been tremendously beneficial for the completion of this project work. I am indeed grateful to them for their sacrifice of time and efforts. Finally, I am appreciative to all individuals who have, in diverse ways, contributed to the success of my university education. God richly bless you all. University of Ghana http://ugspace.ug.edu.gh iv ABSTRACT This study investigated the psychological effects of youth unemployment in Ghana and the buffering role of religiosity and social support. Youths within the ages of 18 and 35 years in the Greater Accra Region of Ghana constituted the research population. A sample of 362 youths were purposively selected for the study. They comprised both the employed (n=172) and the unemployed (n=190). The employed youths served as a control group for comparative analyses. The cross- sectional survey research design was adopted. Standardized measures were used to assess psychological health in terms of depression, cognitive distortions, self-esteem, and suicidality. The Pearson r test, the Linear Regression test, and the Multivariate Analysis of Variance (MANOVA) test were used to analyze the data. The findings showed poorer psychological health among unemployed youths than employed youths. Duration of unemployment significantly predicted poorer psychological health among the youths. Religiosity moderated the psychological effects of youth unemployment except for depression. However, social support predicted but did not moderate the psychological effects of youth unemployment. The findings and their implications are discussed with references to the existing literature and theories. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENTS CONTENT PAGE DECLARATION i DEDICATION ii ACKNOWLEDGEMENT iii ABSTRACT iv TABLE OF CONTENTS v CHAPTER ONE: INTRODUCTION 1 1.1. Background of the study 1 1.2. Statement of the problem 8 1.3. Aims and objectives 9 1.4. Relevance of the study 9 CHAPTER TWO: LITERATURE REVIEW 11 2.1. Theoretical framework 11 2.2. Review of related studies 16 2.3. Rationale for the present study 34 2.4. Key variables 36 2.5. Statement of hypotheses 36 2.6. Operational definitions 36 2.7. Hypothesized conceptual model 37 University of Ghana http://ugspace.ug.edu.gh vi CHAPTER THREE: METHODOLOGY 38 3.1. Population 38 3.2. Sample 38 3.3. Sampling technique 40 3.4. Design 41 3.5. Measures 41 3.6. Inclusion and exclusion criteria 47 3.7. Procedure 48 CHAPTER FOUR: RESULTS 52 4.1. Statistical test for data analysis 52 4.2. Preliminary analyses 53 4.3. Testing of research hypotheses 56 4.4. Summary of findings 70 4.5. Observed conceptual model 70 CHAPTER FIVE: DISCUSSION 72 5.1. Discussion 72 5.2. Limitations of the study 82 5.3. Recommendations 84 5.4. Conclusion 88 University of Ghana http://ugspace.ug.edu.gh vii REFERENCES 89 APPENDICES 98 Appendix I: Questionnaire 98 Appendix II: Consent Form 108 Appendix III: Institutional Approval 110 University of Ghana http://ugspace.ug.edu.gh viii LIST OF TABLES Table Page Table 1: Demographic characteristics of the respondents 39 Table 2: Reliability analyses of scales and subscales in pilot study 49 Table 3: Reliability analyses of scales and subscales in the main study 54 Table 4: Descriptive statistics on key variables 55 Table 5: Correlations among key variables of the study 56 Table 6: Effect of youth unemployment on psychological health 57 Table 7: Effect of youth unemployment duration on depression 58 Table 8: Effect of youth unemployment duration on cognitive distortions 59 Table 9: Effect of youth unemployment duration on self-esteem 59 Table 10: Effect of youth unemployment duration on suicidality 60 Table 11: The moderating effect of religiosity for the relationship between youth unemployment and depression 62 Table 12: The moderating effect of religiosity for the relationship between youth unemployment and cognitive distortions 63 Table 13: The moderating effect of religiosity for the relationship between youth unemployment and self-esteem 64 Table 14: The moderating effect of religiosity for the relationship between youth unemployment and suicidality 65 Table 15: The moderating effect of social support for the relationship between youth unemployment and depression 66 Table 16: The moderating effect of social support for the relationship between youth unemployment and cognitive distortions 67 Table 17: The moderating effect of social support for the relationship between youth unemployment and self-esteem 68 Table 18: The moderating effect of social support for the relationship between youth unemployment and suicidality 69 University of Ghana http://ugspace.ug.edu.gh ix LIST OF FIGURES Figure Page Figure 1: Hypothesized model for the relationship between unemployment, religiosity, social support, and psychological health 37 Figure 2: Path diagram for moderation model 61 Figure 3: Observed model on the psychological effects of youth unemployment and the moderating role of religiosity 70 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.1. Background of the Study Young people constitute a major source of human capital-base that hold and drive the socio- cultural, economic as well as political development of their countries all over the world. The popular cliché that ‘youth are the future leaders’ raises an important question as to the quality of investments being made in young people to prepare them for their role as leaders. Mainstreaming youth is the surest way of achieving effective youth development. Their intellectual abilities coupled with their productive acumen when properly harnessed underpin societal progress. Ghana's present population consists of about 57 percent youths (Ghana Statistical Service (GSS), 2012). This means Ghana has a youthful population structure. Over the past fifty years, the number of youths in the total population has increased from 1.1 million in 1960 to 2.3 million in 1984, to 3.5 million in 2000 (Amankrah, 2005) and to 13.7 million in 2010 (GSS, 2012). Youths constitute about 22.6 percent of the economically active population (GSS, 2012). The share of the youth in the Ghanaian population is also reflected in a corresponding share of unemployment. About 60 percent of the unemployed in Ghana can be found in the youthful age group (Amankrah, 2005). This makes Ghana’s youth unemployment rate one of the highest in the world. According to Amankrah (2005), the proportion of the unemployed youth is higher for females than for males over the past fifty years. Youth unemployment is largely concentrated in the urban areas. Unemployment rate is as high as 30.8 percent in Accra, compared to 11.5 percent in the rural areas and 23.5 percent in other urban areas (Amankrah, 2005). University of Ghana http://ugspace.ug.edu.gh 2 Unemployment (or joblessness) is a situation whereby an individual is unable to find a paid job. According to the International Labor Organization (ILO, 2007), unemployment occurs when people are without work and actively seeking work. Unemployment has a number of negative consequences on those who are affected (Lorenzini & Giugni, 2010). Being deprived of a paid job entails a risk of social exclusion and isolation (Paugam, 2004) but potentially also has a number of negative consequences on personal life and well-being. This is especially true when unemployment is sustained in time. Dooley, Prause and Ham-Rowbottom (2000) argued against simply comparing the unemployed with the employed, because paid jobs can differ greatly in quality. Poorer quality jobs are more likely to be associated with mental health problems than better quality jobs. This point implies that comparing all employment with the unemployed will underestimate the effects of losing a good job. Dooley et al. (2000) addressed this issue by creating a category called “inadequate employment.” It was defined to include involuntary part-time work as well as work for very low wages. Low wages were defined with reference to the poverty threshold for single individuals in the U.S. If weekly earnings were not at least 25 per cent higher than the poverty threshold, wages were considered low. Dooley et al. (2000) defined the unemployed to include discouraged workers, but not other forms of inactivity. They observed that becoming unemployed, inadequately employed, or inactive were all significantly associated with increased depression compared with those remaining adequately employed. University of Ghana http://ugspace.ug.edu.gh 3 1.1.1. Unemployment and psychological health In recent years, psychological well-being has been the focus of intense research attention. Psychological well-being resides within the experiences of the individual (Campbell, Convers, & Rodgers, 1976). It may be defined as the state of feeling healthy and happy, having satisfaction, relaxation, pleasure and peace of mind. It deals with people’s feelings about everyday experiences in life activities. Such feelings may range from negative mental states or psychological strains, such as anxiety, depression, distress, frustration, emotional exhaustion, unhappiness and dissatisfaction, to a state which has been identified as positive mental health. Unemployment research over past decades has shown quite convincingly that unemployment leads to psychological distress and that re-employment improves mental health (Feather, 1990; Ezzy, 1993; Winefield, 1995). In other words, there seems to be a causal relationship between unemployment and poor mental health; the so-called causation hypothesis. However, Schaufeli (1997) made two critical notes. First, the strength of the relationship between unemployment and psychological distress is rather weak. Roughly speaking only 10 to 15% of the variance in distress is explained by employment status (Fryer & Payne, 1986). Second, the fact that unemployment causes psychological distress does not rule out the possibility that high levels of distress might lead to prolonged unemployment. For instance, in a large Australian sample of intended school-leavers, Winefield and Tiggemann (1985) found that depressed mood and poor self-esteem were antecedents as well as effects of unemployment. Hence, poor mental health might cause failure in the labour market. This is also known as reversed causation or the selection hypothesis. The causation hypothesis deals with the negative impact of unemployment on mental health. The selection hypothesis refers to the factors that determine success and failure in the labour market. The University of Ghana http://ugspace.ug.edu.gh 4 former is the focus of the present study. However, Schaufeli (1997) guessed that unemployed’s level of mental health might be one of the selective factors. Studies on the relationship between unemployment and mental health indicate that unemployment adversely affects mental health (Murphy & Athanasou, 1999). The classical study by Jahoda et al. (1933) shows that long-term unemployment upset the whole life of the affected, including a destructuring of their time. Unemployment is related to ill health, including poor mental and physical health, as well as unfavorable health habits (McKee-Ryan, Song, Wanberg, & Kinciki, 2005; Cohen, Kemeny, Zegans, Johnson, Kearney, & Stites, 2007; Paul, 2005; Paul & Moser, 2009). Unemployed individuals tend to have more health problems than employed individuals, but social support is known to facilitate coping with unemployment (Bjarnason & Sigurdadottir, 2003; Kroll, & Lampert, 2009). However, the social networks of unemployed men and women are found to consist of a higher concentration of unemployed members (Russell, 2009). Having unemployed friends may lead to alienation from mainstream values and negative changes in behavior (Russell, 2009). Suicidality may be counted among the negative consequences of unemployment. In Ghana, unemployment has been identified in qualitative studies as one of the causal factors of high suicidal ideation (Adinkrah, 2014; Osafo et al., 2011; Knizek, Akotia, & Hjelmeland, 2010). Of existing studies looking at correlates of suicide risk, many have identified a variety of situational factors (Villanueva, & Braddy, 2009; Pen˜ a et al., 2008; Rew, Thomas, Horner, Resnick, & Beuhring, 2001; Wadsworth & Kubrin, 2007). Negative life events uniquely contribute to the prediction of suicide risk (Chang, Sanna, Hirsch, & Jeglic, 2010). Among these negative life events or situational factors is unemployment. University of Ghana http://ugspace.ug.edu.gh 5 1.1.2. Buffering effects of social support and religiosity Psychological resources are very critical to healthy lifestyle among every individual. However, they are exceptionally relevant to people who are already experiencing negative life events. Negative life events such as unemployment are implicated in psychological disturbances that work to reduce psychological wellbeing (Spinhoven, Elzinga, Hovens, Roelofs, van Oppen, & Zitman, 2011). When experiencing negative life events, individuals use multiple coping styles and resources. Differences in how people respond and adapt to negaive life events can be accounted for by a number of psycho-social factors such as social support and religiosity. Thoits (1995) defined social support as “the functions performed for the individuals by significant others, such as family, friends, and coworkers” (p. 64). These significant others provide instrumental, informational, or emotional assistance. The term ‘social support’ may also be used to refer to various aspects of social relationships (Schwartz & Frohner, 2005). First, social support may be defined in terms of the quantity of social relationships and as such in terms of integration versus isolation. Second, the reciprocal aspects of social support may be expressed in terms of the structure of a person’s social relationships; in that case the term ‘social network’ is often used. Finally, social support is most commonly defined as the qualitative content of relationships, such as the degree to which the social relationships provide emotional concern, understanding, caring, instrumental and practical aid, information and the like. Although these three perspectives on social support need to be distinguished, they are closely intertwined: social relationships must exist in some quantity before they can have a structure and supportive function (Schwartz & Frohner, 2005). Studies have shown that social support is beneficial to health while facing stressful events, although they cannot prevent all damaging effects (Burns, Anstey, & Windsor, 2011; Lakey & Cohen, 2000; Kaul & Lakey, 2003; Gjesfjeld, Greeno, Kim, & Anderson, 2010). University of Ghana http://ugspace.ug.edu.gh 6 Researchers generally believe that the availability of social support plays an important role in influencing the psychological well-being of individuals. Social support consists of social relationships that provide (or can potentially provide) material and interpersonal resources which are valuable to the recipient, such as counseling, access to information and services, sharing of tasks and responsibilities, and skill acquisition (Burns, Anstey, & Windsor, 2011). Furthermore, the category of interpersonal resources often includes a range of emotional supports (e.g., empathy, caring, love, and trust), informational supports (e.g., advice, suggestions, access to information, etc.) and instrumental supports (e.g., aid in kind, sharing of tasks and responsibilities, skills acquisition, among others). These resources, either by themselves or in combination with more concrete material resources, help the recipient cope and adapt to negative life events and support their positive well-being as well (Lakey & Cohen, 2000). Therefore, social support becomes an important factor to consider when assessing adverse effects of youth unemployment. Aside social support, religiosity may also be considered as useful psychological resource that counters the negative consequences of youth unemployment. Shafranske and Maloney (1990) define religiosity as both beliefs and practices relating to an organized religious affiliation or a specified divine power. Psychological well-being have been linked with so many factors including religiosity, having or showing belief in and reverence for God or a deity, as well as participation in activities pertaining to that faith such as attending services/worship regularly and participating in other social activities with one’s religious community have relationship with one’s psychological well-being (Ramirez, Macedo, Sales, Figueiredo, Daher, Araujo, Pargament, Hyphantis, & Carvalho, 2012). Religiosity has been linked to a greater sense of psychological well-being in late adulthood as well as to the ability to better cope with stressful events in middle adulthood (Santrock, 2002). Previous researches pertaining to the subject area University of Ghana http://ugspace.ug.edu.gh 7 of religiosity and psychological well-being provide sound evidence to support the positive association between religiosity and psychological well-being. In Ghana and many other countries, religion remains an integral component of life for individuals confronted with negative life events such as unemployment (Ferraro & Kelley- Moore, 2000; Harrison, Koenig, Hays, Eme-Akwari, & Pargament, 2001; Laubmeier, Zakowski, & Bair, 2004; Stroebe, 2004). Many individuals rely on their religious beliefs and practices to provide meaning to unfortunate circumstances they encounter and to obtain comfort, hope, and social support (Pargament, 1990, 1997). Evidence also indicates that religious beliefs and practices can impact the coping process (Ramirez et al., 2012). Religion may permeate the stress process by influencing the cognitive and behavioral responses for interpreting and handling negative life events (Pargament, 1997). Religion appears to influence the interpretation, appraisal, and attribution of chronic illness (Gordon, Feldman, Crose, Schoen, Griffing, & Shankar, 2002; Siegel, Anderman, & Schrimshaw, 2001). Religion also may to contribute to the coping process by providing coping options through the social, interpersonal, cognitive, spiritual, and behavioral aspects of religious faith (Smith, McCullough, & Poll, 2003). Given these useful purposes of religion, it is necessary to assess the extent to which religion may positively influence individuals’ ability to cope with the negative consequences of unemployment. University of Ghana http://ugspace.ug.edu.gh 8 1.2. Statement of the Problem Being employed is one of the most desirable goals of every individual. Good employment is associated with several benefits. Aside its financial benefits, employment provides a sense of identity (Erikson, 1968), respect, social status, and good mental health. It is therefore not surprising that many empirical studies show that employment has a beneficial effect, specifically upon the mental health of the youth (Bjarnason, & Sigurdardottir, 2003; Schaufeli, 1992). Despite the numerous benefits associated with employment, a number of youths in Ghana remain unemployed. Ghana is currently facing economic crisis and unemployment is one major challenge associated with it. The youths are the most affected group (Amankrah, 2005). Youth unemployment has become a topical issue in public discussions. Evidence suggests that exposure to events such as joblessness are capable of impairing an individual’s psychological well-being (Erikson, 1959, 1982; Jahoda, 1979, 1981, 1982; & Seligman, 1975), and may lead to suicidal thoughts (Osafo, et al., 2011). Though the consequences of unemployment are multifaceted (social, economic, political and psychological), earlier researchers in Ghana (e.g. Dai et al., 2008; ISSER, 2007; Yarquah & Baafi-Frimpong, 2011) have ignored the psychological dimension which may be as devastating as the social, economic and political consequences. There is therefore the need to empirically examine and reveal the negative consequences of youth unemployment in Ghana and to examine the factors that help cushion the youths against the ills in unemployment. University of Ghana http://ugspace.ug.edu.gh 9 1.3. Aim and objectives of the study The aim of the study was to investigate the psychological effects of unemployment among the youths in Ghana and the factors that serve as buffers within the Ghanaian context. The specific objectives deduced from the aim are: i. To assess the impact of unemployment on psychological health among the youths ii. To ascertain the impact of duration of unemployment on psychological health among the youths iii. To examine the buffering effects of social support and religiosity on the relationship between unemployment and psychological health 1.4. Relevance of the Study The study provides knowledge on the negative psychological consequences of youth unemployment in Ghana. Through the literature and findings from the present study, readers will become aware of the adverse effects of youth unemployment in Ghana. Among the psychological effects of youth unemployment are depression, cognitive distortions, low self- esteem, and suicidality. Secondly, the study helps unemployed youths to identify the psychological resources that serve as buffer for the negative consequences of unemployment. It establishes the role of social support and religiosity in coping with the negative consequences of unemployment. Unemployed youths therefore have the opportunity to determine the most significant factor to cushion them against the negative consequences of unemployment. University of Ghana http://ugspace.ug.edu.gh 10 Moreover, the present study is informational for clinical and counseling practices. Findings from the study can assist clinicians and counsellors to correctly diagnose certain psychological problems associated with youth unemployment and to effectively plan interventions for their remedial. Additionally, the study challenges the government and policy makers to timely intervene in the unemployment situation in Ghana. The compelling evidence gathered on the psychological effects of youth unemployment will prompt the government of Ghana to come out with implantation strategies that will provide jobs for the youth in Ghana. Finally, the study adds to the existing literature. Although much has been documented on the psychological effects of youth unemployment, the new dimensions of the present study augment the existing literature. Particularly, the extension of psychological effects to suicidality expands the scope of the literature. In addition, the study also creates avenues for future research. The gaps in the present study provide useful opportunity for further investigations. University of Ghana http://ugspace.ug.edu.gh 11 CHAPTER TWO LITERATURE REVIEW 2.1. Theoretical Framework 2.1.1. Learned helplessness theory (Overmier & Seligman, 1967; Seligman, 1975, 1994) The learned helplessness model proposes that when an individual is repeatedly exposed to an uncontrollable aversive event, the person learns their responses are useless and that they have no control over the situation (Overmier & Seligman, 1967). The organism subsequently generalizes the helplessness to other situations. Overmier and Seligman (1967) and Seligman and Maier (1967) observed learned helplessness in animal laboratory experiments in which dogs were presented with a series of electrical shocks that could be neither escaped nor avoided. After repeatedly being shocked, the dogs eventually learned that despite their efforts, the negative stimulus (shock) could not be avoided. In response to the unavoidable and inescapable event, the dogs became immobile and passively received the shocks. This phenomenon of not responding to the event became known as “learned helplessness.” The dogs eventually generalized the uncontrollable events to those that were controllable and ultimately displayed behavioral activation deficits in response to novel stimuli (Peterson, Buchanan, & Seligman, 1995). 2.1.2. Reformulated Learned Helplessness Theory of Depression (Abramson et al., 1978; Abramson et al., 1989; Peterson & Barrett, 1987; Seligman et al., 1990). The reformulated learned helplessness model was originally formulated by Abramson, Seligman and Teasdale (1978) to explain affective, motivational and cognitive deficits observed in humans following exposure to uncontrollable negative events. According to the reformulated learned helplessness model, a pessimistic attributional style consists of attributing negative events to internal, stable, and global causes. This negative style is related to a helpless University of Ghana http://ugspace.ug.edu.gh 12 reaction. Reactions that accompany helplessness include passivity, sadness, anxiety, hostility, and low self-esteem (Peterson & Seligman, 1984). Any experience that involves learned helplessness might also be related to an individual’s attributional style including athletic performance, health, and career status (Bridges, 2001). Inasmuch, researchers have investigated the relationship between attributional style and psychological health. In general, an internal, stable and global attributional tendency, termed a depressogenic attributional style, is a risk factor for the development of a depressive reaction following a negative life event. However, a depressogenic attributional style should also manifest in behavioural and performance deficits as well as the well-documented affective disturbances. Although the learned helplessness reformulation presented an attributional account of human helplessness and depression, it failed to provide a clear, articulate theory of depression per se. Thus, the reformulated learned helplessness model was further revised and developed by Abramson, Metalsky and Alloy (1989) and termed the hopelessness model of depression. This revision is a hopelessness, rather than a pure attributional theory of depression and is, consequently, more similar to other cognitive theories of depression (e.g., Beck, 1967) than the 1978 reformulation. 2.1.3. Hopelessness theory of depression (Abramson, Metalsky, & Alloy, 1989) According to the Hopelessness Theory of Depression (Abramson, Metalsky, & Alloy, 1998), individuals with maladaptive, or negative, cognitive styles are vulnerable to depression when they encounter negative life events because they assign a negative meaning or consequences to the negative event. Specifically, those who make global and stable attributions, make negative self-inferences, and expect negative consequences following the occurrence of a negative life University of Ghana http://ugspace.ug.edu.gh 13 event are more likely to become depressed. More contemporary versions of this theory (Alloy, Abramson, Safford, & Gibb, 2006; Alloy et al., 2000) have focused primarily on the tendency to make stable and global attributions, called a negative cognitive style. Individuals with the negative cognitive style (NCS) have been found to be at greater risk for depression and suicide across a variety of samples (Abramson et al., 1998). The hopelessness theory of depression posits the existence of a subtype of depression— hopelessness depression— which is hypothesized to have a characteristic cause, symptom profile, course, treatment, and prevention. The symptoms of hopelessness depression are posited to include retarded initiation of voluntary responses (motivational symptom), sad affect, suicidality, lack of energy, apathy, psychomotor retardation, sleep disturbance, difficulty in concentration, and mood-exacerbated negative cognitions. Although low self- esteem and interpersonal dependency also are posited to be part of the clinical picture under specified conditions (Abramson et al., 1989), research work suggests that low or labile self- esteem may best be viewed as part of the causal sequence (Metalsky, Joiner, Hardin, & Abramson, 1993; Roberts & Monroe, 1992). In addition, interpersonal dependency may be a symptomatic feature of hopelessness depression (Metalsky & Joiner, 1997). Based on the above theoretical assumptions, the present study assesses how various psychological variables of mental health are affected by stressors associated with unemployment. Among these variables are depression, dysfunctional thoughts, low self- esteem, and suicidal ideation. The relationships among these variables and their link with stressful life events suggest that they can together act as likely consequences of unemployment. Unemployment, in this context, is conceptualized as a stressful life event for those who are affected. University of Ghana http://ugspace.ug.edu.gh 14 2.1.4. Cognitive behavioral theories (Beck, 1979, 1991, 1995; Beck & Rector, 1998; Ellis, 1987, 1996, 1997) The cognitive-behaviour theories offer a theoretical explanation to how patients’ cognition affects their psychological health (Ellis & Harper 1975; Ellis, 1997; Beck, 1991; Beck & Rector, 1998). The basic assumption underlying these theories is that faulty thought processes and beliefs create problem behaviours and emotional disturbances. The theorists assume that depression is extremely common and serious mental disorder which has an important cognitive component. They stem, at least in part, from distorted and often self-defeating modes of thought. Based on their cognitive component, a number of cognitive therapists focus their attention on the cognitive processes that underlie depression. They therefore propose the use of cognitive-behaviour therapy as an effective means of treating psychological problems such as depression. Beck and Rector (1998) conjectured that underlying dysfunctional beliefs and thoughts do serve as a diathesis for the occurrence of depressions. Beck postulated a negative cognitive triad which was made up of thoughts about the self, one’s surroundings, and about one’s future. This cognitive triad has a role in a person’s experience of depression. Moreover, Beck (1979, 1991) proposed a cognitive behaviour therapy as a curative measure to depression. He argued that depressed individuals engage in illogical thinking and that this underlies their difficulties. Such individuals hold unrealistically negative beliefs and assumptions about themselves, the future, and the world. Beck contended that people cling to these illogical ideas and assumptions no matter what happens. Such distorted thinking leads individuals to experience negative moods – which in turn, increase the probability of more negative thinking. Like Ellis, Beck has been criticized for his exclusive emphasis on the role of cognition on depression at the expense of other potential causes like the role of biology and one’s genetic make-up as well as environmental factors. University of Ghana http://ugspace.ug.edu.gh 15 2.1.5. The Buffer Theory of Social Support (Alloway & Bebbington, 1987; Lin, Woefel & Light 2005) The buffer theory of social support proposed by Alloway and Bebbington (1987) and espoused by Lin, Woefel and Light (2005) suggests that the presence of social relations or support networks moderates effect of adverse environmental stressors that precipitate illness and disease. The theory further proposes that social support may not contribute directly to health outcomes but acts as a ‘‘buffer’’ to protect the individual from harmful effects of one’s environment in times of stress. This suggests that social support may intervene between the stressful event/s and the stress response by attenuating or preventing a stress appraisal response of the individual. Rather than protecting against the effects of a stressful event when it occurs, the buffer theory suggests that a protective effect is achieved by preventing or reducing the amount of psychological risk factors experienced. In the context of this study, unemployment may be perceived as a stressful event whose effects can be minimized through social support. It is expected that unemployed youths with social support will experience less of the negative consequences associated with unemployment. 2.1.6. Afrocentric Framework of Stress and Coping (Utsey, Adams, & Bolden, 2000; Utsey, Brown, & Bolden, 2004). The Afrocentric framework of stress and coping emphasizes collective and communal orientation in coping among Africans and African Americans. Utsey et al. (2000, 2004) identified communally and spiritually based coping to be particularly prevalent among individuals of African descent, reflecting an Afrocentric worldview. These observations find support in a recent coping study of Black Canadians. Joseph and Kuo (2009) reported that spiritual- and ritual-centered coping constituted the most crucial coping strategies adopted by Black Canadians in dealing with interpersonal discrimination (e.g., being looked down on as unintelligent by others). Additionally, in a study of coping with the September 11 attacks, University of Ghana http://ugspace.ug.edu.gh 16 Constantine Alleyne, Caldwell, McRae, and Suzuki (2005) found that both acquiring from and giving support to in-group members and religious coping were an integral part of coping among African Americans. These coping behaviors further underscored the centrality of collectivism and communalism in African Americans. Ghanaian youths are therefore likely to use religion and spirituality to effectively cope with the stress associated with unemployment. Thus, religiosity has been identified as a moderator for the psychological effects of unemployment among the youths in Ghana. 2.2. Review of Related Studies 2.2.1 Unemployment and Social Vices The lack of employment opportunities in poor economies is known to have a lot of social implications. Unemployment among the educated youth has been found to promote rural-urban migration and its attendant vices. It has been observed that lack of job prospects and the likelihood of a desolate future for unemployed young people may contribute to socially deviant behaviour (Sommers 2003). The rise in criminal activities, drug addiction and prostitution among young unemployed is due partly to the combined effects of lack of social networks and insufficient job opportunities (WHO, 2004). Yarquah and Baafi-Frimpong (2012) conducted a study that looked at the social cost of educated youth unemployment in Ghana and its implications for education. The research design used was the descriptive survey type. Out of an estimated population of 16,000 in the Central and Eastern regions of Ghana, which constituted the research areas, a total sample size of 446 was obtained for the study. The snowballing sampling technique was used to select the educated unemployed youth from some popular spots where the youth usually converged. The three study settings of Elmina, Koforidua and Cape Coast were also obtained using the purposive sampling technique. The main factors identified as accounting for the high educated University of Ghana http://ugspace.ug.edu.gh 17 youth unemployment included: lack of adequate learning facilities for quality education; inadequate teachers to teach various subjects in the schools and dropouts from school. In the study of Yarquah and Baafi-Frimpong (2012), it was found that educated youth unemployment led to streetism and its attended social vices such as stealing, drug abuse, and prostitution. The respondents intimated that more emphasis should be placed on technical and vocational education that would equip people with employable skills and that the curricula of schools would have to be tailored to suit the industrial, technological and development needs of the society. Government creating an enabling environment for businesses to thrive; and intensification of the provision of counselling services in schools among others were recommended for dealing with the unemployment menace. A study conducted by Carmichael and Ward (2001) in England and Wales on the link between unemployment and crime suggest that youth unemployment and the different types of crime such as burglary, theft, fraud and forgery and total crime are significantly and positively correlated. According to the International Labour Organization (ILO, 2005), in Africa, unemployment has driven many young women and girls to become social sex workers. Struggling to support families and provide care to sick members of the household, their ability to pursue education is often curtailed. Lack of job opportunities and their disadvantageous social role, both in terms of assets (education and health) and cultural norms, make them more likely to end up as sex workers (ILO, 2005). University of Ghana http://ugspace.ug.edu.gh 18 2.2.2. Unemployment and mental health Schaufeli (1997) investigated two hypotheses: (1) the causation hypothesis that assumes that unemployment leads to poor mental health and (2) the selection hypothesis that assumes that poor mental health reduces the likelihood of finding a job. A prospective longitudinal design was used in order to study two Dutch samples: 635 college graduates and 767 school-leavers. The causation hypothesis was confirmed for school-leavers but not for college graduates. In addition, as expected, employment and further education increased levels of mental health among school-leavers. The selection hypothesis that could only be studied in the graduate sample was not confirmed as far as mental health was concerned. However, it appeared that future employment among graduates was predicted by a positive attitude and an active way of dealing with unemployment. Schaufeli (1997) interpreted the results with reference to the favourable Dutch structural and cultural context that existed at the time the research was conducted. The present study however tests the causation hypothesis but not the selection hypothesis due to limitation in the use of cross-sectional survey. Pharr, Moonie and Bungum (2011) conducted a study to examine the impact of employment status and unemployment duration on perceived health, access to health care, and health risk behaviors. Data from Nevada's 2009 Behavioral Risk Factor Surveillance System (BRFSS) were analyzed. Pharr et al. (2011) compared participants who were unemployed (greater than and less than one year) to those who were employed and those who were voluntarily out of the labor force (OLF). Unemployed participants had significantly worse perceived mental health profiles, were more likely to delay health care services due to cost, and were less likely to have access to health care than employed participants and OLF participants. OLF participants were not significantly different from employed participants. Findings from the study suggested that the impetus for unemployment, be it voluntary or involuntary, may significantly impact a person's mental health. University of Ghana http://ugspace.ug.edu.gh 19 2.2.3. Unemployment and depression Studies have found that many young unemployed people feel marginalized, pessimistic and lacking in control over their lives leading to poor psychological adjustment in life. A study was conducted by UCU lecturers' union (2013) to assess the impact of unemployment on level of depression. The poll examined views of some 1,000 youngsters aged 16-24 across the UK. The review indicated that many of the unemployed participants felt isolated and were lacking in confidence. About half had distorted thinking with 40% feeling that they are not part of society, 36% believing that they will never have a chance of getting a job. One third of the participants were found to experience depression and 39% were found to suffer from stress. Dooley et al. (2000) examined the link between change in employment status and depression. They drew upon data from the National Longitudinal Survey of Youth in the U.S. Their study was based on about 5,000 respondents who were “adequately” employed in 1992, and who were interviewed in 1992 and 1994. Dooley et al. (2000) found that becoming unemployed, inadequately employed or inactive were all significantly associated with increased depression compared with those remaining adequately employed. Being married reduced this impact, while being highly educated increased it. They also looked at the selection effect: being depressed at the first interview was significantly associated with the likelihood of becoming unemployed at the second interview. However, it was not significantly associated with the likelihood of becoming inadequately employed or inactive. In a similar study, Malik (2013) investigated the relationship between depression and unemployed adults from both rural and urban areas. The study was done on the sample of 60 subjects both male and female. In the study a self-report scale by Shamin Karim and Rama Tiwari was used to collect data from the subjects. The results indicated that unemployed participants had higher level of depression than employed participants. Unemployed males had University of Ghana http://ugspace.ug.edu.gh 20 higher level of depression in comparison to unemployed females. The result further showed that depression is higher in urban males and rural females and less in rural males and urban females. Breslin and Mustard (2003) looked at whether the impact of unemployment on mental health varied between young adults and those over age 30. Breslin and Mustard (2003) analyzed longitudinal data from the National Population Health Survey in Canada, from over 6,000 respondents aged 18-55. They were first interviewed in 1994-1995 and followed up two years later. Breslin and Mustard separately examined the effects of causation (from unemployment to mental health problems) and of selection (from mental health problems to unemployment). Mental health was measured in two different ways: with a six-item scale to assess mental distress or with another instrument to assess major depression. When testing the effects of causation, the researchers controlled for change in employment status, gender, age and initial mental health. Breslin and Mustard found that becoming unemployed was associated with an increased likelihood of mental distress for the 31-55 age group, but not for the 18-30 age group. This is consistent with an earlier study by Clark and Oswald (1994), who found that the impact of unemployment on psychological distress was greatest for those aged 30-49. Breslin and Mustard initially obtained similar results when measuring major depression. However, for both groups, becoming unemployed did not have a significant impact on the incidence of major depression when socio-demographic factors and initial depression levels were controlled. As well, people who were unemployed at both time periods did not show increasing distress over time. There was evidence of a selection effect in both age groups: those with high mental distress initially were much more likely than others to become unemployed. Results were not significant with depression as the measure of mental health. University of Ghana http://ugspace.ug.edu.gh 21 In its Employment Outlook, the Organization for Economic Co-operation and Development (OECD, 2008) examined many of these issues through an analysis of longitudinal data in five countries: Australia, Canada, Korea, Switzerland and the United Kingdom (U.K.). The OECD’s analysis used statistical methods to sort out the possibility of a bi-directional cause- and-effect relationship between mental health and change in employment status. The results showed that moving from employment to unemployment or inactive status (out of the labour force) had a large, negative impact on mental health, with a larger impact on men than women. In Canada, Australia, the U.K. and Switzerland, the increase in mental distress was greatest when there was a change from employment to inactivity due to illness. A movement from employment to unemployment also had a significantly negative impact on mental health. The OECD’s analysis also found that when people’s status changed from non-employment to employment, their mental health improved (with exceptions for men in Australia and women in Korea and Switzerland). In Canada, there were significant gains for both men and women, with somewhat greater gains for men. Duration of unemployment mattered, but the impact varied across countries. For example, in the U.K., there was evidence of a “habituation” effect. Psychological distress was greater for those just unemployed or inactive than for those who had been unemployed or inactive for over two years. However, in Australia, long-term unemployment worsened mental health for males. (Duration analysis was not possible for Canada.) Winkelman and Winkelman (1998) looked at the effects of unemployment on self-reported life satisfaction of men aged 20-64. They found that the size of the effect was unrelated to the duration of unemployment. However, their study is quite old and there is the need for replication to assess the psychological effect of duration of unemployment. University of Ghana http://ugspace.ug.edu.gh 22 Paul and Moser (2009) examined the effect of unemployment on mental health with meta- analytic methods across 237 cross-sectional and 87 longitudinal studies. The average overall effect size was d = 0.51 with unemployed persons showing more distress than employed persons. A significant difference was found for several indicator variables of mental health (mixed symptoms of distress, depression, anxiety, psychosomatic symptoms, subjective well- being, and self-esteem). The average number of persons with psychological problems among the unemployed was 34%, compared to 16% among employed individuals. Their moderator analyses demonstrated that men and people with blue-collar-jobs were more distressed by unemployment than women and people with white-collar jobs. Linear and curvilinear moderating effects of the duration of unemployment were also identified. Paul and Moser (2009) further observed that the negative effect of unemployment on mental health was stronger in countries with a weak level of economic development, unequal income distributions, or weak unemployment protection systems compared to other countries. Meta- analyses of longitudinal studies and natural experiments endorsed the assumption that unemployment is not only correlated to distress but also causes it. Seemingly inconsistent longitudinal results of older meta-analyses can be explained by retest artifacts. Paul and Moser (2009) also identified mental-health related selection effects during job loss and job search, but they are weak. With an effect size of d = -0.35 intervention programs for unemployed people were found to be moderately effective in ameliorating unemployment-related distress among continuously unemployed persons. Reine, Novo, and Hammarstrom (2013) analyzed the associations between unemployment and suboptimal self-rated health as well as high alcohol consumption. The also examined the role of possible mediating factors explaining the associations from a gender perspective. Reine, Novo, and Hammarstrom (2013) designed a 14-year longitudinal study with a 96.4% response University of Ghana http://ugspace.ug.edu.gh 23 rate. Their sample consisted of 386 women and 478 men who were either employed or unemployed at 30 years of age. The health outcomes studied were suboptimal self-rated health and high alcohol consumption at 30 years of age. Logistic regression was used for analysis, and the relational theory of gender was used to discuss the findings. Reine, Novo, and Hammarstrom (2013) found a strong relationship between unemployment and suboptimal self-rated health among women, and unemployment and high alcohol consumption among men, even after controlling for health-related selection, potential mediators and background factors. All mediating factors in the model were attributable to suboptimal self-rated health among unemployed women. Two mediating factors were also substantially related to high alcohol consumption among unemployed men. This led them to conclude that long-term unemployment at a young age could have various health effects in men and women. However, their study only showed that the mechanisms behind the health consequences were better understood among women. They therefore recommended that future researchers should attempt developing theories in order to explain how youth unemployment leads to gendered health consequences. 2.2.4. Unemployment and cognitive distortions Leposavić and Leposavić (2009) claimed that the presumption which makes people depressive consists, to an excessive degree, of internal, stable and global attributions to negative occurrences. They believed that negative attributions for unpleasant events are associated with the loss of self-respect which follows. In line with their claim, Leposavić1 and Leposavić2 (2009) conducted study to establish the characteristics of attribution style of depressive patients. Their investigation included 62 subjects. The first group consisted of 32 patients with endogenous depression in remission. The second group included 30 healthy subjects. They University of Ghana http://ugspace.ug.edu.gh 24 tested the characteristics of attribution style, in both groups, using the Attribution Style Questionnaire (ASQ). The results of Leposavić and Leposavić (2009) showed that the group of depressive patients, in comparison with healthy subjects, exhibited a significantly more marked internal attribution for negative events and global internal negative attributions. There was no significant difference between the groups in the stability of these negative attributions, and also the composite score which represents the measure of hopelessness did not make a significant difference between depressive and healthy subjects. They therefore concluded that depressive patients exhibit an inclination towards internal and global attribution for negative events. These negative attributions do not have stable character (i.e. these attributions vary in time). The findings thus suggest that depressive patients are more pessimistic in attribution. Athough, Leposavić and Leposavić (2009) investigated depressive patients, their observation of negative attribution of the patients confirms that negative life events such as illness and even unemployment have the greater tendency to arouse negative thoughts and attributions among the affected. Such negative thoughts may even be worse in hopeless situations such as persistent unemployment. Tiggemann, Winefield, and Winefield (2009) studied the relationship between attributional style and subsequent psychological distress in a prospective design. A group of young adults completed the Attributional Style Questionnaire. Then 1 year later completed a number of psychological measures; negative mood, self-esteem, locus of control, depressive affect, hopelessness, and the General Health Questionnaire-28. Attributional style was found to predict subsequent well-being, but not when the effects of initial well-being were controlled. Nor did the inclusion of negative life events improve the prediction. The results of Tiggemann et al. (2009) were inconsistent with the hopelessness model of depression. University of Ghana http://ugspace.ug.edu.gh 25 Fresco and Alloy (2006) examined the relationship of attributional styles for negative and positive events with depression and anxiety. A sample of 239 college students underwent structured diagnostic interviews and completed self–report measures of attributional style and major life events at two time points separated by approximately four weeks. Using cross– sectional methodology, attributional styles for negative and positive events were compared across current diagnoses of unipolar depression and/or anxiety. Using a prospective design, attributional styles for negative and positive events were assessed as moderators of the relationships between negative and positive life events and levels of subsequent depression symptoms. The tendency to see negative events arising from internal, stable, and global causes and positive events arising from external, unstable, and specific causes, was associated with higher levels of clinician-assessed depression symptoms, particularly when confronted with negative life events or the absence of positive events. Findings indicate that attributional style for positive events contributes to our understanding of cognitive vulnerability to emotional disorders. Shaheen and Alam (2010) also studied psychological distress and its relation to attributional styles and coping strategies among a sample of 300 (150 male and 150 female) eleventh grade students. Shaheen and Alam (2010) found composite attribution for positive events and its three dimensions (i.e. internal-external, stable-unstable and global- specific) to be negatively correlated with psychological distress and composite attribution for negative events and its three dimensions (i.e. internal-external, stable-unstable and global-specific) were positively correlated with psychological distress. They further observed that problem focused coping strategies negatively related to psychological distress and avoidance coping positively related to psychological distress. The results of Shaheen and Alam (2010) provide supportive evidence for the hopelessness model of depression that suggest that negative events often lead individuals to make pessimistic attributions that results in psychological distress. University of Ghana http://ugspace.ug.edu.gh 26 2.2.5. Unemployment and suicide behavior According to WHO (2000), there are fairly strong associations between unemployment rates and suicide rates, but the nature of these associations is complex. The effects of unemployment are probably mediated by factors such as poverty, social deprivation, domestic difficulties and hopelessness. On the other hand, people with mental disorders are more likely to be unemployed than people in good mental health. At any rate, due consideration should be given to the difference in the significance of recent loss of employment and long-term unemployment: greater risk is associated with the former. Fergusson, Boden, and Horwood (2007) examined the association between exposure to unemployment and suicidal behaviours (suicidal ideation and attempted suicide) in a birth cohort of New Zealand young adults using fixed effects logistic and Poisson regression models. Data were gathered on unemployment and suicidal behaviours at annual periods from ages 16- 25 years. At all ages increasing exposure to unemployment was associated with increased risks of suicidal ideation and number of suicide attempts. Following adjustment for fixed effects and time-dynamic covariates, associations between unemployment and suicidal ideation reduced to marginal significance, while the association between unemployment and suicide attempt was not statistically significant. After adjustment, those experiencing six or more months of unemployment in a given year had odds of suicidal ideation that were 1.43 times higher, and rates of suicide attempt that were 1.72 times higher, than those who were not exposed to unemployment. Although unemployment was associated with moderate increases in risks of suicidal behaviours, much of this association was explained by confounding factors. Córdoba-Doña, Sebastián, Escolar-Pujolar, Martínez-Faure, and Gustafsson (2014) observed that although suicide rates increased in some European countries in relation to their current economic crisis and austerity policies, that trend was not observed in Spain. They therefore University of Ghana http://ugspace.ug.edu.gh 27 examined the impact of the economic crisis on suicide attempts, the previously neglected endpoint of the suicidal process, and its relation to unemployment, age and sex. Their study was carried out in Andalusia, the most populated region of Spain, and which has a high level of unemployment. Information on suicide attempts attended by emergency services was extracted from the Health Emergencies Public Enterprise Information System (SIEPES). Suicide attempts occurring between 2003 and 2012 were included, in order to cover five years prior to the crisis (2003–2007) and five years after its onset (2008–2012). Information was retrieved from 24,380 cases (11,494 men and 12,886 women) on sex, age, address, and type of attention provided. Age-adjusted suicide attempt rates were calculated. Excess numbers of attempts from 2008 to 2012 were estimated for each sex using historical trends of the five previous years, through time regression models using negative binomial regression analysis. To assess the association between unemployment and suicide attempts rates, linear regression models with fixed effects were performed. The results of Córdoba-Doña et al.’s study showed a sharp increase in suicide attempt rates in Andalusia after the onset of the crisis, both in men and in women. Adults aged 35 to 54 years were the most affected in both sexes. Suicide attempt rates were associated with unemployment rates in men, accounting for almost half of the cases during the five initial years of the crisis. Women were also affected during the recession period but this association could not be specifically attributed to unemployment. Their study is significant in enhancing our understanding of the potential effects of the economic crisis on the rapidly increasing suicide attempt rates in women and men, and the association of unemployment with growing suicidal behaviour in men. It also provides impetus for research on the suicide effects of the current economic crisis in Ghana since such effects may differ cross-culturally in terms of nature and process. University of Ghana http://ugspace.ug.edu.gh 28 2.2.6. Religiosity as a buffer for the psychological effects of unemployment An increased interest in the effects of religion on mental health and psychological wellbeing is apparent in psychological literature. A number of well conducted clinical and epidemiological studies have shown that the religiosly committed had much less psychological distress than the uncommitted (William, Larson, Buckler, Heckman & Pyle, 1991). Ismail and Desmukh (2012) explored the link between religiosity and psychological well-being in a model of Pakistani Muslims. They used a sample 65 men and 85 women with an age range from 18 to 60 years. Reliable with previous research, their study suggested that a strong, negative relationship indeed existed between religiosity and loneliness that individual with high religiosity are less likely to be lonely and will not be expressing anxiety in all situations. A strong positive relationship was also found between religiosity and life satisfaction. Thus, the results of their study supported the existing claims on the relationship between religiosity and different factors of psychological well-being that individuals who highly participate in religious activities and engagement with institutions like religion have positive life satisfaction because they are involved in activities that satisfy their needs and serve to temper individuals’ desire and thereby help them to achieve better psychological well-being. Bergin’s (2001) review of empirical literature on the relationship between religiosity and mental health provides evidence that average effects are generally positive, although not dramatic. His review indicates a number of correlations between religious affiliation and positive psychological functioning. He found out from the review that religious beliefs and practices contribute substantially to the formation of personal moral criteria and sound moral judgment. The regular practice of religion encouraged such beneficial effects on mental health, such as less depression, higher self-esteem and greater family and marital happiness. University of Ghana http://ugspace.ug.edu.gh 29 2.2.7 Social support as a buffer for the psychological effects of unemployment Social support is consistently identified as buffering the effects of life events on wellbeing outcomes in clinical samples (Ames & Roitzsch, 2000) and the general population (Falcon, Todorova, & Tucker, 2009). Haden, Scarpa, Jones and Ollendick (2007) maintained that quality of interpersonal relationships was a significant moderator of negative stressful events and individual wellbeing. Their findings supported the postulations of the buffer theory of social support. Moreover, a significant literature highlights the role of positive psychological function on wellbeing in organizational, epidemiological and clinical contexts across the lifespan (Burns & Machin, 2012; Huppert & Whittington, 2003; Ruini, Belaise, Brombin, Caffo, & Fava, 2006). Lorenzini and Giugni (2010) investigated the role of social support for diminishing the negative consequences of long-term unemployment. They believe that young unemployed who benefit from different kinds of social support will be less affected by financial distress, anxiety, and unhappiness. In addition, they expected to find a greater effect of the closest circles of support: first, partner support, then family support, and finally friend support. They used quantitative data derived from a telephone survey carried between February and August 2010 on a representative sample of young long-term unemployed and precarious youth, plus a control group of regularly employed youth. All three groups included people aged between 18 and 34 residing in the canton of Geneva, which displayed consistently the highest unemployment rates in Switzerland. Long-term unemployment was defined as having been without a job for at least one year. Here they focused on unemployed and regularly employed youth. The sample size for these two groups was, respectively 124 and 320. The analysis of Lorenzini and Giugni (2010) provided evidence of an impact of the social support on the personal welfare of young long-term unemployed. Such an impact, however, University of Ghana http://ugspace.ug.edu.gh 30 varied according to the type of social support and the specific aspect of well-being under study. While no effect was observed on financial distress, partner support seems to reduce anxiety- related problems, and finally all three types of social support contributed in some way to increase the level of happiness of young unemployed. According to Burns and Machin (2013), negative life events are associated with poor wellbeing and mental health outcomes. Following a diathesis-stress model, Burns and Machin (2013) tested whether psychological functioning and quality of interpersonal relationships moderated the effect of life events on subjective wellbeing. Using data from a young and middle-aged adult sample (n = 364) drawn from an Australian university-student population, the researchers discovered that that life events were associated with negative but not positive wellbeing outcomes. Perceived impact of life events was a stronger predictor of wellbeing than was the number of life events. Psychological functioning and quality of interpersonal relationships were associated with both wellbeing dimensions but only quality of interpersonal relationships moderated the effect of life events on wellbeing. In conclusion, perceived impact of life events was more strongly related to wellbeing than number of life events. Interpersonal relationships moderate the effect of life events with those reporting higher levels of quality of interpersonal relationships reporting less decrement in negative affect following stressful life events. Bjarnason and Sigurdardottir (2003) claimed that psychological distress is a serious problem among unemployed youth, and may lead to various social and psychological problems. They therefore examined patterns of distress among previously unemployed youth that have experienced five different labor market outcomes over a period of 6 months in Denmark, Finland, Iceland, Norway, Scotland and Sweden. They found that moving beyond unemployment is associated with less distress, in particular among those who have found permanent employment, but also among those who have found temporary employment, have returned to school, or are staying at home. Perceptions of material deprivation and parental University of Ghana http://ugspace.ug.edu.gh 31 emotional support directly affected distress in all labor market outcomes, and mediated the effects of various other factors on such distress. The effects of socio-demographic characteristics, living arrangements, unemployment history and attitudes, and parental support were found to be specific to gender and labor market outcomes, while the effects of material deprivation were uniform across all such categories. The direct effect of parental support in the study of Bjarnason and Sigurdardottir (2003) suggests that social support reduces the psychological effects of unemployment. Consequently, Bjarnason and Sigurdardottir (2003) recommended further studies to disentangle structural and individual effects, the causal complexities involved in processes of social support, and to determine the extent to which such models equally predict psychological distress among the unemployed and other groups of youth. Using both quantitative and qualitative data drawn from an ongoing EU-funded research project, Lorenzini and Giugni (2010) conducted a study with two main goals. Firstly, their study ascertained if and to what extent long-term unemployment disrupts the personal life of young unemployed. They focused more specifically on three aspects of their personal welfare: financial distress, anxiety-related health problems, and the overall level of happiness. Secondly, they aimed to assess the impact of social support on these three aspects of the personal life of young unemployed. Here, the authors focused on three kinds of social support: by the partner, by the family, and by close friends. Their analysis showed that long-term unemployment has a number of important consequences on the personal life of young people on all three counts, producing financial distress, creating anxiety-related health problems, and diminishing the overall level of happiness. Young long- term unemployed were not very well on all three aspects when compared to youth who have a regular job. In addition, they found evidence of an impact of the social support on the personal University of Ghana http://ugspace.ug.edu.gh 32 welfare of young long-term unemployed. However, not all three sources of social support played a role and not for all three aspects of well-being. Specifically, no effect was observed on financial distress, partner support seems to reduce anxiety-related problems, and all three types of social support contributed in some way to increase the level of happiness of young unemployed. This finding also illustrates the crucial role that social support plays in the psychological health of individuals who experience negative life events such as unemployment. Åslund, Larm, Starrin and Nilsson (2014) argued that financial stress is an important source of distress and is related to poor mental and physical health outcomes. They investigated whether tangible social support could buffer the effect of financial stress on psychological and psychosomatic health. Individuals with high financial stress and low tangible social support had six to seven times increased odds ratios for low psychological well-being and many psychosomatic symptoms. By contrast, individuals with high financial stress and high tangible social support had only two to three times increased odds ratios for low psychological well- being and three to four times increased odds ratios for many psychosomatic symptoms, suggesting a buffering effect of tangible social support. Consistent with the buffering hypothesis, there were significant interactions between financial stress and social support, particularly in relation to low psychological well-being. They concluded that social support had its strongest effect at high levels of financial stress. The question whether the altering of our social networks may improve physical health is important for the prevention of ill health in people experiencing financial stress. Strengthening social networks may have the potential to influence health-care costs and improve quality of life. According to Burns and Machin (2013), negative life events are associated with poor wellbeing and mental health outcomes. Following a diathesis-stress model, Burns and Machin (2013) tested whether psychological functioning and quality of interpersonal relationships moderated the effect of life events on subjective wellbeing. Using data from a young and middle-aged University of Ghana http://ugspace.ug.edu.gh 33 adult sample (n = 364) drawn from an Australian university-student population, the researchers discovered that that life events were associated with negative but not positive wellbeing outcomes. Perceived impact of life events was a stronger predictor of wellbeing than was the number of life events. Psychological functioning and quality of interpersonal relationships were associated with both wellbeing dimensions but only quality of interpersonal relationships moderated the effect of life events on wellbeing. In conclusion, perceived impact of life events was more strongly related to wellbeing than number of life events. Interpersonal relationships moderate the effect of life events with those reporting higher levels of quality of interpersonal relationships reporting less decrement in negative affect following stressful life events. Some researchers and practitioners have promoted support groups as a way to improve psychological well-being. Contrary to expectations, however, support groups do not always seem to improve patients' mental health. No consistent improvements in quality of life or depression scores were found in an eight-week mental health patients’ peer-support programme (Uccelli, Mohr, Battaglia, Zagami & Mohr, 2004). Additionally, individual cognitive behavioural therapy (CBT) and antidepressants were both found to be significantly more effective than group therapy at reducing depression among 63 mental health patients (Mohr, Boudewyn, Goodkin, Bostrom & Epstein, 2001). Importantly, however, these studies fail to assess the extent to which patients subjectively identify with the support group itself. Wakefield, Bickley and Sani (2013) believed that mental illness is associated with various psychological problems, including depression and anxiety. Whilst patients’ support groups are intended to improve mental health, this goal is not always achieved. Wakefield, Bickley and Sani, (2013) investigated 152 patients with multiple sclerosis who were recruited via UK multiple sclerosis support groups. Their study included measures of support group identification, depression, anxiety and satisfaction with life, as well as control variables (education level and age). The results of their study revealed that support group identification University of Ghana http://ugspace.ug.edu.gh 34 was significantly linked to depression, anxiety and satisfaction with life. Moreover, group identification explained a significant amount of variance in addition to that explained by education and age on each health outcome. Repeating the analysis to compare each of the three main sub-types of multiple sclerosis revealed these effects to be present for individuals with Relapsing–Remitting (RR) and Primary Progressive (PP) multiple sclerosis, but not for those with Secondary Progressive (SP) multiple sclerosis. Based on the above findings, Wakefield, Bickley and Sani, (2013) suggested that highly identifying with a support group has important positive outcomes for the psychological wellbeing of multiple sclerosis patients. This has implications for practicing clinicians. That is, people with mental illness should be encouraged to engage with support groups, but more must be done to ensure they subjectively identify with these groups, rather than merely attending them. 2.3. Rationale for the Present Study The definition of unemployment in the literature does not reflect the present unemployment situation of Ghanaian youths. The literature focusses on individuals with job loss as participants. Whilst this category of individuals falls within the scope of unemployment, it must be noted that many unemployed youths in Ghana may not have been previously employed and might be experiencing long term effects of unemployment, and among the employed youths many are earning below their expectations and are grossly dissatisfied with their present jobs. The present study therefore examines the extent to which the length of unemployment affects the psychological health of the youths in Ghana. It also seeks to examine quality of employment in terms of the adequately employed and the inadequately employed. The inadequately employed refers to youths who have temporally settled on menial jobs and are seeking better University of Ghana http://ugspace.ug.edu.gh 35 alternative jobs. The adequately employed are youths who are employed in jobs that provide them with expected satisfaction and income. Findings on the psychological effects of unemployment have not been consistent across countries. For instance, it is known in the literature that the duration of unemployment matters, but the impact varies across countries (Organization for Economic Cooperation and Development (OECD), 2008). Therefore, we cannot simply assume that the psychological effects of unemployment, as documented in other countries, can be generalized in whole to unemployed youths in Ghana. In this regard, Winefield and Fryer (1996) have stressed the importance of taking into consideration the historical and societal setting when research findings on unemployment and mental health are interpreted. For instance, Paul and Moser (2009) observed that the negative effect of unemployment on mental health was stronger in countries with a weak level of economic development, unequal income distributions, or weak unemployment protection systems compared to other countries. By all standards, Ghana can be judged to fall in the former category. Methodologically, most of the studies on suicide and suicidal ideation in Ghana have exclusively relied on qualitative design to identify unemployment as one of the risk factors for suicide (Adinkrah, 2014; Osafo et al., 2011; Knizek, Akotia, & Hjelmeland, 2010). Given the inherent limitations of qualitative studies, the present study is being conducted as a sequel to the earlier qualitative studies with the objective of testing earlier findings with quantitative data from a large sample. This will help to establish the validity of qualitative research findings and increase their utility/ generalizability Finally, suicidal ideation has not prominently featured in the literature on the psychological effects of unemployment. Earlier studies have cited depression, low self-esteem, and anxiety as the major psychological consequences of unemployment and social support as a buffer. This University of Ghana http://ugspace.ug.edu.gh 36 present study however seeks to expand the scope of research by investigating suicidal ideation as a negative consequence of unemployment and religiosity and social support as buffers in the Ghanaian society. 2.4. Key variables 2.4.1. Independent variables: Employment status and duration of unemployment 2.4.2. Moderators: Religiosity and social support 2.4.3. Dependent variables: Depression, cognitive distortions, self-esteem, and suicidality 2.5. Statement of Hypotheses 1. There will be poorer psychological health among unemployed youths than employed youths. 2. Duration of unemployment will significantly predict poorer psychological health among the youths. 3. Religiosity will moderate the relationship between youth unemployment and psychological health. 4. Social support will moderate the relationship between youth unemployment and psychological health. 2.6. Operational Definitions  Psychological health: Depression, cognitive distortions, self-esteem, and suicidality  Employment status: The state of being either employed or unemployed.  Unemployment: The state of being without a paid job for a period.  Employment: The state of being actively engaged in a paid job. University of Ghana http://ugspace.ug.edu.gh 37  Duration of unemployment: The number of years a person has stayed unemployed since age 18 and/or the year he or she completed full-time studies.  Social support: Support provided to a person by family, friends and significant others  Religiosity: Involvement in and active practice of religious faith and belief in and bond with a supernatural being or force 2.7. Hypothesized Conceptual Model Psychological Health Figure 1. Hypothesized model for the relationship between unemployment, religiosity, social support, and psychological health The model above predicts that unemployment affects the psychological health of the youths. In addition, the model depicts social support and religiosity as moderating variables for the relationship between unemployment and psychological health. Religiosity  Depression  Cognitive distortions  Self-esteem  Suicidality Unemployment Social Support University of Ghana http://ugspace.ug.edu.gh 38 CHAPTER THREE METHODOLOGY 3.1. Population Youths in the Greater Accra region of Ghana between the ages of 18 and 35 constituted the population for the research. The Greater Accra region is the most populous region for unemployed youths. It is the capital region of the country. It attracts most young people who are seeking for greener pastures. The population is sub-categorized into the employed and unemployed youths. Employed youths are youths who are actively engaged in paid jobs. Unemployed youths are youths who have not been in active employment for a period not less than one year. The population consists of both males and females with varied characteristics. 3.2. Sample A total of 362 youths were sampled for the study. All respondents were selected from the Greater Accra Region which is the most populous region for unemployed youths. Greater Accra Region is the capital region of Ghana and it attracts most young men who are seeking for greener pastures. The sample consisted of 52.5% unemployed youths (n=190) and 47.5% employed youths (n=172) who served as control. The ages of the participants ranged from 18 years to 35 years (M = 25.72, SD = 4.60). In terms of gender composition, there were 51.9% females (n=188) and 48.1% males (n=174) in the study. In terms of marital status, 81.8% of the respondents were single (n=296), 15.5% of them were married (n=56), and the remaining 2.8% of them were divorced (n=10). Christians constituted 89.8% of the sample (n=325), Muslims constituted 7.7% (n=28), Traditionalists also constituted 1.1% (n=4), and those in other religions made up of 1.4% of the sample (n=5). With regard to educational background, 50.3% of the respondents had tertiary education (n=182), 36.7% had secondary education University of Ghana http://ugspace.ug.edu.gh 39 (n=133), 10.5% had basic education (n=38), and 2.5% had no formal education (n=9). Table 1 below provides description of the sample in terms of frequencies and percentages. Table 1 Demographic Characteristics of the Respondents Variable Frequency (N = 362) Percentage (100%) Gender Males 188 51.9 Females 174 48.1 Age M = 25.72, SD = 4.60 Marital Status Single 296 81.8 Married 56 15.5 Divorced 10 2.8 Religious Affiliation Christianity 325 89.8 Islamic 28 7.7 Traditionalist 4 1.1 Others 5 1.4 Educational Background No formal education 9 2.5 Basic education 38 10.5 Secondary education 133 36.7 Tertiary education 182 50.3 Employment status Unemployed 190 52.5 Employed 172 47.5 Inadequately employed* 94 26.0 Adequately employed* 78 21.5 *Two levels of employment The sample size was determined based on power analysis. Power refers to the sensitivity of a study design to detect true significant findings when using statistical analysis. Power is determined by the significance level chosen, the effect size and the sample size. A small effect size requires a University of Ghana http://ugspace.ug.edu.gh 40 large sample size for significant differences to be detected, while a large effect size requires a relatively small sample to be detected (Anderson, 2003). Power analysis for a MANOVA with two levels and four dependent variables was conducted in G*Power to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, and a large effect size (f = 0.40), a medium effect size (f = 0.25), and a small effect size (f = 0.10) (Faul, Erdfelder, Buchner, A., & Lang, 2013). Based on the aforementioned assumptions, the desired sample size was 20 for large effect size, 44 for a medium effect size, and 264 for a small effect size. In the current study, power analysis for a MANOVA with two levels and four dependent variables was conducted in G*Power to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, and a small effect size (f = 0.10). Based on the aforementioned assumptions, the desired sample size for the present study is 126. In addition, power analysis for a Multiple Regression for moderation analyses was conducted with 1 degree of freedom and 3 predictors in G*Power to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, and a small effect size (f = 0.02). Based on the aforementioned assumptions, the desired sample size for the present study is 395. Although the researcher targeted 395 respondents, only 362 valid cases were obtained for the study. 3.3. Sampling technique Purposive and snowball sampling techniques were used for the sample selection. The purposive sampling technique ensures that participants are selected based on specific characteristics or traits that best suit a study. Here, the researcher purposely searched out for youths who were either employed or unemployed. In order to reach the unemployed youths, the researcher adopted the snowball sampling technique. This technique requires a researcher to first identify one or two members of a group and then use them as informants to reach other members of the University of Ghana http://ugspace.ug.edu.gh 41 group. Since unemployed youths are likely to establish friendship among themselves (Russell, 2009), the researcher used unemployed friends to identify and select a significant number of unemployed youths. Thus, a significant number of unemployed youths was selected through the use of the snowball technique. 3.4. Design Cross-sectional survey design was used in this study. A survey research, according to Punch (2005), attempts to use questionnaires to survey individuals across a broad spectrum in a social research situation. The design is concerned with the relative accuracy with which a target variable is influenced by different situational and demographic variables. The cross-sectional survey design was appropriately chosen for the present study because questionnaires were used to collect data on the research variables from a cross-section of the research population at a single point in time. In addition, the design was deemed appropriate because the study involved respondents of different sex, age, employment status and socio-economic background. 3.5. Measures Questionnaires were used as the instrument for data collection on the research variables. Each questionnaire had six sections. The first five sections comprised standardized scales that measured the key variables of the study. The last section of the questionnaire measured the demographic variables of the respondents such as gender, age, marital status, religious affiliation, educational background and employment status. Three types of employment status were measured in terms of respondents’ self-ratings. Respondents were asked to indicate whether they are currently fully employed, somehow employed, or unemployed, depending on how they judge their own employment situation. Respondents who indicated somehow University of Ghana http://ugspace.ug.edu.gh 42 employed constituted the category of “inadequate employment”. Respondents who indicated fully employed constituted the category of “adequate employment”. Both categories constituted “employed youths” for the study. If unemployed, respondents were asked to indicate the number of years they have stayed unemployed. This provided the measure for “duration or length of unemployment.” However, for the purpose of scoring, employed youths were assigned the numerical value of “0” as duration of unemployment whilst unemployed youths were assigned actual values that indicated the number of years they have remained unemployed. Details of the scales used in designing the questionnaire are provided below. 3.5.1. Hopelessness Depression Symptom Questionnaire (HDSQ; Metalsky & Joiner, 1997; α =.93) The Hopelessness Depression Symptom Questionnaire (HDSQ; Metalsky & Joiner, 1991) was expressly designed to measure the symptoms of hopelessness depression. The HDSQ is a 32- item self-report measure that allows investigators to examine individual and combined symptoms of hopelessness depression. There are a total of eight subscales and each measures a different symptom of hopelessness depression. Each symptom is measured by a cluster of four items. The symptoms are the following: (a) Motivational Deficit (retarded initiation of voluntary responses; items 1-4); (b) Interpersonal Dependency (items 5-8); (c) Psychomotor Retardation (items 9-12); (d) Anergia (items 13-16); (e) Apathy/Anhedonia (items 17-20); (f) Insomnia (items 21-24); (g) Difficulty in Concentration/Brooding (items 25-28); and (h) Suicidality (items 29-32). This is in line with the direction of Guadagnoli and Velicer (1988) and Stevens (1992) that factors should contain more than three items to be internally reliable and valid. A Cronbach’s alpha score of 0.7 or higher is usually regarded as indicative of acceptable internal reliability (DeVellis, 1991). The alpha coefficients for each subscale in a sample of 435 were: (a) Motivational Deficit (retarded initiation of voluntary responses; alpha University of Ghana http://ugspace.ug.edu.gh 43 = .70); (b) Dependency (alpha = .72); (c) Psychomotor Retardation (alpha = .74); (d) Anergia (alpha = .86); (e) Apathy/Anhedonia (alpha = .75); (f) Insomnia (alpha = .81); (g) Difficulty in Concentration/Brooding (alpha = .80); and (h) Suicidal ideation/impulses (alpha = .86). The alpha coefficient for the full HDSQ was .93 (Metalsky & Joiner, 1997). The HDSQ was adapted for the study. The cluster of four items were modified into a single statement each and was scored on a 4-point Likert scale. Sampled items of the modified HDSQ are “I have stopped trying to get what I want,” “My speech is slowed down,” “I can concentrate as well as usual,” and “I am having impulses to kill myself.” The modified HDSQ was scored as strongly disagree = 1 mark, disagree = 2 marks, agree was assigned 3 marks, and strongly agree = 3 marks. The first 28 items were used to measure depression. The remaining 4 items of the scale (items 29-32) which originally measure suicidality were extracted and added to the Suicidal Behaviors Questionnaire-Revised (SBQ-R) to measure suicidality at the point of scoring in the present study. Total scores of the modified scale (all items inclusive) ranged from 32 (minimum) to 128 (maximum). Lower scores reflected lower level of hopelessness depression and higher scores reflected higher levels of hopelessness depression. 3.5.2. The Suicidal Behaviors Questionnaire-Revised (SBQ-R; Osman et al., 2001; α =.88) The Suicidal Behaviors Questionnaire-Revised (SBQ-R; Osman, Bagge, Gutierrez, Konick, Kopper, & Barrios, 2001) assesses the history of suicide behaviors, suicidal ideation within the past year, frequency of suicidal ideation, previous suicide attempts, and the likelihood of future attempts. Sampled items from the scale are “Have you ever thought about or attempted to kill yourself?” and “How likely is it that you will attempt suicide someday?” Higher risk of future suicidal behavior is associated with higher scores. A cut-off score of 7 has been shown to distinguish suicidal from non-suicidal individuals (Osman et al., 2001). The University of Ghana http://ugspace.ug.edu.gh 44 reported reliability of this measure is 0.88 for adolescent inpatient samples and 0.87 for high school samples (Osman et al., 2001). Among adolescent inpatient sample, Osman et al. (2001) observed a Cronbach's alpha of 0.88 for the full scale. Inter-correlations amongst items ranged from .62 (items 3 and 4) to .70 (items 1 and 2). Total score on the scale differentiated between suicidal and non-suicidal adolescents with an odds ratio of 2.19. They further observed a Cronbach's alpha of 0.87 among a high school sample. Inter-correlations amongst items ranged from .48 (items 3 and 4) to .82 (items 1 and 2). The SBQ-R was scored on a frequency response scale. Responses ranged from never to very often. However, the pattern of responses were not consistent across all items. Total scores ranged from 4 (minimum) to 18 (maximum). Higher scores reflected higher suicidality and lower scores reflected lower suicidality. Items of the SBQ-R were added to the Suicidality subscale of the HDSQ. The merger of the two scales provided a measure for suicidality in the present study. 3.5.3. The Centrality of Religiosity Scale (CRS; Huber & Huber, 2012; α= 0.96) The Centrality of Religiosity Scale (CRS; Huber & Huber, 2012) is a measure of the centrality of religiosity. The CRS consists of 15 items divided into five subscales with three items per subscale. The five subscales are intellect (items 1, 5, and 9), ideology (items 10, 11, and 14), experience (items 3, 7, and 8), private practice (items 4, 6, and 13), and public practice (items 2, 12, and 15). Intellect subscale is purely cognitive, understood as interest in religious issues. Ideology subscale deals with religious beliefs that express the subjectively perceived probability of the existence of transcendent reality. Experience subscale concerns religious experience, understood as a sense of divine presence and intervention in the participant’s life. Private Practice subscale is about prayer which establishes contact with transcendent reality. University of Ghana http://ugspace.ug.edu.gh 45 Public Practice subscale deals with worship which involves the social rooting of religiosity and participation in religious services. Each of the above dimensions has been operationalized by means of three questions focusing on frequency (e.g., How often do you usually pray?) and subjective importance (e.g., How important is personal prayer for you?). In three studies, reliabilities of the individual dimensions ranged from 0.80 to 0.93, and from 0.92 to 0.96 for the whole CRS-15 (Huber, 2007). Responses are given on a 5-point rating scale. In the case of questions about importance, the possible responses are: not at all, not very much, moderately, quite a bit, and very much so. With respect to questions about frequency, responses are indicated by choosing one of the following expressions: never, rarely, occasionally, often, and very often, which is tantamount to scoring between 1 and 5 points. The sum total of points scored in all of the five dimensions range from 15 (minimum) to 75 (maximum). These constitute a measure of the centrality of religiosity. Higher scores reflect higher religiosity and lower scores reflect lower religiosity. 3.5.4. The Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et al., 1988; α = .95) Multidimensional Scale of Perceived Social Support (MSPSS; (Zimet, Dahlem, Zimet, & Farley, 1988) is a 12-item self-report assessment instrument designed to measure levels of perceived social support from three perspectives: family, friends, and significant other. Items 3, 4, 8, and 11 measure support from family. Items 6, 7, 9, and 12 measure support from friends. Items 1, 2, 5, and 10 measure support from a significant other. Sampled items are “My family is willing to help me make decisions, “I have friends with whom I can share my joys and sorrows,” and “There is a special person who is around when I am in need.” University of Ghana http://ugspace.ug.edu.gh 46 Each source of social support is assessed using four specific questions and was rated on a 7- point Likert scale from 1 = (Very Strongly Disagree), 2 = (Strongly Disagree), 3 = (Mildly Disagree), 4 = (Neutral), 5 = (Mildly Agree), 6 = (Strongly Agree), and 7 = (Very Strongly Agree). A total score was obtained by summing all of the items and scores ranged from 12 to 84. Higher scores indicated higher perceptions of social support and lower scores indicated lower perceptions of social support. Internal consistency reported for the scale is between .80 and .95. 3.5.5. Rosenberg’s Self-esteem Scale (Rosenberg, 1965; α = .86) The Rosenberg Self-esteem Scale (Rosenberg, 1965) measures individual’s self-esteem. The scale is a 10-item self-report scale designed to measure global self-esteem with a Cronbach alpha reliability range of .79 to .86. Some items in the scale are “I feel that I have a number of good qualities,” “I feel I do not have much to be proud of,” and “At times I think I am no good at all.” Responses are provided on a 4-point Likert scale ranging from “Strongly Agree” (with 4 marks), “Agree” (3 marks), “Disagree” (2 marks) and “Strongly Disagree” (1). Items 3, 5, 8, 9, and 10 are reverse scored in which a “Strongly Agrees” response attracts 1 mark, “Agree” with 2 marks, “Disagree” with 3 marks, and “Strongly Disagree” with 4 marks. Scores on the scale are continuous. Possible total scores per participant range from 10 (lowest self-esteem) to 40 (highest self-esteem). University of Ghana http://ugspace.ug.edu.gh 47 3.5.6. Automatic Thoughts Questionnaire (ATQ; Hollon & Kendall, 1980; α=.96) The Automatic Thoughts Questionnaire (ATQ; Hollon & Kendall, 1980) is used to measure the frequency of automatic negative thoughts associated with depression. It identifies the covert self-statements usually reported by depressives as being representatives of the kinds of their cognitions. The ATQ contains 30 items. Each item is a negative thought and the respondents are asked to rate the frequency with which they recall experiencing these thoughts during the previous week. Among the items of ATQ are “I’m no good,” “I wish I were a better person” and “There must be something wrong with me.” Frequency ratings are made on a five-point scale from 1 (not at all) to 5 (all the time). Total scores range from 30 (little or no distortions) to 150(maximum distortions). Hollon and Kendall (1980) reported both a split-half, odd-even correlation coefficient of 0.97 and an alpha coefficient of 0.96. The ATQ was employed in the present study to measure cognitive distortions. 3.6. Inclusion and Exclusion Criteria 3.6.1. Inclusion Criteria Youths who are capable of active employment and who are seeking or searching for employment qualified for the study. 3.6.2. Exclusion Criteria Youths with chronic illness, physical or mental disabilities were not allowed to partake in the study. Youths who are engaged in full time studies were also not allowed to partake in the study. Again, youths in apprenticeship were excluded from the study. Finally, youths less than 18 years were excluded from the study based on legal age limit for employment in Ghana. University of Ghana http://ugspace.ug.edu.gh 48 3.7. Procedure 3.7.1. Preliminary Preparation The researcher prepared the research proposal, questionnaires, consent forms and other supporting documents and applied for ethical clearance. The ethical clearance for the study was obtained from the Ethics Committee for Humanities (ECH) in the University of Ghana. In addition, the researcher obtained a letter of introduction from the Department of Psychology of University of Ghana. These letter of clearance and letter of introduction introduced the researcher to prospective research participants and also established the purpose of the study. 3.7.2. Pilot Study The researcher first conducted a pilot study prior to the full-scale study. The purpose of the pilot study was to test the appropriateness of the research instruments. It was also meant to provide information on the kind and amount of resources that will be needed for the full-scale study. Finally, the pilot study was to help estimate time duration for the completion of each questionnaire and the possible duration for the entire data collection exercise for the main work. There were 35 participants involved in the pilot study. The participants were selected from the Greater Accra region. They included males and females within the ages of 18 and 35 years. Participants spent an average of 20 minutes to complete each questionnaire. The data for the pilot study was collected within a week. The researcher appreciated the time and effort of the participants after the completion of the questionnaire. The reliability analysis was conducted on the scales and subscales used in the study. The result of the pilot study led the researcher to modify and adapt some scale items to fit the present study. Summary result of the analysis is presented in Table 2 below. University of Ghana http://ugspace.ug.edu.gh 49 Table 2 Reliability Analyses of Scales and Subscales in the Pilot Study Scales/subscales Number of Items Cronbach’s alpha (α) HDSQ 32 .91 Motivational deficit 4 .41 Interpersonal dependency 4 .69 Psychomotor Retardation 4 .64 Anergia 4 .56 Apathy/Anhedonia 4 .49 Insomnia 4 .88 Difficulty in Concentration/Brooding 4 .71 Suicidality 4 .80 SBQ-R 4 .86 ATQ 30 88 Self-esteem 10 .80 CRS 15 .81 Intellect 3 .58 Ideology 3 .59 Experience 3 .81 Private practice 3 .65 Public practice 3 .74 MSPSS 12 .88 Family 4 .87 Friends 4 .69 Significant other 4 .91 N = 35 3.7.3. Data Collection After the pilot study, the researcher prepared a total of 400 questionnaires and administered them to the participants. The modified questionnaires were used to collect data for the full- scale study. Out of the 400 questionnaires, 374 were retrieved. However, only 362 of them were complete and therefore considered valid for the analysis. University of Ghana http://ugspace.ug.edu.gh 50 The researcher employed the services of three research assistants who were unemployed university graduates. The research assistants were instrumental in the data collection and data entry. The researcher and his assistants located unemployed youths at their ideal locations or homes through diligent search. The employed youths were located at their work organizations. Here, the researcher sought permission from the authorities of the organizations within which the youths were employed. With permission granted, the researcher and the research assistants provided the prospective participants with the informed consent forms to read and decide whether or not they were willing and ready to participate in the study. Oral explanations and clarifications were also provided about the study. Whilst some participants gave written consents, others gave their oral consents. Only those who gave their consents were given the research questionnaires to complete. The researcher allowed ample time for each participant to complete the questionnaire. An average time of 20 minutes were spent on a questionnaire. The researcher and his assistants expressed their gratitude to the participants for their time and energy spent on the study. All the completed questionnaires were assembled together for the purposes of data analyses. 3.7.4. Ethical Issues The researcher adhered to all ethical issues in research in the course of the study. First the researcher sought institutional approval for the study. This was done by obtaining ethical clearance from the Ethics Committees for Humanities in University of Ghana. Second, the researcher obtained participants’ informed consent before their enrolment into the study. Prospective participants were provided with detailed information about the study, including the possible costs and benefits associated with their participation. In addition, the ethics of anonymity and confidentiality were also respected in this study. The ethics of anonymity requires a researcher to avoid disclosure of the identity of research University of Ghana http://ugspace.ug.edu.gh 51 participants. The ethics of confidentiality also requires a researcher to avoid public disclosure of the data or information obtained from a participants, especially when such data is tied to the person’s identity. These two ethical principles were strictly upheld in this study. Finally, the researcher upheld the ethics of beneficence. This ethical principle requires that the research being conducted should be beneficial to the participants and to society in general. The study was used as an advocacy tool to create public awareness to the ills associated with youth unemployment in Ghana. Policies and interventional measures are recommended to address such ills, thereby improving the psychological wellbeing of the youths. University of Ghana http://ugspace.ug.edu.gh 52 CHAPTER FOUR RESULTS 4.1. Statistical Tests for Data Analysis The research data was analyzed in both descriptive and inferential statistics. The descriptive statistics was used for the analyses of demographic data into frequency and percentage scores. Moreover, the researcher used the descriptive statistics for the preliminary analyses of the data in checking for their normality. The normality of the data obtained for the study was analyzed in terms of skewness and kurtosis. Normality was accepted when Skewness and Kurtosis are between -1 and +1 (Tabachnick & Fidell, 2001). The achievement of normality warranted the use of parametric statistical tests for further analyses on the research hypotheses. Finally, the researcher employed the descriptive statistics to generate means and standard deviation scores on the key variables of the study. The main inferential statistical tests used in this study are the Pearson Product Moment Correlation (Pearson r) test, the Multivariate Analyses of Variance (MANOVA) test, and the Linear Regression test. The Pearson r test was used for inter-correlational analyses among the key variables and the measuring instruments in order to establish their reliability coefficients. Primarily, the Pearson r test was used to compute for the coefficients of internal consistency (Cronbach’s α) to establish the reliability of each of the scales and subscales in the questionnaire. Measures were satisfactory if their alpha reliability values range from .70 to 1.00. This is in accordance with the suggestion of Nunnally (1978) that the satisfactory coefficient alpha should be equal or higher than .70 if a set of items can constitute a reliable scale. The MANOVA test was used to analyze the first research hypotheses. The MANOVA test is most appropriate when comparing mutually exclusive groups on two or more dependent University of Ghana http://ugspace.ug.edu.gh 53 variables that are measured at least on the interval scale. In the first hypothesis, two mutually exclusive groups were compared on four dependent variables (i.e., depression, cognitive distortions, self-esteem, and suicidality). The dependent variables were measured on the interval scale. Thus, the adoption of the MANOVA test became most appropriate. The remaining four hypotheses were analyzed with Linear Regression test. The Linear Regression test is the most appropriate test in determining the extent to which several independent variables predict one dependent variable which is measured on at least the interval scale. Again, the Regression test is useful for establishing moderation effects among variables. The researcher employed the Regression test to determine the extent to which duration of unemployment predicts depression, cognitive distortions, self-esteem, and suicidality among the youths. The Regression test was again used for moderation analysis in the fourth hypothesis. The Statistical Package for the Social Sciences (SPSS) version 21.0 was used for the data analysis. The analysis was run in two main stages. The first was the preliminary analysis and second involved testing the hypotheses of the study. The following are the details of the analysis. 4.2 Preliminary Analysis The preliminary analysis was done in three steps. These included reliability analysis of scales and subscales, the analysis of normality in terms of skewness and kurtosis, descriptive analysis, and inter-correlations among the key variables of the study. Coefficient of internal consistency (Cronbach’s α) was first computed to establish the reliability of each of the scales and subscales in the questionnaire. Measures had significant reliability coefficients (see Table 3). Nunnally (1978) suggests that the coefficient alpha should be equal or higher than .70 if a set of items University of Ghana http://ugspace.ug.edu.gh 54 can constitute a reliable scale. Except for subscales, all scale had alpha reliability coefficients higher than .70. Table 3 Reliability Analyses of Scales and Subscales in the Main Study Scales/subscales Number of Items Cronbach’s alpha (α) DSQ 32 .93 Motivational deficit 4 .63 Interpersonal dependency 4 .69 Psychomotor Retardation 4 .78 Anergia 4 .69 Apathy/Anhedonia 4 .63 Insomnia 4 .81 Difficulty in Concentration/Brooding 4 .75 Suicidality 4 .88 SBQ-R 4 .84 ATQ 30 .94 Self-esteem 10 .73 CRS 15 .92 Intellect 3 .69 Ideology 3 .81 Experience 3 .75 Private practice 3 .74 Public practice 3 .77 MSPSS 12 .88 Family 4 .83 Friends 4 .75 Significant other 4 .85 N = 362 Regarding issues of normality, all the variables were normally distributed (see Table 4). Normality is accepted when Skewness and Kurtosis approach 0, and preferably lie between -1 and +1 (Tabachnick & Fidell, 2001). The data therefore supported the use of parametric statistical tests. The next step involved computing the descriptive statistics of the study University of Ghana http://ugspace.ug.edu.gh 55 variable. This was done in terms of minimum scores, maximum scores, means and standard deviations. Results from this analysis are also presented in Table 4 below. Table 4 Descriptive Statistics on Key Variables Variables Minimum Maximum Mean SD Skewness Kurtosis Duration of Unemployment 0 13 4.56 2.149 1.01 2.46 Depression 28 112 49.56 13.61 .99 .77 Cognitive Distortions 30 139 57.61 19.95 .95 .26 Self-esteem 13 50 38.06 7.05 -.54 -.18 Suicidality 7 35 10.40 5.46 .09 .21 Religiosity 15 75 58.98 11.54 .81 .70 Social Support 12 84 54.81 14.60 -.39 -.24 N=362 Finally, the Pearson r correlation inter-correlation analysis was conducted. Here, the researcher correlated all the key variables of the study to determine the strength and direction of their relationships. The correlation matrix revealed that all the variables significantly correlated with each other (see Table 5). Duration of unemployment correlated positively with depression (r = .26, p < .01), cognitive distortions (r = .33, p < .01), suicidality (r = .29, p < .01), and negatively correlated with self-esteem (r = -.24, p < .01), religiosity (r = -23, p < .01), and social support (r = -.12, p < .05). University of Ghana http://ugspace.ug.edu.gh 56 Table 5 Correlations among Key Variables of the Study Variables 1 2 3 4 5 6 7 1. Unemployment duration - 2. Depression .26** - 3. Cognitive distortion .33** .63** - 4. Self-esteem -.24** -.46** -.62** - 5. Suicidality .29** .68** .62** -.47** - 6. Religiosity -.23** -.44** -.50** .47** -.41** - 7. Social support -.12* -.32** -.34** .33** -.28** .27** - **p < .01, *p < .05; N = 362 4.3. Testing of Research Hypotheses Hypothesis 1: There will be poorer psychological health among unemployed youths than employed youths. Hypothesis 1 compares unemployed youths and employed youths on psychological health. Psychological health was measured in terms of depression, cognitive distortions, self-esteem, and suicidality. Given that there are four dependent variables, the MANOVA test was deemed most appropriate for the analysis of the first hypothesis. Table 6 shows the summary results of the MANOVA test on Hypothesis 1. University of Ghana http://ugspace.ug.edu.gh 57 Table 6 Effects of Youth Unemployment on Psychological Health Measure Employment status n Mean SD df F Sig. η2 Depression Employed 172 47.95 12.65 1/360 4.60 .033 .01 Unemployed 190 51.01 14.30 Cognitive distortions Employed 172 54.00 17.84 1/360 11.04 .001 .03 Unemployed 190 60.88 21.20 Self-esteem Employed 172 39.28 6.46 1/360 10.09 .002 .03 Unemployed 190 36.95 7.39 Suicidality Employed 172 9.50 4.65 1/360 9.18 .003 .03 Unemployed 190 11.22 6.00 Pillai's Trace: V=.04; F(4, 357) = 3.65; p = 006, Partial Eta Squared = .04 Using Pillai’s trace, there was a significant effect of unemployment on depression, cognitive distortions, self-esteem, and suicidal ideation (V = .04, F(4, 357) = 3.65, p < 01, Partial η2= .04). Separate univariate ANOVAs on the outcome variables revealed significant effects of unemployment on depression, (F(1, 360) = 4.60, p < .05, Partial η2= .01), cognitive distortions (F(1, 360) = 11.04, p = .001, Partial η2= .03), self-esteem (F(1, 360) = 10.09, p < .01, Partial η2= .03) and suicidality (F(1, 360) = 9.18, p < .05, Partial η2= .03). The mean scores show that unemployed youths were more depressed (M=51.01, SD = 14.30) than employed youths (M=47.95, SD = 12.65). They were also more cognitively distorted (M=60.88, SD=21) than employed youths (M=54.00, SD=17.84), had lower self-esteem (M=36.95, SD=7.39) than employed youths (M=39.28), and had higher level of suicidality (M=11.22, SD=6.00) than employed youths (M=9.50, SD=4.65). These results confirm that there is poorer psychological health among unemployed youths than employed youths. University of Ghana http://ugspace.ug.edu.gh 58 Hypothesis 2: Duration of unemployment will significantly predict poorer psychological health among the youths. Hypothesis 2 examined the extent to which duration of unemployment predicted psychological health (i.e., depression, cognitive distortions, self-esteem, and suicidality). The Simple Liner Regression test was used to run this analysis. Duration of unemployment was regressed on each of the dimensions of psychological health. Summary results of the regression analysis are presented in Tables 7, 8, 9, and 10 for depression, cognitive distortions, self-esteem, and suicidality respectively. Table 7 Effect of Youth Unemployment Duration on Depression Model B S. E. Β (Constant) 47.032 .857 Unemployment duration 1.614 .323 .255*** F(1, 360) =25.017***; R2 =.065; ***p =.000 The results of the simple linear regression test in Table 7 show a significant regression model with duration of unemployment accounting for 6.5% variance in depression among the youths (F(1, 360) = 25.017, R2 = .065, p < .001). Duration of unemployment positively predicted depression among the youths (β = .255, p < .001). University of Ghana http://ugspace.ug.edu.gh 59 Table 8 Effect of Youth Unemployment Duration on Cognitive Distortions Model B S. E. Β (Constant) 52.762 1.225 Unemployment duration 3.098 .461 .334*** F(1, 360)= 45.120***; R2 =.111; ***p =.000 Table 8 presents summary results of the linear regression test on the effect of duration of unemployment on cognitive distortions among the youths. Per the table, there was a significant regression model with duration of unemployment accounting for 11.1% variance in cognitive distortions among the youths (F(1, 360) = 45.120, R2 = .111, p < .001). Duration of unemployment positively predicted cognitive distortions among the youths (β = .334, p < .001). Table 9 Effect of Youth Unemployment Duration on Self-Esteem Model B S. E. Β (Constant) 39.285 .446 Unemployment duration -.784 .168 -.239*** F(1, 360)= 21.827***; R2 =.057; ***p =.000 The linear regression results in Table 9 depicts the effect of duration of unemployment on the self-esteem of the youth. The results reveal a significant regression model with duration of unemployment accounting for 5.7% variance in self-esteem among the youths (F(1, 360) = 21.827, R2 = .057, p < .001). Duration of unemployment negatively predicted self-esteem among the youths (β = -.239, p < .001). University of Ghana http://ugspace.ug.edu.gh 60 Table 10 Effect of Youth Unemployment Duration on Suicidality Model B S. E. Β (Constant) 9.252 .340 Unemployment duration .736 .128 .290*** F(1, 360) = 32.943***; R2 =.084; ***p =.000 Table 10 also displays the linear regression results on the effect of youth unemployment duration on suicidality. The summary results reveal that there was a significant regression model with duration of unemployment accounting for 8.4% variance in suicidality among the youths (F(1, 360) = 32.943, R2 = .084, p < .001). Duration of unemployment positively predicted suicidality among the youths (β = .290, p < .001). In sum, duration of youth unemployment positively predicts depression, cognitive distortions, and suicidality and negatively predicts self-esteem. In effect, there is confirmation of the second research hypothesis that duration of unemployment will significantly predict poorer psychological health among the youths. Testing for moderation The testing of Hypotheses 3 and 4 involved moderating analysis. These two hypotheses sought to find out whether religiosity and social support moderate the relationship between unemployment and psychological health measured in terms of depression, cognitive distortions, self-esteem, and suicidality. Thus, the proposed moderators were religiosity and social support, the independent variable was unemployment, and the dependent variable was psychological health (i.e., depression, cognitive distortion, self-esteem, and psychological health). Unemployment was a categorical variable with two levels (i.e., employed and University of Ghana http://ugspace.ug.edu.gh 61 unemployed). It was therefore dummy coded in which case employed category was assigned the value of “0” since this category served as the baseline. The unemployed category was assigned the value of “1” since it reflected the treatment condition. As a requirement for testing for moderation effect, there should be a relationship between the predictor variable(s) and the criterion variables (Holmbeck, 1997). In testing for Hypotheses 3 and 4, the model of Baron and Kenny (1986) was followed. According to Baron and Kenny (1986), a common framework for illustrating moderating effect is the use of causal path analysis. They proposed three causal paths (a, b, and c) as illustrated on the diagram below. Predictor a b Moderator Criterion c Predictor X Moderators Figure 2. Path diagram for moderation model (Baron & Kenny, 1986) Per the model, path a involves regressing the IV on the DV, path b involves regressing the moderator on the DV, and path c involves regressing the interaction term of the IV x Moderator on the DV. The moderator hypothesis is supported if the interaction term (path c) is significant (Baron & Kenny, 1986). To obtain the interaction term, the independent and the moderating variables are centered to reduce the effect of multicollinearity between predictor variables. It is also helpful when predictors do not have a meaningful zero point. Finally, multilevel models with centered predictors tend to be more stable, and estimates from these models can be treated as more or less independent of each other (Aiken & West, 1991, Field, 2009). In centering, the University of Ghana http://ugspace.ug.edu.gh 62 mean of the variable was subtracted from the individual scores. The results of the moderating analysis are shown in the tables below. Hypothesis 3: Religiosity will moderate the relationship between youth unemployment and psychological health. Table 11 The Moderating Effect of Religiosity in the Relationship between Youth Unemployment and Depression Model B S. E. β Step 1 (Constant) 47.953 1.032 Unemployment 3.057 1.425 .112* Step 2 (Constant) 78.536 3.513 Unemployment 1.775 1.296 .065 Religiosity -.507 .056 -.430*** Step 3 (Constant) 78.323 3.614 Unemployment 1.786 1.298 .066 Religiosity -.504 .058 -.427*** Unemployment x Religiosity -.030 .116 -.012 R2 = .013*, .195***, and .196*** for steps 1, 2, and 3 respectively; ∆R2= .183 and .000 for steps 2 and 3 respectively; ***p < .001, *p < .05 The results of the hierarchical regression in Table 11 show the moderating effect of religiosity in the relationship between youth unemployment and depression. Step 1 of the model shows that youth unemployment significantly accounted for 1.3% variance in depression among the youths (R2 = .013, p < .05). Unemployment positively predicted depression (β = .112, p < .05). In step 2, religiosity accounted for 18.3% variance in depression (∆R2 = .183, p < .001). Religiosity negatively predicted depression among the youths (β = -.427, p < .001). Finally, step 3 shows that the interaction term of unemployment and religiosity did not account for a significant variance in depression among the youths (∆R2 = .000, p > .05). The interaction term University of Ghana http://ugspace.ug.edu.gh 63 could not significantly predict depression (β = -.012, p > .05). In effect, religiosity was not a significant moderator for the relationship between unemployment and depression among the youths. Table 12 The Moderating Effect of Religiosity in the Relationship between Youth Unemployment and Cognitive Distortions Model B S. E. Β Step 1: (Constant) 54.000 1.500 Unemployment 6.879 2.071 .172*** Step 2: (Constant) 104.195 4.943 Unemployment 4.774 1.823 .120** Religiosity -.832 .079 -.482*** Step 3: (Constant) 100.270 5.005 Unemployment 4.983 1.798 .125** Religiosity -.774 .080 -.448*** Unemployment x Religiosity -.549 .161 -.156*** R2 = .030***, .259***, and .282*** for steps 1, 2, and 3 respectively; ∆R2= .229*** and .029*** for steps 2 and 3 respectively; ***p < .001, **p < .01 The hierarchical regression results in Table 12 show the moderating effect of religiosity in the relationship between youth unemployment and cognitive distortions. Step 1 of the model shows that youth unemployment significantly accounted for 3.0% variance in cognitive distortions among the youths (R2 = .030, p = .001). Unemployment positively predicted cognitive distortions (β = .172, p = .001). In step 2, religiosity accounted for 22.9% variance in cognitive distortions (∆R2 = .229, p < .001). Religiosity negatively predicted cognitive distortions among the youths (β = -.482, p < .001). Finally, step 3 shows that the interaction term of unemployment and religiosity significantly accounted for 2.9% variance in cognitive distortions among the youths (∆R2 = .029, p = .001). The interaction term negatively predicted cognitive distortions (β = -.156, p = .001). In effect, religiosity emerged as a significant moderator for the relationship between unemployment and cognitive distortions among the youths. University of Ghana http://ugspace.ug.edu.gh 64 Table 13 The Moderating Effect of Religiosity in the Relationship between Youth Unemployment and Self-Esteem Model B S. E. β Step 1: (Constant) 39.279 .531 Unemployment -2.326 .732 -.165** Step 2: (Constant) 22.336 1.772 Unemployment -1.616 .654 -.115* Religiosity .281 .028 .460*** Step 3: (Constant) 23.238 1.812 Unemployment -1.664 .651 -.118* Religiosity .267 .029 .438*** Unemployment x Religiosity .126 .058 .102* R2 = .027**, .236***, and .246*** for steps 1, 2, and 3 respectively; ∆R2=.209*** and .010* for steps 2 and 3 respectively; ***p < .001, **p < .01, *p < .05 Table 13 displays the hierarchical regression results on the moderating effect of religiosity in the relationship between youth unemployment and self-esteem. Step 1 of the model shows that youth unemployment significantly accounted for 2.7% variance in self-esteem among the youths (R2 = .027, p < .01). Unemployment negatively predicted self-esteem (β = -.165, p < .01). In step 2, religiosity accounted for 20.9% variance in self-esteem (∆R2 = .229, p < .001). Religiosity positively predicted self-esteem among the youths (β = .460, p < .001). Finally, step 3 of the model reveals that the interaction term of unemployment and religiosity significantly accounted for 1.0% variance in self-esteem among the youths (∆R2 = .010, p < .05). The interaction term positively predicted self-esteem (β = .102, p < .05). This means that religiosity emerged as a significant moderator for the relationship between unemployment and self-esteem among the youths. University of Ghana http://ugspace.ug.edu.gh 65 Table 14 The Moderating Effect of Religiosity in the Relationship between Youth Unemployment and Suicidality Model B S. E. β Step 1: (Constant) 9.500 .412 Unemployment 1.721 .568 .158** Step 2: (Constant) 20.847 1.422 Unemployment 1.245 .524 .114* Religiosity -.188 .023 -.398*** Step 3: (Constant) 20.109 1.453 Unemployment 1.284 .522 .118* Religiosity -.177 .023 -.374*** Unemployment x Religiosity -.103 .047 -.107* R2 = .025**, .181***, and .192*** for steps 1, 2, and 3 respectively; ∆R2=.156*** and .011* for steps 2 and 3 respectively; ***p < .001, **p < .01, *p < .05 The hierarchical regression results in Table 14 depicts the moderating effect of religiosity in the relationship between youth unemployment and suicidality. Step 1 of the model shows that youth unemployment significantly accounted for 2.5% variance in suicidality among the youths (R2 = .025, p < .01). Unemployment positively predicted suicidality (β = .158, p < .01). In step 2, religiosity accounted for 15.6% variance in suicidality (∆R2 = .156, p < .001). Religiosity negatively predicted suicidality among the youths (β = -.398, p < .001). Finally, step 3 reveals that the interaction term of unemployment and religiosity significantly accounted for 1.1% variance in suicidality among the youths (∆R2 = .011, p < .05). The interaction term negatively predicted suicidality (β = -.107, p < .05). Thus, religiosity emerged as a significant moderator for the relationship between unemployment and suicidality among the youths. In sum, religiosity was not a significant moderator for the relationship between unemployment and depression. However, religiosity did moderate the relationship between unemployment and the remaining three dimensions of psychological health namely cognitive distortions, self- University of Ghana http://ugspace.ug.edu.gh 66 esteem, and suicidality. This implies that, largely, Hypothesis 3 was supported. Hypothesis 4: Social support will moderate the relationship between youth unemployment and psychological health. Table 15 The Moderating Effect of Social Support in the Relationship between Youth Unemployment and Depression Model B S. E. β Step 1: (Constant) 47.953 1.032 Unemployment 3.057 1.425 .112* Step 2: (Constant) 64.513 2.704 Unemployment 3.164 1.348 .116* Social Support -.303 .046 -.325*** Step 3: (Constant) 64.764 2.716 Unemployment 3.167 1.348 .116* Social Support -.308 .046 -.331*** Unemployment x Social Support .093 .093 .050 R2 = .013*, .118***, and .121*** for steps 1, 2, and 3 respectively; ∆R2= .106*** and .002 for steps 2 and 3 respectively; ***p < .001, *p < .05 Table 15 shows the hierarchical regression analysis on the moderating effect of social support in the relationship between youth unemployment and depression. Step 1 of the hierarchical model shows that youth unemployment significantly accounted for 1.3% variance in depression among the youth (R2 = .013, p < .05). Unemployment positively predicted depression (β = .112, p < .05). In step 2, social support accounted for 10.6% variance in depression (∆R2 = .106, p < .001). Social support negatively predicted depression among the youths (β = -.325, p < .001). Finally, step 3 shows that the interaction term of unemployment and social support accounted for an insignificant variance in depression among the youths (∆R2 = .002, p > .05). The University of Ghana http://ugspace.ug.edu.gh 67 interaction term could not significantly predict depression (β = -.050, p > .05). In effect, social support was not a significant moderator for the relationship between unemployment and depression among the youths. Table 16 The Moderating Effect of Social Support for the Relationship between Youth Unemployment and Cognitive Distortions Model B S. E. Β Step 1: (Constant) 54.000 1.500 Unemployment 6.879 2.071 .172*** Step 2: (Constant) 79.209 3.907 Unemployment 7.041 1.948 .177*** Social Support -.462 .067 -.338*** Step 3: (Constant) 79.020 3.928 Unemployment 7.039 1.950 .176*** Social Support -.458 .067 -.335*** Unemployment x Social Support -.070 .135 -.026 R2 = .030***, .144***, and .145*** for steps 1, 2, and 3 respectively; ∆R2= .114*** and .001 for steps 2 and 3 respectively; ***p < .001 The hierarchical regression results in Table 16 show the moderating effect of social support in the relationship between youth unemployment and cognitive distortions. Step 1 of the model shows that youth unemployment significantly accounted for 3.0% variance in cognitive distortions among the youths (R2 = .030, p = .001). Unemployment positively predicted cognitive distortions (β = .172, p = .001). In step 2, social support accounted for 11.4% variance in cognitive distortions (∆R2 = .114, p < .001). Social support negatively predicted cognitive distortions among the youths (β = -.338, p < .001). Finally, step 3 shows that the interaction term of unemployment and social support accounted for insignificant variance in cognitive distortions among the youths (∆R2 = .001, p > .05). The interaction term did not predict cognitive distortions (β = -.026, p > .05). In effect, social support did not emerge as a significant University of Ghana http://ugspace.ug.edu.gh 68 moderator for the relationship between unemployment and cognitive distortions among the youths. Table 17 The Moderating Effect of Social Support in the Relationship between Youth Unemployment and Self-Esteem Model B S. E. β Step 1: (Constant) 39.279 .531 Unemployment -2.326 .732 -.165** Step 2: (Constant) 30.637 1.388 Unemployment -2.382 .692 -.169*** Social Support .158 .024 .328*** Step 3: (Constant) 30.568 1.395 Unemployment -2.383 .693 -.169*** Social Support .160 .024 .331*** Unemployment x Social Support -.026 .048 -.026 R2 = .027**, .135***, and .135*** for steps 1, 2, and 3 respectively; ∆R2= .107*** and .001 for steps 2 and 3 respectively; ***p < .001, **p < .01 Table 17 displays the hierarchical regression results on the moderating effect of social support in the relationship between youth unemployment and self-esteem. Step 1 of the model shows that youth unemployment significantly accounted for 2.7% variance in self-esteem among the youths (R2 = .027, p < .01). Unemployment negatively predicted self-esteem among the youths (β = -.165, p < .01). In step 2, social support accounted for 10.7% variance in self-esteem (∆R2 = .107, p < .001). Social support positively predicted self-esteem among the youths (β = .328, p < .001). Finally, step 3 of the model reveals that the interaction term of unemployment and social support did not account for a significant variance in self-esteem among the youths (∆R2 = .001, p > .05). The interaction term failed to predict self-esteem among the youths (β = -.026, p > .05). This means that social support was not a significant moderator for the relationship between unemployment and self-esteem among the youths. University of Ghana http://ugspace.ug.edu.gh 69 Table 18 The Moderating Effect of Social Support in the Relationship between Youth Unemployment and Suicidality Model B S. E. β Step1: (Constant) 9.500 .412 Unemployment 1.721 .568 .158** Step 2: (Constant) 15.190 1.095 Unemployment 1.758 .546 .161*** Social Support -.104 .019 -.279*** Step 3: (Constant) 15.132 1.101 Unemployment 1.757 .546 .161*** Social Support -.103 .019 -.276*** Unemployment x Social Support -.021 .038 -.028 R2 = .025**, .103***, and .103*** for steps 1, 2, and 3 respectively; ∆R2= .078*** and .001 for steps 2 and 3 respectively; ***p < .001, **p < .01 The hierarchical regression results in Table 18 depicts the moderating effect of social support in the relationship between youth unemployment and suicidality. Step 1 of the model shows that youth unemployment significantly accounted for 2.5% variance in suicidality among the youths (R2 = .025, p < .01). Unemployment positively predicted suicidality (β = .158, p < .01). In step 2, social support accounted for 7.8% variance in suicidality (∆R2 = .078, p < .001). Social support negatively predicted suicidality among the youths (β = -.279, p < .001). Finally, step 3 reveals that the interaction term of unemployment and social support accounted for insignificant variance in suicidality among the youths (∆R2 = .001, p > .05). The interaction term did not predict suicidality (β = -.028, p > .05). Thus, social support was not a significant moderator for the relationship between unemployment and suicidality among the youths. In sum, social support was not a significant moderator for the relationship between youth unemployment and all four dimensions of psychological health namely depression, cognitive distortions, self-esteem, and suicidality. In this way, Hypothesis 4 was not confirmed. University of Ghana http://ugspace.ug.edu.gh 70 4.4. Summary of Findings The following are the main findings obtained from the study. 1. There was poorer psychological health among unemployed youths than employed youths. Compared to employed youths, unemployed youths had higher levels of depression, cognitive distortions, and suicidality and lower level of self-esteem. 2. Duration of unemployment significantly predicted poorer psychological health among the youths. Longer duration predicted higher levels of depression, cognitive distortions, and suicidality and lower level of self-esteem. 3. Religiosity moderated the relationship between youth unemployment and psychological health except depression dimension. 4. Social support did not moderate the relationship between youth unemployment and psychological health. 4.5. Observed Conceptual Model Psychological Health β = .11* β = .17*** β = -.16*** β = -.17** β = .10* β = .16** β = -.11* ***p < .001, **p < .01, *p < .05 Figure 3. Observed model on the psychological effects of youth unemployment and the moderating role of religiosity Depression Cognitive distortions Self-esteem Suicidality Unemployment Religiosity University of Ghana http://ugspace.ug.edu.gh 71 The model in figure 3 shows that unemployment positively predicted depression, cognitive distortions, and suicidality but negatively predicted self-esteem. Religiosity significantly moderated the relationship between youth unemployment and depression, cognitive distortions, and suicidality. However, religiosity did not moderate the relationship between youth unemployment and depression. University of Ghana http://ugspace.ug.edu.gh 72 CHAPTER FIVE DISCUSSION, RECOMMENDATIONS, AND CONCLUSION 5.1 Discussion The purpose of the study was to investigate the psychological effects of unemployment among the youths in Ghana and the factors that serve as buffers within the Ghanaian context. The objectives of the study were to examine the impact of unemployment on the psychological health among the youths, to ascertain the impact of duration of unemployment on psychological health among the youths, and to examine the buffering effects of religiosity and social support on the relationship between unemployment and psychological health. Psychological health was defined in terms of depression, cognitive distortions, self-esteem, and suicidality. Four hypotheses were formulated to test the research objectives. The findings obtained are discussed below. 5.1.1. Effects of youth unemployment on psychological health The first hypothesis predicted that there will be poorer psychological health among unemployed youths than employed youths. Confirmatory evidence was obtained from the data analysis. Consistent with prediction, there was poorer psychological health among unemployed youths than employed youths. Compared to employed youths, unemployed youths had higher levels of depression, higher cognitive distortions, higher suicidality and lower level of self- esteem. These results confirm the causation hypothesis which predicts negative impact of unemployment on mental health (Schaufeli, 1997). Earlier studies (e.g., Feather, 1990; Ezzy, 1993; Winefield, 1995; Winkelman & Winkelman, 1998; Dooley et al., 2000) and recent studies (Breslin & Mustard, 2003; Bjarnason & Sigurdadottir, 2003; Kroll, & Lampert, 2009; University of Ghana http://ugspace.ug.edu.gh 73 Lorenzini, & Giugni, 2010) have shown quite convincingly that unemployment leads to psychological distress and that re-employment improves mental health. This points to a greater likelihood for causal relationship between unemployment and poor mental health. In Ghana, the youth unemployment situation is pathetic. The freeze on employment coupled with the economic crisis makes life unbearable for the youth who are unemployed. Making earns-meat becomes nearly impossible for many young men and women who do not have reliable source of support. The economic hardships in the country and the financial constraint of the unemployed youth might largely account for their poor mental health in terms of lower self-esteem, and higher levels of depression, cognitive distortions, and suicidality. Unemployment is therefore an undesirable situation which needs to be controlled in order to restore good psychological health among the youth. Beyond Ghana, unemployment is known to negatively affect mental health. In its Employment Outlook, the Organization for Economic Co-operation and Development (OECD, 2008) observed among five countries (Australia, Canada, Korea, Switzerland and U.K.) that moving from employment to unemployment or inactive status (out of the labour force) had a large, negative impact on mental health, with a larger impact on men than women. In Canada, Australia, the U.K. and Switzerland, the increase in mental distress was greatest when there was a change from employment to inactivity due to illness. A movement from employment to unemployment also had a significantly negative impact on mental health. However, the small effect size observed in the present study confirms the two explanations provided by Schaufeli (1997). First, Schaufeli (1997) noted that the strength of the relationship between unemployment and psychological distress is rather weak. Roughly speaking only 10 to 15% of the variance in distress is explained by employment status (Fryer and Payne, 1986). Second, the fact that unemployment causes psychological distress does not rule out the University of Ghana http://ugspace.ug.edu.gh 74 possibility that high levels of distress might lead to prolonged unemployment. In fact, small effect size was observed on all four dimensions of psychological health (i.e. depression, cognitive distortions, self-esteem, and suicidality). Whilst unemployed youth suffer high levels of depression and lower self-esteem, their cognition may be marred with negative thoughts and distorted view of the self. In the view of Joiner, Alfano, and Metalsky (2003), individuals suffering from negative life events often possess negative self-schemas – negative conceptions of their own traits, abilities, and behaviour. As a result, such individuals tend to have lower psychological wellbeing. They are usually depressed, anxious, and stress. Unemployed youths are likely to adopt negative framework for the interpretation of their circumstance and such interpretation is likely to make them become even more depressed. It is therefore not surprising to witness that unemployed youths had higher depression and cognitive distortions coupled with lower self-esteem. To a greater extent, the findings of the study provide empirical support to both the Reformulated Learned Helplessness Theory of Depression (Abramson et al., 1978; Abramson et al., 1989; Peterson & Barrett, 1987; Seligman et al., 1990) and the Hopelessness Theory of Depression (Abramson, Metalsky, & Alloy, 1998). The Reformulated Learned Helplessness Theory of Depression explains affective, motivational and cognitive deficits observed in humans following exposure to uncontrollable negative events such as unemployment. According to the reformulated learned helplessness model, a pessimistic attributional style consists of attributing negative events to internal, stable, and global causes. This negative style is related to a helpless reaction. Reactions that accompany helplessness include passivity, sadness, anxiety, hostility, and low self-esteem (Peterson & Seligman, 1984). The theory predicts that an internal, stable and global attributional tendency, termed a depressogenic attributional style, is a risk factor for the development of a depressive reaction following a negative life event such as chronic illness. University of Ghana http://ugspace.ug.edu.gh 75 Similarly, the Hopelessness Theory of Depression (Abramson et al., 1998) explains that individuals with maladaptive, or negative, cognitive styles are vulnerable to depression when they encounter negative life events such as unemployment. Such individuals tend to assign a negative meaning or consequences to the negative event. Those who make global and stable attributions, make negative self-inferences, and expect negative consequences following the occurrence of a negative life event are more likely to become depressed. Individuals with the negative cognitive style develop a greater risk for depression and suicide (Abramson et al., 1998). The above theoretical explanation points out that once unemployed youths develop distorted views of the self, hopelessness depression becomes inevitable. It is therefore not surprising that as significant positive relationship was found between cognitive distortions and depression in the study. This confirms the theoretical view that the adoption of depressogenic attributional style is likely to lead to lower psychological wellbeing. There is therefore the need to help shape the cognition of unemployed youths to adopt a more positive framework for the assessment of themselves. Negative views of the self will only predict greater levels of depression. It is in this direction that Schaufeli (1997) recommended the adoption of positive attitude and an active way of dealing with unemployment, instead of brooding over one’s weakness. 5.1.2. Effect of duration of youth unemployment on psychological health The second hypothesis predicted that duration of unemployment will significantly predict poorer psychological health among the youths. In line with prediction, it was observed that duration of unemployment significantly predicted poorer psychological health among the youths. Longer duration predicted higher levels of depression, cognitive distortions, and suicidality and lower level of self-esteem and vice versa. University of Ghana http://ugspace.ug.edu.gh 76 The second finding shows that the duration of unemployment predicts more psychological distress than the mere form of unemployment. This implies that it is not only unemployment that is worrying but the duration as well. The negative consequences of unemployment increase with the duration of unemployment. As the years run, the economic difficulties of unemployed youths get worse and their psychological problems increase alongside. This is especially true in the Ghanaian context where the national economy is steadily declining with no hope of rescue. The unemployed youths in Ghana therefore have no hope for improvement over their situations. The loss of hope can only predicts higher levels of depression, cognitive distortions, and suicidality as well as lower levels of self-esteem. The literature highlights the negative consequences of sustainable unemployment. Lorenzini and Giugni (2010) observed that long-term unemployment has a number of important consequences on the personal life of young people on three counts: producing financial distress, creating anxiety-related health problems, and diminishing the overall level of happiness. Young long-term unemployed are not very well on all three aspects when compared to youth who have a regular job. Again, in their economic outlook, OECD (2008) provided evidence from five countries (Australia, Canada, Korea, Switzerland and U.K.) to show that duration of unemployment mattered, but the impact varied across countries. For example, in the U.K., there was evidence of a “habituation” effect. Psychological distress was greater for those just unemployed or inactive than for those who had been unemployed or inactive for over two years. However, in Australia, long-term unemployment worsened mental health for males. In the present study, the “habituation” effect was not the case among unemployed Ghanaian youths. Instead, the Australian observation that long-term unemployment worsened mental health was the case. This also shows that the effect of duration of unemployment is located in specific economies, therefore unemployment research must be placed in specific social and cultural context as suggested by Winefield and Fryer (1996). University of Ghana http://ugspace.ug.edu.gh 77 5.1.3. Buffering effect of religiosity on the relationship between youth unemployment and psychological health Recently, the field of psychology has begun to display a growing interest in religious coping methods and their implications for health and well-being. Empirical studies have yielded an interesting picture of the relationship between religious coping and physical and mental health. Coupled with this growing interest is the ascription of high religiosity among Ghanaians. It was on these grounds that the current study investigated the buffering effect of religiosity in the psychological effects of youth unemployment. It was predicted in the third hypothesis that religiosity will moderate the relationship between youth unemployment and psychological health. Except for depression, religiosity moderated the relationship between youth unemployment and psychological health. Religiosity was measured in holistic aspects involving the intellect, ideology, personal experience, private practice, and public practice. Previous studies have suggested that increased religious participation leads to enhanced well-being over time (Strawbridge, Shema, & Cohen, 2001). Substantial literature demonstrates positive effects of religious beliefs on psychological well-being. Religiosity significantly predicts psychological wellbeing among individuals with acute forms of stress (Lazarus & Folkman, 1984; Lakey & Cohen, 2000). Psychological well- being is deeply related to the individual’s religious beliefs, which offer a rich source of recourse to consider (Joshi, Kumari, & Jain, 2008). Religiosity plays a major part in the life of an individual. It can provide hope in despair. In daily life, people report that they are able to experience deep peace even in the midst of mental distress (Underwood & Teresi, 2002). According to Moberg (2009), happiness is greater and psychological stress is lower for those who attend religious services regularly. In Ghana, it is not rare to find most youths crowding the churches, seeking for divine interventions to their problems. Paramount among these problems is unemployment. Perhaps, the decision to attend University of Ghana http://ugspace.ug.edu.gh 78 frequent church services and mosque rests on the belief of the youth that religiosity/ spirituality can help them overcome their unemployment problem. The study indeed reveals religiosity as an effective coping mechanisms help to reduce psychological distress associated with unemployment. Religion may mitigate psychological distress by influencing the cognitive and behavioral responses for interpreting and handling negative life events confronting unemployed youths. According to literature, religion appears to influence the interpretation, appraisal, and attribution of stressful events (Gordon et al., 2002; Siegel et al., 2001). It contributes to the coping process by providing coping options through the social, interpersonal, cognitive, spiritual, and behavioral aspects of religious faith (Hathaway & Pargament, 2001). This explains why religiosity could moderate the relationship between unemployment and cognitive distortions, self-esteem, and suicidality. Perhaps, it is for this reason that unemployed youths are more religious, prayed more often, and engage in ritual activities more. The finding provides evidence in support of the Afrocentric framework of stress and coping which emphasizes collective and communal orientation in coping among Africans and African Americans. Indeed, collective religious beliefs and collective practices impact the coping process (Utsey et al., 2000, 2004). Communally and spiritually based coping are particularly useful among individuals of African descent, reflecting an Afrocentric worldview. These observations have found support in a recent coping study of Black Canadians (Constantine et al., 2005; Joseph & Kuo, 2009). Joseph and Kuo (2009) for instance reported that spiritual- and ritual-centered coping constituted the most crucial coping strategies adopted by Black Canadians in dealing with interpersonal discrimination (e.g., being looked down on as unintelligent by others). Constantine et al. (2005) also found that both acquiring from and giving support to in-group members and religious coping were an integral part of coping among African Americans. The overwhelming enrollment of unemployed youths in Ghana in most University of Ghana http://ugspace.ug.edu.gh 79 churches also supports the view that Ghanaian unemployed youths seek for collective and communal worship as well as spiritual- and ritual-centered as most crucial coping strategies to address the negatives associated with unemployment. The present study confirms that religiosity is an effective coping mechanism against the negative consequences of youth unemployment except for depression. The exception can be explained on the basis of the assertions of Greenfield and Nadine (2007) that associations between more frequent formal religious participation and psychological well-being are largely contingent upon the dimension of psychological well-being. Furthermore, findings from previous studies that simultaneously have examined multiple dimensions of psychological well-being suggest that different patterns of association between religiosity and well-being are likely to emerge across diverse dimensions of psychological wellbeing (Maselko & Kubzansky, 2006). Based on these empirical evidences it is plausible that religiosity could only moderate specific aspects of psychological health and not psychological health in general. However, even with depression where religiosity was an insignificant moderator, religiosity still emerged as a significant predictor. It was observed that religiosity negatively predicted depression among the youths. This finding indicates that irrespective of one’s employment status, religiosity remains an important factor to reduce depression. 5.1.4. Buffering effect of social support on the relationship between youth unemployment and psychological health The role of social support in psychological distress is well established in the literature (Carlson & Perrewe, 2009). Some studies suggest that regardless of stress level, social support has a positive effect (Thoits, 2003), whereas others suggests that the degree of social support an individual has in a situation may determine the stress level (Carlson & Perrewe, 2009). The fourth hypothesis predicted that social support will moderate the relationship between youth University of Ghana http://ugspace.ug.edu.gh 80 unemployment and psychological health. Contrary to prediction, social support did not moderate the relationship between youth unemployment and psychological health, though it was found to be a significant predictor for all four dimensions of psychological health. Social support negatively predicted depression, cognitive distortions, and suicidality and positively predicted self-esteem. The fourth finding shows that though social support influences psychological health of the youth, it is not exceptionally relevant to unemployed youths in Ghana. The lack of moderation depicts the lack of interaction between social support and employment status in influencing psychological health. This conclusion is consistent with studies that suggest that though social support is beneficial to health while facing stressful events, it cannot prevent all damaging effects (Burns, Anstey, & Windsor, 2011; Lakey & Cohen, 2000; Kaul & Lakey, 2003; Gjesfjeld, Greeno, Kim, & Anderson, 2010). The study thus justifies the view of Burns, Anstey, and Windsor (2011) that the availability of social support plays an important role in influencing the psychological wellbeing of individuals. The finding establishes the fact that social support has a significant positive effect on psychological health without regard to employment status. Higher levels of social support predicted lower levels of depression, cognitive distortions, and suicidality as well as higher self-esteem. Lower levels of lower social support on the other hand predicted higher levels of depression, cognitive distortions, and suicidality and lower self-esteem. The above observation is consistent with the significant number of literature that highlights the positive role of social support on psychological wellbeing in organizational, epidemiological and clinical contexts across the lifespan (Burns & Machin, 2012; Huppert & Whittington, 2003; Ruini, Belaise, Brombin, Caffo, & Fava, 2006). The observation has some practical implications. It suggests the need to encourage individuals to establish strong links and University of Ghana http://ugspace.ug.edu.gh 81 affiliations with friends, families and significant others. By so doing, individuals will be able to broaden their social support base and be in a position to enjoy its benefits. The finding also creates the awareness of families and the larger society to identify provision of social support as a critical means of ensuring the psychological wellbeing of their relatives suffering from poor mental health. The fourth finding however failed to confirm the buffer theory of social support (Alloway & Bebbington, 1987; Lin, Woefel & Light, 2005) which proposes that the presence of social relations or support networks moderates effect of adverse environmental stressors that precipitate poor psychological health. The current finding thus could not proof the buffering effect of social support in negative life situations of mental health among unemployed youths in Ghana. Indeed, in various ways social support is consistently identified as cushioning the effects of life events on wellbeing outcomes in clinical samples (Ames & Roitzsch, 2000) and even the general population (Falcon, Todorova, & Tucker, 2009). The inability to prove this point in the present study requires further investigation in this regard. Future studies can help to ascertain the buffering effect of social support in the psychological effects of youth unemployment. The lack of moderating effect of social support in the psychological effects of youth unemployment suggests that one does not necessarily experience negative life events before benefiting from the positive impact of social support. Social support is important to all individuals irrespective of the nature of one’s life situation. The present finding regarding the lack of moderating effect social support in the psychological effects of youth unemployment is not consistent with the finding of Lorenzini and Giugni (2010) who observed a significant impact of the social support on the personal welfare of young long-term unemployed in Switzerland. Such an impact, however, varied according to the type University of Ghana http://ugspace.ug.edu.gh 82 of social support and the specific aspect of well-being under study. While no effect was observed on financial distress, partner support seems to reduce anxiety-related problems, and finally all three types of social support contributed in some way to increase the level of happiness of young unemployed. It must however be noted that the present study investigated psychological health in terms of depression, cognitive distortions, self-esteem, and suicidality. The outcome variable of Lorenzini and Giugni (2010) were financial distress, anxiety, and unhappiness. The differences in outcome variables might explain the disparity in findings between the present study and that of Lorenzini and Giugni (2010). 5.2. Limitations of the Study The study failed to examine the psychological effects of quality employment among the youth. The definition and use of ‘employment’ in the present study was narrowed and not reflective of the definition suggested by Dooley et al. (2000). According to Dooley et al. (2000), it is less useful to simply compare the unemployed with the employed, because paid jobs can differ greatly in quality. Poorer quality jobs are more likely to be associated with mental health problems than better quality jobs. This implies that there is the need to assess the psychological effects of quality employment together with unemployment. However, this could not be done in the present study due to inappropriate sample compositions for these categories of youths. Whilst unemployed youth constituted 52.5% (n=190) of the total sample of 362 youths, inadequately employed youths constituted 26.0% (n=94) and adequately employed youths constituted 21.5% (n=78) of the sample. Thus, the uneven distribution of adequately employed, inadequately employed, and unemployed youths undermined comparative analysis in this regard. University of Ghana http://ugspace.ug.edu.gh 83 The age range of the sample is considerably large (18 years to 35 years) and is potentially inclusive of 2 to 3 developmental life stages which may pose a problem in both the categorization of the entire sample as “youth” as well as the meaning of employment to individuals at different stages of their life. Moreover, the social support measure (MSPSS) may be too generalized and in need of being more specific to employment. Social support is a multidimensional variable inclusive of types and sources of support. Generalized social support may be of limited value in understanding its relationship with other psychosocial constructs. The use of generalized social support measure in the present study might have accounted for the lack of moderating effect of social support for the relationship between unemployment and psychological health. Another limitation emerges from the survey research design employed in the study. Due to the nature of the survey design used in this study, strict causal relationships cannot necessarily be established. In other words, since the survey research design does not allow of the control of extraneous variables as in the case of experiment, the researcher cannot certainly assume that the independent variables solely accounted for the variations in psychological health among the youths. Thus, all inferences or conclusions drawn in this study only depict descriptive relationships but not causal relationship. In this regard, it becomes difficulty to validly prove the causality hypothesis in unemployment research. According to Winefield and Fryer (1996), sorting out the effects of unemployment on mental health is complicated by the fact that the cause-and-effect relationship can work in both directions: unemployment may worsen mental health, and mental health problems may make it more difficult for a person to obtain and/or hold a job. The latter is referred to as the “selection” effect. However, where data are available on the same individuals over time (longitudinal data), it is possible to use statistical methods that sort out the causation. For University of Ghana http://ugspace.ug.edu.gh 84 example, researchers can control for the mental health of individuals before job loss to examine the effects of job loss on mental health. The present study could not do so due to the use of the cross-sectional survey design. The present study only looked at the relationship between unemployment and psychological health at a point in time, in which case the findings can be difficult to interpret with causal inferences. 5.3. Recommendations The recommendations from the study are in two-folds. First, the researcher articulates practical recommendations based on the findings from the study. Second, the researcher offers theoretical recommendations to shape the conduct of future research on the psychological effects of youth unemployment. 5.3.1. Recommendations for Practice The following are practical recommendations aimed at solving the problem of unemployment and improving psychological health among the youths. First, there is the need to make the youths employable by bridging the gap between theory and practice. The onus lies on academic institutions to design and structure their course content to fit local needs in national context. It is only by this that graduates can acquire and use requisite skills to serve society through innovations and self-creation of opportunities. Academic institutions should also place more emphasis on practical training, especially with industrial attachment, to expose graduates to adequate practical work to provide them with work experience prior to the completion of their academic programme. In addition, course structure should be reviewed regularly in line with demands of industry and the country’s development goals. University of Ghana http://ugspace.ug.edu.gh 85 There is also the need for proactive interventional measures that seek to create more job opportunities for the youth. Government should encourage private sector partnership in job creation. This can be a good step to create more job opportunities to absorb the army of graduates in search for non-existing jobs in the country. This however can best be described as an immediate interventional measure with short-term effect. Since agricultural remains the backbone of the economy, government ought to come out with policies and incentives that will attract the youths in the agricultural industry. Students should be re-oriented to see all jobs as important so as to strive to make the best out of them. Particularly, Agriculture can be made a core course for both basic and secondary education. In this way, most citizens will be able to grasp the basic skills and technologies useful in boosting agricultural production. Moreover, local industries need to be promoted to absorb most of the unemployed youths in the local communities. As Amankrah (2000a & b) suggested, there is the need to formalize and promote community-based apprenticeship training schemes in all districts with the support of the Government to take care of the youth who have dropped out or have not been to school. Technical institutes or vocational institutes need to be equipped to undertake apprenticeship in the formal or informal industry. In addition, the registration of apprenticeship providers, the standardization of content, duration of training, and certification in conformity with industry and identifiable trade associations will make apprenticeship and trade become attractive to the youth. This no doubt will be a significant boost to the local industry and increase national productivity in long run. Since financial constraints are among the pathways from unemployment to mental health difficulties, adequate access to and levels of employment insurance benefits can be helpful in University of Ghana http://ugspace.ug.edu.gh 86 reducing the incidence of mental health problems among the unemployed and, thereby, facilitating re-employment. Based on the first research finding that unemployed youths experience poorer psychological health than employed youths, there is the need to extend clinical services to unemployed youths. Unemployed youths are known to suffer lower self-esteem and higher depression, cognitive distortions, and suicidality. The significant association among these mental health variables makes it critical to provide assessable and timely clinical interventional services to the unemployed. The researcher proposes the use of cognitive-behaviour therapy as an effective treatment option to shape the cognition of unemployed youths and to make them become less depressed. There is the need to for clinical psychologists to help the unemployed to replace their distorted thoughts with more rational and logical thoughts in order to maximize their psychological health. They need to develop positive views of themselves and trust in their potentials. Particularly, unemployed youths with higher suicidal tendencies need to be identified and assisted to prevent future calamities. 5.3.2. Recommendation for Future Research On the bases of the findings and the identified limitation of this study, the following recommendations are made for future research consideration. First, based on the direction provided by Dooley et al. (2000), it is necessary for future researchers to assess the psychological effects of quality employment by comparing the unemployed, the inadequately employed, and the adequately employed youths on psychological health. Adequately employed individuals should comprise youths whose jobs and salaries commensurate with their qualifications and who are satisfied with their jobs. University of Ghana http://ugspace.ug.edu.gh 87 Inadequately employed individuals should comprise youths whose jobs and salaries do not commensurate their qualifications. Such individuals are dissatisfied with their jobs and may be seeking for alternative jobs where they can find satisfaction. Similarly, in order to clearly determine the effect size of youth unemployment, there is the need to distinguish two categories of unemployment. These are the ‘never employed before’ and ‘previously employed.’ The former refers to individuals who have never had any chance to work for salary or who have never been engaged in a job that provides regular income. The latter refers to individuals who were previously employed but have lost their jobs. This is necessary because previous jobs could provide certain accumulated benefits for the unemployed youths and this is likely to obscure the psychological effects of youth unemployment. Despite the overwhelming evidence social relations or support networks moderates the effect of adverse life events that precipitate poor psychological health (Alloway & Bebbington, 1987; Lin, Woefel & Light, 2005), the present study could not proof this buffering effect of social support in negative psychological effects of unemployment among the youths in Ghana. To be able to unveil the buffering effect of social support, there is the need for future researchers to examine specific dimensions of social support as moderators for the psychological effects of youth unemployment. Useful dimensions to consider may be social support from friends, family, and significant others (Zimet et al., 1988). University of Ghana http://ugspace.ug.edu.gh 88 5.4. Conclusion The study has clearly revealed the psychological effects of unemployment among the youths in Ghana and the factors that serve as buffers within the Ghanaian context. Consistent with expectation, there was poorer psychological health among unemployed youths than employed youths. Compared to employed youths, unemployed youths had higher levels of depression, cognitive distortions, and suicidality and lower level of self-esteem. Additionally, duration of unemployment significantly predicted poorer psychological health among the youths. Longer duration predicted higher levels of depression, cognitive distortions, and suicidality and lower level of self-esteem. Moreover, except for depression, religiosity emerged as a significant moderator for the relationship between youth unemployment and psychological health. Though social support significantly predicted psychological health, it could not moderate the relationship between youth unemployment and psychological health. In effect, the buffer theory of social support was not supported. Given the negative psychological consequences of youth unemployment, there is a need for the government of Ghana to plan and implement pragmatic youth programme that will contain Ghanaian youths in search for job. For the meantime, unemployed youths need to be encouraged to actively participate in religious activities in order to reap its rich benefits. University of Ghana http://ugspace.ug.edu.gh 89 REFERENCES Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A theory- based subtype of depression. Psychological Review, 96, 358-372. Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49 – 74. Adinkrah, M. (2014). Confessions: suicidal ideation on a Ghanaian radio program. Journal of public health and epidemiology, 6(7), 229-234. Alloway, R., & Bebbington, P. (1987). The buffer theory of social support – A review of the literature. Psychological Medicine, 17(1), 91–108. Amankrah, J. Y, (2000a). Policy Focus for Improving the Productivity of the Informal Sector in Ghana: Report prepared for NACVET as background paper towards the development of a national policy for Vocational and Technical Education in Ghana, December 2000. Amankrah, J. Y. (2000b). Improving Productivity in the Informal Sector in Ghana” Policy Issues and Strategies for Training in the Informal Sector” Report prepared for The World Bank, December 2000. Amankrah, J. Y. (2005). Ghana decent work statistical indicators: Fact finding study. ILO Working Paper, ILO, Geneva, September 2005. Ames, S. C., & Roitzsch, J. C. (2000). The impact of minor stressful life events and social support on cravings: A study of inpatients receiving treatment for substance dependence. Addictive Behaviors, 25(4), 539–547. Anderson, T. W. (2003). An introduction to multivariate statistical analysis. New York: Wiley. Åslund, C., Larm, P., Starrin, L., & Nilsson, K. W. (2014). The buffering effect of tangible social support on financial stress: influence on psychological well-being and psychosomatic symptoms in a large sample of the adult general population. International Journal for Equity in Health, 13, 85 Bergin, A. E. (2001). Values and religious issues in psychotherapy and mental health. American Psychologist, 46, 394 - 403. University of Ghana http://ugspace.ug.edu.gh 90 Bjarnason, T., & Sigurdardottir, T. J. (2003). Psychological distress during unemployment and beyond: social support and material deprivation among youth in six northern European countries. Social Science & Medicine, 56, 973–985 Blume, A. W., Resor, M. R., Villanueva, M. R., & Braddy, L.D. (2009). Alcohol use and comorbid anxiety, traumatic stress, and hopelessness. Addictive Behaviors, 34, 709–713. Breslin, F. C., & Mustard, C. (2003). Factors influencing the impact of unemployment on mental health among young and older adults in a longitudinal, population-based survey. Scandinavian Journal of Work, Environment and Health, 29(1), 5 – 14. Burns, R. A. & Machin, M. A. (2013). Psychological wellbeing and the diathesis-stress hypothesis model: The role of psychological functioning and quality of relations in promoting subjective well-being in a life events study. Personality and Individual Differences 54, 321–326 Burns, R. A., & Machin, M. A. (2012). Moving beyond the pleasure principle: Within and between-occasion effects of employee eudaimonia within a school organizational climate context. Journal of Vocational Behavior, 80(1), 118–128. Burns, R. A., Anstey, K. J., & Windsor, T. D. (2011). Subjective well-being mediates the effects of resilience and mastery on depression and anxiety in a large community sample of young and middle-aged adults. Australian and New Zealand Journal of Psychiatry, 45(3), 240-248. Campbell, A., Converse, P. E., & Rodgers, W. L. (1976). The quality of American life: Perceptions, evaluations, and satisfactions. New York: Russell Sage Foundation. Chang, E., C., Sanna, L. J., Hirsch, J., K., & Jeglic, E. L., (2010). Loneliness and Negative Life Events as Predictors of Hopelessness and Suicidal Behaviors in Hispanics: Evidence for a Diathesis-Stress Model. Journal of Clinical Psychology, 66(12), 1242—1253. Clark, A. E., & Oswald, A. J. (1994). Unhappiness and unemployment. The Economic Journal, 104, 648-659. Cohen, F. , Kemeny, M. E., Zegans L. S. , Johnson P., Kearney K. A., & Stites, D. P. (2007). Immune function declines with unemployment and recovers after stressor termination. Psychosomatic Medicine, 69, 225-234. University of Ghana http://ugspace.ug.edu.gh 91 Constantine, M. G., Alleyne, V. L., Caldwell, L. D., McRae, M. B., & Suzuki, M. B. (2005). Coping responses of Asian, Black, and Latino/Latina New York City residents following the September 11, 2001 terrorist attacks against the United States. Cultural Diversity & Ethnic Minority, 11, 293-308. Córdoba-Doña, J. A., Sebastián, M. S., Escolar-Pujolar, A., Martínez-Faure, J. E. & Gustafsson, P. E. (2014). Economic crisis and suicidal behaviour: the role of unemployment, sex and age in Andalusia, Southern Spain. International Journal for Equity in Health, 13, 55 – 61. Dooley, D, Prause, J., & Ham-Rowbottom, K. A. (2000). Underemployment and depression: longitudinal relationships. Journal of Health and Social Behaviour, 41(4), 421 – 436. Erikson, E.H. (1968). Identify: Youth and crisis. London: Faber and Faber. Ezzy, D. (1993). Unemployment and mental health: A critical review. Social Science and Medicine, 37, 41–52. Falcon, L. M., Todorova, I., & Tucker, K. (2009). Social support, life events, and psychological distress among the Puerto Rican population in the Boston area of the United States. Aging & Mental Health, 13(6), 863–873. Feather, N. T. (1990). The psychological impact of unemployment. New York: Springer. Fergusson, D. M., Boden, J. M. & Horwood, L. J. (2007). Unemployment and suicidal behaviour in a New Zealand birth cohort: A fixed effects regression analysis. CRISIS, 28(2), 95-101. Ferraro, K.F. & Kelley-Moore, J.A. (2000). Religious consolation among men and women: Do health problems spur seeking? Journal for the Scientific Study of Religion, 39(2), 220-235. Fryer, D. & Payne, R. L. (1986). Being unemployed: A review on the literature on the psychological experience of unemployment. In International Review of Industrial and Organizational Psychology, Cooper, C. L. and Robinson, I. (Eds). New York: Wiley. Ghana Statistical Service (GSS; 2012). Population and housing census: Summary report of final results. Accra: Ghana Statistical Service. University of Ghana http://ugspace.ug.edu.gh 92 Gjesfjeld, C.D., Greeno, C. G., Kim, K. H., & Anderson, C.M. (2010) Economic stress, social support, and maternal depression: Is social support deterioration occurring? Social Work Research, 34, 135-143. Gordon, P.A., Feldman, D., Crose, R., Schoen, E., Griffing, G., & Shankar, J. (2002).The role of religious beliefs in coping with chronic illness. Counseling and Values, 46, 162-174. Haden, S. C., Scarpa, A., Jones, R. T., & Ollendick, T. H. (2007). Posttraumatic stress disorder symptoms and injury: The moderating role of perceived social support among veterans with multiple sclerosis. Rehabilitation Psychology, 49, 106–113. Harrison, M.O., Koenig, H.G., Hays, J.C., Eme-Akwari, A.G., & Pargament, K.I. (2001).The epidemiology of religious coping: A review of recent literature. International Review of Psychiatry, 13, 86-93. Helliwell, J. F., & Putnam, R. D. (2004). The social context of well-being. Philosophical Transactions of the Royal Society, 359, 1435-1446. Huber, S. (2007). Are religious beliefs relevant in daily life? In Religion Inside and Outside Traditional Institutions, edited by Heinz Streib. Lieden: Brill Academic Publishers, 211– 230. Huppert, F. A., & Whittington, J. E. (2003). Evidence for the independence of positive and negative well-being: Implications for quality of life assessment. British Journal of Health Psychology, 8(1), 107–122. ILO (International Labour Organization) (2005). “Youth: Pathways to decent work. Promoting youth employment – Tackling the challenge” Geneva. Ismail, Z., & Desmukh, S. (2012). Religiosity and Psychological Well-Being. International Journal of Business and Social Science, 3, 11. ISSER (2007). Registered unemployment in Ghana. Accra: University of Ghana Press. Jahoda, M. (1979). The impact of unemployment in the 1930’s and the 1970’s. Bulletin of the British Psychological Society, 32, 309-314. Jahoda, M. (1981). Work, employment, and unemployment: values, theories and approaches in social research. American Psychologist, 36, 184- 19 1, University of Ghana http://ugspace.ug.edu.gh 93 Jahoda, M. (1982). Employment and unemployment. Cambridge. MA: Cambridge University Press. Joseph, J., & Kuo, B. C. H. (2009). Black Canadians’ coping responses to racial discrimination. Journal of Black Psychology, 35, 78-101. Joshi, S., Kumari, S., & Jain, M. (2008). Religious Belief and Its Relation to Psychological Well-being. Journal of the Indian Academy of Applied Psychology, 34(2), 345-354. Kaul, M., & Lakey, B. (2003). Where is the support in perceived support? The role of generic relationship satisfaction and enacted support in perceived support's relation to low distress. Journal of Social and Clinical Psychology, 22(1), 59-78. Knizek, B. L., Akotia, C. S. & Hjelmeland, H. (2010). A Qualitative study of attitudes toward suicide prevention among psychology in Ghana. OMEGA, 62(2), 169-186 Lakey, B., & Cohen, S. (2000). Social support theory and measurement. In S. Cohen, L. Underwood, & B. Gottlieb (Eds.), Measuring and intervening in social support. New York: Oxford University Press. Laubmeier, K.K., Zakowski, S.G., & Bair, J.P. (2004). The role of spirituality in the psychological adjustment to cancer: A test of the transactional model of stress and coping. International Journal of Behavioral Medicine, 11(1), 48-55. Lin, N., Woefel, M. W., & Light, S. C. (2005). The buffering effect of social support subsequent to an important life event. Journal of Health Social Behavior, 26, 247- 63. Lorenzini, J., & Giugni, M. (2010). Youth Coping with Unemployment: The Role of Social Support. Paper for the YOUNEX Swiss workshop on “Youth, Unemployment, Precariousness, and Exclusion in Switzerland”, Geneva, 15 October 2010. Maselko, J., & Laura D. K., (2006). Gender Differences in Religious Practice, Spiritual Experiences and Health: Results from the General Social Survey. Social Science and Medial 62, 2848-2860. McKee-Ryan, F. M., Song, Z., Wamberg, C. R., & Kinicki, A. J. (2005). Psychological and physical well-being during unemployment: A meta-analytic study. Journal of Applied Psychology, 90, 53-76. University of Ghana http://ugspace.ug.edu.gh 94 Metalsky, G. I., & Joiner, Jr., T .E. (1997). The Hopelessness Depression Symptom Questionnaire Cognitive. Therapy and Research, 21(3), 359-384 Mohr D. C., Boudewyn A. C., Goodkin D. E., Bostrom A., Epstein L. (2001). Comparative outcomes for individual cognitive-behavior therapy, supportive–expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. Journal of Consultation Clinical Psychology, 69, 942–9. Murphy, G. C., & Athanasou, J. A. (1999). The effect of unemployment on mental health. Journal of Occupational and Organizational Psychology, 72, 83-99. Organization for Economic Co-operation and Development (OECD, 2008). Are all jobs good for your health? The impact of work status and working conditions on mental health. OECD Employment Outlook. Chapter 4. Osafo, J., Hjelmeland, H., Akotia, C. S. & Knizek, B.L. (2011). Social injury: An interpretative phenomenological analysis of the attitudes towards suicide of lay persons in Ghana. International Journal Qualitative Study Health Well-being, 6, 8708. Osman, A., Bagge, C. L., Gutierrez, P. M., Konick, L. C., Kopper, B. A., & Barrios, F. X. (2001). The Suicidal Behaviors Questionnaire-Revised (SBQ-R): Validation with clinical and nonclinical samples. Assessment, 8(4), 443-454. Pargament, K.I. (1990). God help me: Toward a theoretical framework of coping for the psychology of religion. Research in the Social Scientific Study of Religion, 2,195-224. Pargament, K.I. (1997). The psychology of religion and coping: Theory, research, and practice. New York: The Guilford Press. Paugam,S.(2004). Laprécarité professionnelle: effets individuels et sociaux. In G. Fournier et B. Bourassa, Entretiendu CRIEVAT Laval. Québec: Pressesdel’ Université Laval Paul, K. I. (2005). The negative mental health effect of unemployment: Meta-analyses of cross- sectional and longitudinal data. Unpublished doctoral dissertation, University of Erlangen- Nürnberg. Paul, K. I., & Moser, K. (2009). Unemployment impairs mental health: Meta-analyses. Journal of Vocational Behavior 74, 264–282 University of Ghana http://ugspace.ug.edu.gh 95 Ramirez, S. P., Macedo, D. S., Sales, P. M., Figueiredo, S. M., Daher, E. F., Araujo S. M., Pargament, K. I., Hyphantis, T. N, & Carvalho, A.F. (2012). The relationship between religious coping, psychological distress and quality of life in hemodialysis patients. Journal of Psychosomatic Research, 72,129-135. Reine, I., Novo, M. & Hammarstrom, A. (2013). Unemployment and ill health – A gender analysis: Results from a 14-year follow-up of the Northern Swedish Cohort. Public Health 127, 214 – 222. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Ruini, C., Belaise, C., Brombin, C., Caffo, E., & Fava, G. A. (2006). Well-being therapy in school settings: A pilot study. Psychotherapy and Psychosomatics, 75(6), 331–336. Ryan, R.M., & Deci, E.L. (2002). Overview of self-determination theory: An organismicdialectical perspective. In: E. L. Deci & R. M. Ryan (Eds.), Handbook of self- determination research (pp. 3–33). Rochester, NY: University of Rochester Press. Schaufeli, W. B. & Van Yperen, N. W. (1992). Unemployment and psychological distress among graduates: A longitudinal study. Journal of Occupational and Organizational Psychology, 65, 291 –305. Schaufeli, W. B. (1997). Youth unemployment and mental health: some Dutch findings. Journal of Adolescence, 20, 281–292. Seligman, M.E.P. (1975). Helplessness: On depression, development, and death. San Francisco: W.H. Freeman. Shafranske, E., & Maloney, H. (1990). Clinical psychologists' religious and spiritual orientations and their practice of psychotherapy. Psychotherapy, 27, 72‐78. Siegel, K., Anderman, S.J. & Schrimshaw, E.W. (2001). Religion and coping with health related stress. Psychology and Health, 16, 631-653. Smith, T. B., McCullough, M. E., & Poll, J. (2003). Religiousness and depression: Evidence for a main effect and the moderating influence of stressful life events Psychological bulletin, 129(4), 614–636. University of Ghana http://ugspace.ug.edu.gh 96 Spinhoven, P., Elzinga, B. M., Hovens, J. G., Roelofs, K., van Oppen, P., & Zitman, F. G. (2011). Positive and negative life events and personality traits in predicting course of depression and anxiety. Acta Psychiatrica Scandinavica, 124(6), 462–473. Strawbridge, William J., Sarah J. Shema, Richard D. Cohen, & George A.Kaplan, (2001). Religious Attendance Increases Survival by Improving and Maintaining Good Health Behaviours, Mental Health and Social Relationships. Society of Behavioural Medicine, 23, 68-74. Stroebe, M.S. (2004). Religion in coping with bereavement: Confidence of convictions or scientific scrutiny? The International Journal for the Psychology of Religion, 14(1), 23-36. Thoits, P. A. (1995). Stress, Coping, and Social Support Processes: Where Are We? What Next? Journal of Health and Social Behavior, 35, 53-79. Thoits, P. A. (2003). Personal Agency in the Accumulation of Multiple Role-Identities. Pp. 179-94 in Advances in Identity Theory and Research, edited by P. J. Burke, T. J. Owens, R. Serpe, and P. A. Thoits. New York: Kluwer Academic/Plenum. Uccelli, M., Mohr, M., Battaglia M. A, Zagami P, Mohr, D.C. (2004). Peer support groups in multiple sclerosis: current effectiveness and future directions. Multiple Sclerosis, 10, 80– 84. Utsey, S. O., Adams, E. P., & Bolden, M. (2000). Development and initial validation of the Africultural Coping Systems Inventory. Journal of Black Psychology, 26(2), 194-215. Utsey, S. O., Brown, C., & Bolden, M. A. (2004). Testing the structural invariance of the Africultural Coping Systems Inventory across three samples of African descent populations. Educational and Psychological Measurement, 64(1), 185-195. Wakefield, J. R. H., Bickley, S. & Sani, F. (2013). The effects of identification with a support group on the mental health of people with multiple sclerosis. Journal of Psychosomatic Research 74, 420–426 WHO (2000). Preventing Suicide a Resource for General Physicians Mental and Behavioural Disorders Department of Mental Health World Health Organization: WHO worldwide initiative for the prevention of suicide. Geneva. WHO/MNH/MBD/00.1. Original: English University of Ghana http://ugspace.ug.edu.gh 97 Willams, R. W., Larson, D. B., Bukler, R. E., Heckman. R. C., & Pyle, C. M. (1991). Religion and psychological distress in a community sample. Social Science Medicine, 32, 1257- 1262. Winefield, A. H. & Fryer, D. (1996). Some emerging threats to the validity of research on unemployment and mental health. Australian Journal of Social Research, 2, 115–128. Winefield, A. H. (1995). Unemployment: Its psychological costs. In International Review of Industrial and Organizational Psychology, Vol 10, Cooper, C. L. and Robertson, I. T. (Eds). London: Wiley, pp. 169–212. Yarquah, J. A. & Baafi-Frimpong, S. (2012). Social Cost of Educated Youth Unemployment in Ghana and Its Implications for Education. Centrepoint Humanities Edition, 14(1), 122- 143. Yarquah, J. A., & Baafi-Frimpong, S. (2011) Social Cost of Educated Youth Unemployment in Ghana and Its Implications For Education. Centrepoint Humanities Edition, 14(1), 122- 143. University of Ghana http://ugspace.ug.edu.gh 98 APPENDICES APPENDIX I: QUESTIONNAIRE This is a study about the psychological effects of youth unemployment in Ghana. The researcher is conducting this study in partial fulfilment for the award of MPhil Degree in Psychology. Despite its academic orientation, the study has some benefits to society. The study will serve as a tool for social action, as it is meant to emphasize the need for prompt interventional measures for the ills associated with youth unemployment in Ghana. This Questionnaire is designed to assist the researcher measure the extent to which unemployment affects individual’s psychological health and suicidal ideation. Respondents are assured of complete anonymity and confidentiality. Under no circumstance will any part of the responses be used against any participant. Participation or completion of the questionnaire is voluntary. Participants can withdraw from the study at any point. Your willingness to participate in this study is very much appreciated. Section A: The Hopelessness Depression Symptom Questionnaire Instructions: On this questionnaire is a set of statements. Please read each statement and TICK or CHECK (√) the response which describes you best for the past TWO WEEKS. BE SURE TO READ EACH STATEMENT BEFORE MAKING YOUR CHOICE. Use the following response scale. Response Scale: 1. Strongly disagree 2. Disagree 3. Neutral 4. Agree 5. Strongly agree Item Strongly Strongly Disagree Neutral Agree 1 2 3 4 5 1. I have stopped trying to get what I want. 2. I am passive when it comes to getting what I want these days. 3. I have given up trying to accomplish what is important to me. 4. My motivation to get things done is as good as usual. 5. I need little or no support from other people. 6. I don't rely on other people to do things for me. University of Ghana http://ugspace.ug.edu.gh 99 Item Strongly Strongly Disagree Neutral Agree 1 2 3 4 5 7. I am not overly dependent on other people. 8. I am a burden to other people 9. I am doing things in "slow motion" these days. 10. I walk around like a zombie these days 11. My speech is slowed down 12. My thoughts are slowed down 13. My energy is lower than usual 14. I can't get things done as well as usual 15. I have as much energy as usual 16. I get tired out more easily than usual 17. I enjoy things as much as usual 18. When doing things I normally enjoy, I have as much fun as usual 19. When it comes to the things in life that count, I am as interested as usual 20. I don’t enjoy sex as much as usual 21. I have trouble falling asleep 22. I have trouble sleeping through the night 23. I wake up early in the morning and have trouble falling back to sleep 24. I can fall asleep as well as usual 25. My concentration is as good as usual 26. I can concentrate as well as usual University of Ghana http://ugspace.ug.edu.gh 100 Items Strongly Strongly Disagree Neutral Agree 1 2 3 4 5 27. I brood about unpleasant events these days 28. I am usually distracted by unpleasant thoughts 29. I have thoughts of killing myself 30. I am considering possible ways of committing suicide 31. I have little or no control over suicidal thoughts 32. I am having impulses to kill myself Section B: The Suicidal Behaviors Questionnaire-Revised (SBQ-R) Please check the number beside the statement or phrase that best applies to you. 1. Have you ever thought about or attempted to kill yourself? (check one only) 1. Never 2. It was just a brief passing thought 3a. I have had a plan at least once to kill myself but did not try to do it 3b. I have had a plan at least once to kill myself and really wanted to die 4a. I have attempted to kill myself, but did not want to die 4b. I have attempted to kill myself, and really hoped to die 2. How often have you thought about killing yourself in the past year? (check only only) 1. Never 2. Rarely (1 time) 3. Sometimes (2 times) 4. Often (3-4 times) 5. Very often (5 or more times) 3. Have you ever told someone that you were going to commit suicide, or that you might do it? (check one only) 1. No 2a. Yes, at one time, but did not really want to die 2b. Yes, at one time, and really wanted to die 3a. Yes, more than once, but did not want to do it 3b. Yes, more than once, and really wanted to do it 4. How likely is it that you will attempt suicide someday? (check one only) 0. Never 1. No chance at all 2. Rather unlikely 3. Unlikely 4. Likely 5. Rather likely 6. Very likely University of Ghana http://ugspace.ug.edu.gh 101 Section C: The Multidimensional Scale of Perceived Social Support (MSPSP) Instructions: The items are divided into factor groups relating to the source of the social support, namely family, friends or significant other. We are interested in how you feel about the following statements. Read each statement carefully. Indicate how you feel about each statement. The rating scale is as follows: 1. Very strongly disagree 2. Strongly disagree 3. Mildly disagree 4. Neutral 5. Mildly agree 6. Strongly agree 7. Very strongly agree Items 1 2 3 4 5 6 7 1. There is a special person who is around when I am in need. 2. There is a special person with whom I can share my joys and sorrows. 3. My family really tries to help me. 4. I get the emotional help and support I need from my family. 5. I have a special person who is a real source of comfort to me. 6. My friends really try to help me. 7. I can count on my friends when things go wrong. 8. I can talk about my problems with my family. 9. I have friends with whom I can share my joys and sorrows. 10. There is a special person in my life who cares about my feelings. 11. My family is willing to help me make decisions. 12. I can talk about my problems with my friends. University of Ghana http://ugspace.ug.edu.gh 102 SECTION D: Automatic Thoughts Questionnaire (A T Q) Instructions: Listed below are varieties of thought that pop into people’s heads. Please read each thought and indicate how frequently, if at all, the thought occurred to you over the last week. Please read each item carefully and WRITE in the appropriate box in following fashion 1. Not at all 2. Sometimes 3. Moderately often 4. Often 5. All the time 1. I feel like I’m up against the world. 2. I’m no good. 3. Why can’t I ever succeed? 4. No one understands me. 5. I’ve let people down. 6. I don’t think I can go on. 7. I wish I were a better person. 8. I’m so weak. 9. My life’s not going the way I want it to go. 10. I’m so disappointed at myself. 11. Nothing feels good anymore. 12. I can’t stand this anymore. 13. I can’t get started. 14. What’s wrong with me? 15. I wish I were somewhere else. 16. I can’t get things together. 17. I hate myself. 18. I’m worthless. 19. I wish I could just disappear. 20. What’s the matter with me? 21. I’m a loser. 22. My life is a mess. 23. I’m a failure. 24. I’ll never make it. 25. I feel so helpless. 26. Something has to change. 27. There must be something wrong with me. 28. My future is bleak. 29. It’s just not worth it. 30. I can’t finish anything. 4 1 2 3 5 University of Ghana http://ugspace.ug.edu.gh 103 Section E: The Centrality of Religiosity Scale (CRS) Instructions: Please read all of the statements in the table below and select the response that best describes you in terms of your religiosity. Be sure to read each of the statements before making your choice. never rarely occasionally often very often 1. How often do you think about religious issues? 2. How often do you take part in religious services? 3. How often do you experience situations in which you have the feeling that God or something divine wants to communicate or to reveal something to you? 4. How often do you pray? 5. How often do you keep yourself informed about religious questions through radio, television, internet, newspapers, or books? 6. How often do you pray spontaneously when inspired by daily situations? 7. How often do you experience situations in which you have the feeling that God or something divine is present? 8. How often do you experience situations in which you have the feeling that God or something divine intervenes in your life? not at all not very much moderately quite a bit very much so 9. How interested are you in learning more about religious topics? 10. To what extent do you believe that God or something divine exists? 11. To what extend do you believe in an afterlife— e.g. immortality of the soul, resurrection of the dead or reincarnation? 12. How important is to take part in religious services? 13. How important is personal prayer for you? 14. In your opinion, how probable is it that a higher power really exists 15. How important is it for you to be connected to a religious community? University of Ghana http://ugspace.ug.edu.gh 104 Section F: Rosenberg Self-Esteem Scale For each of the following statements, please mark or check the response that best reflects yourself in relation to each statement. Items Strongly Disagree Somewhat Disagree Neutral Somewhat Agree Strongly Agree 1. I feel that I am a person of worth. 2. I feel that I have a number of good qualities. 3. All in all, I am inclined to feel that I am a failure. 4. I am able to do things as well as most other people. 5. I feel I do not have much to be proud of. 6. I take a positive attitude toward myself. 7. On the whole, I am satisfied with myself 8. I wish I could have more respect for myself. 9. I certainly feel useless at times. 10. At times I think I am no good at all. Section G: Demographic Information Kindly provide information to the following biographical indicators. 1. Gender: Male Female 2. Age __________________ 3. Marital Status: Single Married Divorced Widowed 4. Religion: Christianity Islam Tradition Others (specify) ……………………………….. 5. Place of residence (ie. village, town, city)____________________________________ 6. Region of residence (e.g., Central region, Ashanti region, etc) _________________________ University of Ghana http://ugspace.ug.edu.gh 105 7. Education Background : a. No formal education b. Basic education c. Secondary education d. Tertiary education (Specify, eg. polytechnic, university) ___________________ 8. Employment status: a. Fully employed b. Somehow employed c. Unemployed 9. If unemployed, indicate the number of years you have been unemployed _______________ 10. If unemployed, indicate your source of livelihood ___________________________________________________________________________ __________________________________________________________________________ 11. If unemployed, have you ever been employed previously? Yes No 12. If previously employed, kindly describe the job and the position you held: i. Job_______________________________________________________________ ii. Position held_______________________________________________________ Thank you very much!!! University of Ghana http://ugspace.ug.edu.gh 106 APPENDIX IV: CONSENT FORM UNIVERSITY OF GHANA OFFICE OF RESEARCH, INNOVATION AND DEVELOPMENT Ethics Committee for Humanities (ECH) PROTOCOL CONSENT FORM Section A- BACKGROUND INFORMATION Title of Study: Psychological effects of youth unemployment in Ghana: A case study of the Greater Accra Region Principal Investigator: Christopher M. Amissah Principal Supervisor: Dr. Maxwell Asumeng Co-Supervisor: Dr. Kingsley Nyarko Address: Department of Psychology, University of Ghana, Legon Certified Protocol Number ECH 049/14-15 Section B– CONSENT TO PARTICIPATE IN RESEARCH Introduction This Consent Form contains information about the research named above. In order to be sure that you are informed about being in this research, we are asking you to read (or have read to you) this Consent Form. You will also be asked to sign it (or make your mark in front of a witness). We will give you a copy of this form. This consent form might contain some words that are unfamiliar to you. Please ask us to explain anything you may not understand. Reason for the Research You are being asked to take part in a research that assesses the psychological effects of youth unemployment. General Information about Research This research will mainly involve the use of primary data in the form of responses provided to standardized measures in the questionnaire. Specifically, the study seeks to assess your Official Use only Protocol number University of Ghana http://ugspace.ug.edu.gh 107 psychological health (depression, self-esteem, hopelessness, and negative thoughts) and suicidal ideation as consequences of long-term unemployment. The study also examine how social support and religiosity serve as buffers to the psychological effects of unemployment among the youths in Ghana. Your Part in the Research If you agree to be in the research, you will be required to respond to a set of questions that inquire about your feelings and psychological state. Possible Risks No physical risk is anticipated in this research. However, your participation may cause you some psychological and emotional discomfort. This is because you will be required to produce information about certain sensitive psychological issues such as depression, anxiety, and suicidal thoughts and behavior. In acknowledgement of this danger, the researcher has instituted remedial measures. Depending on the degree of discomfort you may suffer, the researcher will employ the service of professional psychologists to help in counselling in order to forestall psychological and emotional stability to you. The discussion will help address some personal problems you may bring across, thereby improving you general psychological wellbeing. Moreover, your participation in this research will also deny you some precious time for yourself. The study is likely to take about 30 minutes of your time. Within such period, the researchers will engage you for the purpose of the research. Possible Benefits Your participation will help you gain greater consciousness of your psychological state. The outcome of the study will also be useful in planning interventions for individuals who may have poorer psychological health due to long-term unemployment. Recommendations will also be made to policy makers about the need create job opportunities for the youths in order to maximize their psychological health. Finally, the study will contribute to the body of knowledge in this area of research. If You Decide Not to Be in the Research You are free to decide if you want to be in this research or not. It is purely voluntary. University of Ghana http://ugspace.ug.edu.gh 108 Confidentiality The data from this study is for academic purposes only and will be kept strictly and completely confidential. Your personal information will not be associated with the data nor with any written reports, presentation, or publications that may develop from this study. Any future use of the data will be for the same purposes and will be subjected to the same confidentiality guidelines. Compensation You will not be paid for participating in this research since it is purely voluntary. Leaving the Research You may leave the research at any time. If you choose to take part, you can change your mind at any time and withdraw. If You Have a Problem or Have Other Questions In case you encounter any problem during this research or have any questions about the research, please call Mr. Christopher M. Amissah, the prinicipal investigator on 0242875615, or Dr. Maxwell Asument, the principal supervisor on 0248674405, or Dr. Kingsley Nyarko, the co-supervisor on 0548006675, all in the Psychology Department of University of Ghana, Legon. Your rights as a participant This research has been reviewed and approved by the Ethics Committee of Institute of Statistical, Social and Economic Research of the University of Ghana. The committee is the recognized body of University of Ghana that reviews research studies in order to help protect participants. If you have any questions about your rights as a research participant you may contact the committee via the following email addresses: ech@isser.edu.gh or dopai-tetteh@ug.edu.gh University of Ghana http://ugspace.ug.edu.gh 109 VOLUNTEER AGREEMENT I have read and understood the above document describing the benefits, risks and procedures for the research entitled “Psychological effects of youth unemployment in Ghana.” I have been given an opportunity to have any questions about the research answered to my satisfaction. I agree to participate as a volunteer. ___________________________ ________________________________________ Date Name and Signature of volunteer If volunteers cannot read the form themselves, a witness must sign here: I was present while the benefits, risks and procedures were read to the volunteer. All questions were answered and the volunteer has agreed to take part in the research. _______________________ ______________________________________ Date Name and Signature of Witness I certify that the nature and purpose, the potential benefits, and possible risks associated with participating in this research have been explained to the above individual. _______________________ ____________________________________ Date Name and Signature of Principal Investigator University of Ghana http://ugspace.ug.edu.gh 110 University of Ghana http://ugspace.ug.edu.gh