Understanding flooding events in Ghana: a social marketing and self-determination theory perspective Ernest Yaw Tweneboah-Koduah, Matilda Adams, Michael Nana Amoakoh and Stephen Mahamah Braimah Department of Marketing and Entrepreneurship, University of Ghana Business School, Accra, Ghana Abstract Purpose – The purpose of this study is to deepen the understanding of individuals’ flood disaster mitigation behaviour through the theoretical lens of self-determination theory (SDT). Specifically, the paper examines the influence of autonomy, competence and relatedness on individuals’ self-determined motivation for flood disaster mitigation. The study also examines whether self-determined motivation influences individuals’ flood disaster mitigation behaviour adoption. Design/methodology/approach – Through a cross-sectional survey design, the study used a quantitative research approach for data collection and analyses. Specifically, data was analysed using the structural equation modelling technique. Findings – The result of this study revealed that competence was the strongest predictor of self-determined motivation, followed by autonomy. The result further shows that self-determined motivation positively influences individual’s adoption of flood mitigation behaviour. However, the results showed that relatedness does not predict self-determined motivation for flood disaster mitigation. Practical implications – To enhance the effectiveness of flood disaster mitigation social marketing campaigns, the SDT psychological needs constructs (particularly, competence and autonomy) can aid in assessing the motivation of the target population towards adopting mitigation behaviours. An assessment of motivation will help understand perception of flood risk and behavioural evaluation. The consideration of autonomy and competence on motivation will aid in developing effective campaigns to satisfy the needs of the target population. Originality/value – The paper adds to the limited research that used SDT to understand a social marketing phenomenon. Keywords Social marketing, Self-determined motivation, Autonomy, Competence, Relatedness, Flood disaster mitigation behaviour Paper type Research paper 1. Introduction Over the past decade, flooding has become one of the most threatening natural disasters confronting many parts of the world. Floods can cause widespread devastation, which results in loss of life and damages to personal property and public health infrastructure in both developed and developing worlds (Mensah and Ahadzie, 2020). In 2019, for example, a report from the World Health Organisation (WHO) suggests that between 1998 and 2017, floods affected more than 2 billion people worldwide (WHO, 2019). The 2019 WHO report as well as other scholarly sources indicate that developing countries are particularly vulnerable to flood disasters (WHO, 2019; AlQahtany and Abubakar, 2020; Fatti and Patel, 2013; JSOCM 12,4 534 Received 13March 2021 Revised 22 September 2021 25 January 2022 19 April 2022 Accepted 26April 2022 Journal of Social Marketing Vol. 12 No. 4, 2022 pp. 534-555 © EmeraldPublishingLimited 2042-6763 DOI 10.1108/JSOCM-03-2021-0049 The current issue and full text archive of this journal is available on Emerald Insight at: https://www.emerald.com/insight/2042-6763.htm http://dx.doi.org/10.1108/JSOCM-03-2021-0049 Glago, 2019). In Ghana, research suggests that flood disasters rank the second highest among natural disasters after epidemics (Addo and Adeyemi, 2013; Ansah et al., 2020). This phenomenon has become a major hindrance to sustainable development in the country as governments and individuals have to bear many losses associated with flood disasters. Thus, if the country is to achieve sustainable development, the adoption of flood mitigation behaviours is a necessary condition for the implementation of effective policies. Ghana recorded an unprecedented flood event in June 2015, which led to the death of 150 people (Graphic Online, 2020; Amoako and Inkoom, 2018) and a massive destruction of properties and economic losses (Erman et al., 2018) costing the nation about US$116,340.22. Since this experience, a huge focus has been shifted to providing effective solutions to help mitigate flooding disasters in Ghana. For example, the Government of Ghana, through relevant agencies such as the Ministry of Works and Housing and Ministry of Local Government and Rural Development, have implemented some engineering interventions that aim at dredging and desilting major rivers and streams to increase their capacities (Modern Ghana, 2021), costing the taxpayer US$32m annually. Yet, it appears that mitigating flood disasters is still a daunting task for government and other stakeholders as there has not been much significant improvement in flooding events. A study by Amoako and Inkoom (2018) revealed that heavy rainfall is one of the major causes of urban flooding. However, previous research in Ghana and other countries demonstrate that other anthropogenic activities further exacerbate it. Some of these activities include building on waterways (Nair and Abraham, 2018; Glago, 2021), inadequate drainage systems (Logah et al., 2014), poor urban and structural planning (Owusu-Ansah, 2016), removal of urban vegetation (Owusu-Ansah, 2016), rural–urban migration (Amoako and Frimpong Boamah, 2015) and poor attitude of Ghanaians towards proper waste management practices (Tabiri, 2015; Alhassan et al., 2018). For example, Alhassan et al. (2018) explain that the poor attitude of most Ghanaians with regard to the dumping of solid waste into nearby gutters and streams results in drainage blockages that prevent the free flow of flood water, thereby increasing flood disasters during the rainy season. The problem is thus ultimately one of a behavioural issue and thus requires a behavioural change approach towards its mitigation (Valois et al., 2020). However, to date, little research (if any) has been conducted to understand the behavioural factors that influence individuals’ adoption of flood mitigation behaviours. In this paper “mitigation behaviours” refer to any activity that individuals perform before a flood to reduce or eliminate the impact of or loss caused by it”. Previous studies have focused primarily on the nature and extent of floods (Gyekye, 2013; Okyere et al., 2013), and causes of flood and mitigation measures (Asumadu- Sarkodie et al., 2015), institutional and social dimension (Korah and Cobbinah, 2019) and flood risk management strategies (Ahadzie and Proverbs, 2011) to the neglect of understanding the behavioural factors underlying mitigation behaviours. Consequently, understanding the behavioural and psychological mechanisms that influence flood mitigation activities among individuals could lead to a better and a more effective flood disaster mitigation programme in Ghana. Social marketing, thus, presents itself as a potential framework that could be used in this research enquiry as it seeks to understand and influence behaviour change of the target audience for social good (Spotswood et al., 2012; Gruneklee et al., 2016). Previous studies have applied the concept and seen enormous successes in diverse contexts, including recycling; food waste; transport; energy conservation; environmental protection; and proper waste management practices (Haghighatjoo et al., 2020; Carins and Rundle-Thiele, 2014; Haj-Salem and Al-Hawari, 2021; Tweneboah-Koduah et al., 2019; Abotchie and Shokar, 2009; Hodgkins et al., 2019; Doku, 2012), among others, in both the developed and developing worlds. In spite of social Understanding flooding events in Ghana 535 marketing’s successful application in these aforementioned studies, it appears its application to the adoption of flood disaster mitigation behaviours is relatively silent in the literature and this study seeks to address that. According to Ayikwa et al. (2020), social marketers resort to theoretical frameworks to help understand and explain the motivation or influential factors underlying people’s behaviour change adoption. Although several behavioural theories exist, the social marketing literature is replete with the repeated use of traditional behavioural theories such as the theory of planned behaviour (Ajzen, 1991), stages of change model (Prochaska et al., 2013) and the health belief model (Julinawati et al., 2013), to the neglect of other relevant behavioural theories [such as the self-determination theory (SDT)]. Most of the aforementioned traditional behavioural theories describe motivation through extrinsic means. However, research suggests that because human behaviour is complex, it cannot always be well explained by extrinsic motives (Gilal et al., 2019). For instance, an individual may be more willing to engage in behaviour because of his or her own inherent interest (self-determined motivation) in the behaviour and not because of any extrinsic motive. To that extent, some scholars (Dibb, 2014; Luca and Zervas, 2016) in social marketing have called for additional theories to better understand the complex nature of pro-social issues, which in the long run will offer a huge promise to intervention effectiveness. In an attempt to address the foregoing research gaps, the current study uses the SDT (Deci and Ryan, 2000), which is a human motivation theory to understand and predict a social marketing phenomenon (i.e. the adoption of flood disaster mitigation behaviours) in Ghana. Thus, the objectives of this study are twofold. First, the study seeks to examine the three psychological needs of SDT (i.e. autonomy, competence and relatedness) as predictors of self-determined motivation towards flood disaster mitigation behaviours. In this study, we operationalise flood disaster mitigation behaviours as a range of environmentally friendly actions, such as “proper waste disposal practices”, “refraining from building on waterways and flood plain areas”, “planting of trees”, etc. Second, the study examines whether self-determined motivation influences flood disaster mitigation behaviour adoption. The paper makes significant contributions to both theory and practice. First, the study adds to the social marketing literature by drawing on the SDT to help gain a better understanding into why people adopt flood mitigation behaviours. As this study, to the best of the authors’ knowledge, is among the few studies that have attempted to fully test the SDT theory of predictors in a social marketing context, it makes significant contributions to theory by providing essential insights that extend our understanding of flood mitigation behaviour adoption from a developing country perspective. Second, the study contributes to practice, as the findings will provide important information to implementers of social marketing programs, policymakers and social marketers aiming to promote flood mitigation behaviours among Ghanaians, to better understand the strategies and communication messages they need to adopt in promoting mitigation behaviours. The remainder of this article is structured in the following order: Section 2 focusses on presenting the theoretical arguments of the appropriateness of the SDT in predicting behaviour, which subsequently leads to the development of hypotheses on the relationships within the model. Section 3 presents the research methodology, which is followed by analysis and results in Section 4. The discussions, practical and theoretical implications as well as the limitations and future research directions are also presented in Section 5. 2. Theoretical context and hypotheses development As indicated earlier, the current study draws on Ryan and Deci’s (2017) SDT as it is a leading theory of human motivation (Gilal et al., 2019). SDT suggests that there are three JSOCM 12,4 536 basic psychological needs, autonomy (have choice and control), competence (feeling competent and capable) and relatedness (feeling affiliated with others in the context) which shape self-determined motivation (Ryan and Deci, 2017; Slemp et al., 2018; Van den Broeck et al., 2016). The theory also posits that the extent to which these three needs are supported and satisfied largely dictates the level of self-determined motivation and that when these needs are met, participants experience more self-determined motivation and more positive outcomes (Sweeney et al., 2014). SDT has been extensively applied to explain and predict human motivation in a variety of contexts, including health, education, recycling, e-waste, pro-environmental behaviours and in public policy to encourage better consumer decisions (Hiratsuka et al., 2017; Pentecost et al., 2017; Evans, 2015; Sweeney et al., 2014; Webb et al., 2013; Chen, 2014; Gilal et al., 2019; Baxter and Pelletier, 2020). The theory has also been applied in commercial marketing in the areas of customer loyalty (Lin et al., 2009) and relational marketing (Dholakia, 2006). Although SDT has been acknowledged in social marketing, it has not been frequently used by social marketers (Binney et al., 2006; Mitchell et al., 2017). To this end, this study explores the potential of SDT to address a social marketing phenomenon specifically, flood mitigation behaviour, which is currently limited in the literature. We, therefore, argue in this current study that the basic psychological needs of autonomy, competence and relatedness are predictors of self-determined motivation. We also argue that self-determined motivation influences individuals’ flood disaster mitigation behaviours. The relationships within the model are further discussed in the sub-sections using existing literature. 2.1 Hypotheses development 2.1.1 Autonomy and self-determined motivation to engage in behaviour. In the view of Gilal et al. (2019), autonomy refers to the need to feel self-directed and to act volitionally with a sense of choice. Put differently, autonomy need is satisfied when individuals feel they have the power to control their own actions and behaviour (Deci and Ryan, 2000). For example, if individuals perceive pressure from other people such as city officials to engage in flood disaster mitigation behaviours, this would be a controlled situation that would result in perceived low autonomy, whereas a lack of pressure from outside sources would result in high perceived autonomy arising from a sense of personal choice, which in turn leads to self-determined motivation. For instance, in examining the relationship between autonomy and self-determined motivation, most findings established that autonomy was a good predictor of self-determined motivation (Deci and Ryan, 2000; Lavergne et al., 2010; Osbaldiston and Sheldon, 2003; Standage et al., 2003). For instance, in the physical activity setting, Standage et al. (2003), found autonomy to positively influence students’ self-determined motivation towards physical activity. A study by Cooke et al. (2016) in the education setting also established that the use of autonomy supportive language in relation to recycling led to increased self-determined motivation which was related to better results on a quiz about recycling. Similarly, Lavergne et al.’s (2010) study in a pro- environmental context also found that autonomy predicted self-determined motivation for the pro-environmental behaviour. Following from the discussion above, we expect that in this research context, individuals’ perceived autonomy in flood disaster mitigation will positively influence their self-determined motivation for flood disaster mitigation behaviour. The study, therefore, hypothesises that: H1. Individuals’ autonomy in flood disaster mitigation will have a positive significant influence on their self-determinedmotivation for flood disaster mitigation behaviour. 2.1.2 Competence and self-determined motivation to engage in behaviour. Within the SDT, competence denotes the need to feel “effective and confident with respect to some behaviour Understanding flooding events in Ghana 537 or goal” (Deci and Ryan, 2014, p. 55) and achieving desired outcomes (Engström and Elg, 2015). The need for competence arises when individuals perceive that they are capable or have the ability to perform a determined task and it is related to seeking challenges that are optimal to one’s abilities (Ryan and Deci, 2017). Chen et al. (2015) suggest that feelings of competence are necessary for individuals to approach optimal challenges that allow them to learn and develop. For example, in this research context, perceived competence may be the feeling that one has the skills, or access to the skills and information about flood disaster mitigation behaviour. The experience of low competence would be not having these skills, and not knowing where to find information or help regarding flood disaster mitigation. Some scholars intimate that competence is similar to the concept of perceived self-efficacy, which refers to beliefs about whether a person can achieve certain outcomes or not (Bandura, 1997). Considerable research on SDT shows that when people’s needs for competence are met, they are more likely to have self-determined motivation (Niemec and Ryan, 2009; Ryan and Deci, 2017; Gagne, 2003). For example, a study by Niemec and Ryan (2009) in an educational practice setting revealed that competence facilitates students’ self-determined motivation towards academic performance. Similarly, in a meta-analysis of 184 SDT-based studies in the health domain, Ng et al. (2012) found support for the relationship between competence and self-determined motivation. On the contrary, Cooke et al. (2016) did not find competence to influence self-determined motivation. In spite of Cooke et al. (2016) indicating that competence is not a predictor of self-determined motivation, Engström and Elg (2015) found that competence is the strongest predictor of self-determined motivation. Consistent with Engström and Elg (2015) and SDT, we expect that in this research context, individuals’ competence in flood disaster mitigation will more likely influence their self-determined motivation towards flood mitigation behaviour. We, therefore, formulate the following hypotheses: H2. Individuals’ competence in flood disaster mitigation will have a positive significant influence on their self-determined motivation for flood disaster mitigation behaviour. 2.1.3 Relatedness need and self-determined motivation to engage in behaviour. Relatedness represents the need to “feel personally accepted by significant others, and to feel cared for by others and caring for them” (Deci and Ryan, 2014, p. 52). In flood mitigation context, perceptions of relatedness may result from interactions from a variety of individuals such as family, friends, neighbours, community members and others involved in flood disaster mitigation behaviours. Huang et al. (2019) highlight that in relatedness, people perceive themselves to be socially accepted and connected to a social group they regularly interact with such as family, peers, colleagues and significant others. Because people usually want to feel connected to their social groups, they start taking in and internalising the values of such groups and accept them as theirs, which subsequently influences their decision-making (Ryan and Deci, 2017). An earlier study by Darner (2007) found that there is a relationship between relatedness and self-determined motivation. This finding is corroborated by Pacewicz et al. (2020) and Qian et al. (2020). For example, Qian et al. (2020) in their study of fans’ psychological processes of exports consumption found relatedness to be a stronger predictor of self-determined motivation for exports consumption. In line with the foregoing, we expect that individuals’ perceived relatedness in flood disaster mitigation will positively influence their self-determined motivation for flood disaster mitigation behaviour. The following hypothesis is, therefore, formulated: JSOCM 12,4 538 H3. Individuals’ relatedness in flood disaster mitigation will have a positive significant influence on their self-determined motivation for flood disaster mitigation behaviour. 2.1.4 Self-determined motivation and flood disaster mitigation behaviour. From the SDT perspective, many studies have found self-determined motivation to be a good predictor of behaviour (Weinstein and Ryan, 2010; Hagger and Chatzisarantis, 2009; Kaiser et al., 2020; Ryan and Deci, 2017). For instance, it has been found that people with higher self- determined motivation were more likely to perform pro-environmental behaviours that were perceived to be challenging or difficult (Pelletier et al., 2011; Sheldon et al., 2016; Venhoeven et al., 2013; Webb et al., 2013). Consistent with previous SDT studies, we expect that individuals with higher self-determined motivation towards flood disaster mitigation will more likely participate in flood disaster mitigation behaviours. Thus, we hypothesise the following: H4. Higher self-determined motivation towards flood disaster mitigation has a positive significant influence on flood disaster mitigation behaviour. Our proposed research model in Figure 1 theorises that the basic psychological needs of autonomy, competence and relatedness relating to flood disaster mitigation have a positive influence on individuals’ self-determined motivation for the behaviour. The model also argues that self-determined motivation will positively influence floodmitigation behaviour. 3. Methodology 3.1 Sampling and data collection The study population comprised people who reside in Accra and Kumasi (i.e. Greater Accra and Greater Kumasi Metropolis respectively). The selection of these two cities for the survey was based on the fact that evidence suggests they are themost vulnerable cities in Ghana, as they record the most flood events every year (Mensah and Ahadzie, 2020). Because of the COVID-19 lockdown in Ghana at the time, we purposely sampled the respondents using an online survey. This approach was adopted because according to Fowler (2014), an online survey is suitable when the study’s population is geographically dispersed. The use of online survey has become increasingly popular because it offers a comparatively more cost- effective method for reaching large groups of potential respondents in a short turnaround time (Sanders et al., 2007; Couper, 2000). Its popularity, however, does not mean that the method is inherently valid and reliable as according to De Gregorio and Sung (2010), there are some problems associated with this sampling method (e.g. representativeness issues). Irrespective of this limitation, the current study used this method because it was the best option considering the “abnormal times we currently find ourselves in”. Participants were Figure 1. Research model H3 H2 H1 H4 Understanding flooding events in Ghana 539 invited to complete the online survey via a link that was shared on Facebook and on WhatsApp groups. As indicated earlier, only sample cases residing in Accra and Kumasi and aged between 18 and above were eligible to participate in the study and were used for analysis. Thus, as an inclusion criterion, the initial part of the questionnaire screened the respondents’ eligibility including age and location of respondents. The Facebook accounts and WhatsApp groups were purposely selected to ensure that the respondents reside in Greater Accra and Kumasi Metropolis. No incentive was offered for participation and participants were assured of the anonymity of their responses. To ensure that no duplicate was used for analysis, the questionnaires returned were screened and only those that were not considered as duplicate were used for analysis. The survey was conducted between March andApril, 2020. In all, 294 usable responses were obtained for the analysis. 3.2 Measures The online survey consisted of two sections, Sections A and B. Section A involved the background information of the respondents, while Section B looked at measures regarding need autonomy, competence, relatedness, motivation and behaviour. The need autonomy, competence and relatedness items were all adapted from Ng et al. (2012). Participants rated their agreement on a seven-point Likert Scale from 1 – strongly disagree to 7 – strongly agree. The motivation towards the environment scale by Pelletier et al. (1998) was adapted to measure participants’ motivation for engaging in flood mitigation behaviours. Respondents read the question “why are you engaging in flood mitigation behaviour?” and were asked to indicate the extent to which each item corresponds to their personal motives for engaging in the behaviour on a seven-point scale ranging from 1 – does not correspond at all to 7 – corresponds completely. Four motivation items were included (e.g. “because I like the feelings I have when I am doing things that will mitigate flood disaster”. Finally, behaviour was adapted from Ghani et al. (2013) and was measured with four items. Prior to the online survey administration, the questionnaire was sent electronically to five experts in the area of social marketing, to ensure content validity. Additionally, the questionnaire for the online survey was pilot tested with 18 people who bore similar characteristics as the participants of the study. Based on the comments received, minor changes were made to the wording and positioning of the questions, and finally shared on the social media platforms. To address issues regarding ethics, respondents were asked to read the informed consent form. After reading the form, those who agreed to participate in the study completed the online survey and all responses to this survey were anonymous. Non-response bias (Armstrong and Overton, 1977) was assessed, comparing differences between early and late respondents. The results indicated that no significant differences exist between the two groups. Hence, non-response bias was considered not an issue in the current study. After three weeks of data collection, 322 responses were retrieved from the electronic version of the questionnaire. After data cleaning, we removed 28 ineligible responses and incomplete questionnaires. The study therefore used 294 valid and fully completed responses for further analysis. All statistical data were analysed using SPSS 23 and AMOS 23. The final sample comprised 149 males (51%) and 145 females (49%). Approximately 33% of the respondents were between the ages of 31 and 40years. The majority of the sample had degree- level education (69%). This finding means that the majority of our sample were literate and so they could have cognitive capability to comprehend the statements (see Table 1). 3.3 Common method bias Because our data were collected based on self-reports, we examined the potential effects of common method bias (CMB). We conducted a Harman single-factor test (Podsakoff et al., 2003), JSOCM 12,4 540 which is one of the most common techniques used in addressing common method variance. Confirmatory factor analysis was used to verify whether significant method variance was present. Our findings indicated that our single-factor model exhibited a poor fit (root mean square error of approximation (RMSEA) = 0.18; comparative fit index (CFI) = 0.61; normed fit index (NFI) = 0.052). Additionally, no single factor accounted for more than 30% of the variance, which is less than Podsakoff and Organ’s (1986) 50% threshold, suggesting that the data for the study did not suffer from CMB. 4. Data analyses and results 4.1 Profile of the respondents Table 2 provides the description of the measures used in the study with their mean, standard deviation (SD) and standard error (SE) values. From the table, it is observed that the statement “currently, I have found alternative ways to dispose my waste to mitigate flood disaster” recorded the highest mean value of 4.25 (SD = 0.83). This means that most of the study’s participants agreed with this statement. On other hand, most of the study’s participants disagreed with the statement “because helping to mitigate flood disaster is a sensible thing to do” (mean = 2.79, SD = 1.08). In this study, the authors used structural equation modelling (SEM) to verify the proposed theoretical model and research hypotheses using the two-step approach recommended in SEM (Anderson and Gerbing, 1988). Thus, in Stage 1, the study evaluated the measurement model by running confirmatory factor analysis (CFA) to validate the measurement scales of all five constructs (autonomy, competence, relatedness, self- determined motivation and flood mitigation behaviour) in the research model. Second, the Table 1. Demographic characteristics of respondents Characteristics Frequency (%) Gender Male 149 50.7 Female 145 49.3 Total 294 100 Age 21–30 90 30.6 31–40 98 33.3 41–50 79 26.9 51–60 24 8.2 Above 60 3 1.0 Total 294 100 Education Junior high school 2 7 Senior high school 4 1.4 Diploma 14 4.8 Degree 203 69.0 Others 71 24.1 Total 294 100 Marital status Single 130 44.2 Married 160 54.4 Separated 3 1.0 Divorced 1 0.3 Total 294 100 Understanding flooding events in Ghana 541 structural equation model was estimated to test the hypotheses proposed in the research model. 4.2 Confirmatory factor analyses and structural equation modelling The confirmatory factor analysis was carried out using the Analysis of Moments of Structures (AMOS) software package (Version 22) as well as the maximum likelihood estimation procedure. The first measurement model tested specified five latent variables which were allowed to correlate freely with each other. The fit indices for the initial measurement model suggested a considerable misfit with a Chi square value of 261.24; df = 83; and an RMSEA value of 0.097; hence, the model needed to be re-specified (Bagozzi and Yi, 2012). Accordingly, Table 2. Descriptive statistics of constructs measures Scale item Variable code Mean SD SE Autonomy I feel I am pursing goals that are my own in flood disaster mitigation Auto1 3.9014 1.11787 0.06520 I feel I participate in flood disaster mitigation behaviour willingly Auto2 3.8605 1.04389 0.06088 I choose to participate in flood disaster mitigation behaviour according to my own free Auto3 3.7245 1.13685 0.06630 Competence I feel I am good at performing flood disaster mitigation behaviour Comp1 3.8401 0.96965 0.05655 I feel I can overcome challenges relating to flood disaster mitigation behaviours Comp2 4.0612 0.92167 0.05375 I feel I have the ability to perform well in flood disaster mitigation behaviour Comp3 3.2925 1.02001 0.05949 Relatedness I feel I show concern for others when I perform flood disaster mitigation behaviour Relate1 3.4014 1.11863 0.06524 In performing flood disaster mitigation behaviour, I feel I have a close relationship with other people Relate2 3.4796 1.06026 0.06184 I feel cared for when I perform flood disaster mitigation behaviour Relate3 3.6190 1.03741 0.06050 Behaviour I have joined the campaign to educate others on flood disaster mitigation issues in my community Beh1 3.1259 1.03578 0.06041 I am currently not interested in activities that could cause flood disaster Beh2 2.8503 1.44456 0.08425 I currently do not dispose waste into drainage channels because of flood disasters Beh3 4.2075 1.09036 0.06359 Currently, I have found alternative ways to dispose my waste to mitigate flood disaster Beh4 4.2517 0.82903 0.04835 Self-determined motivation I partake in flood disaster mitigation behaviour because it is fun SDM1 2.8503 0.99042 0.05776 I partake in flood disaster migration behaviour because it is a reasonable thing to do SDM2 2.8265 1.05829 0.06172 I partake in flood disaster mitigation behaviour because I feel guilty if I do not engage in it SDM3 2.7993 1.07589 0.06275 I partake in flood disaster mitigation because I feel is a way I have chosen to contribute to flood disaster mitigation SDM4 2.9524 1.14381 0.06671 JSOCM 12,4 542 the initial measurement model was subjected to modification according to the sizes of the factor loadings. Based on the suggestion of the AMOS output, one item (SDM 4) was removed because it recorded a factor loading less than 0.50 (Hair et al., 2013) and this significantly improved the model fit (x 2 [239.874]; df = 109; x 2/df = 2.201; NFI = 0.934; Tucker-Lewis index (TLI) = 0.953 and CFI = 0.963, RMSEA = 0.064; refer to Table 3). The fit indices for the improvedmodel suggest that the research model fitted the data satisfactorily. 4.2.1 Reliability and construct validity. The study further assessed the measurement reliability and validity of the constructs in the researchmodel. Reliability of the itemswere assessed using Cronbach’s alpha and composite reliability. Both Cronbach’s alpha values and composite reliability scores exceeded the 0.70 thresholds recommended by Krabbe (2016). To establish convergent validity for each construct, the study used factor loadings and average variance extracted (AVE) to show that measures actually measured their respective or intended constructs. The results indicated that all the factor loadings as well as the AVE values were above the acceptable levels of 0.50 (Hair et al., 2006; Awwad and Agti, 2011). The t-values of all construct itemswere greater than 1.96, suggesting significance at p< 0.001, thus, indicating good convergent validity for all the constructs aswell as high reliabilities for all the items (seeTable 4 andFigure 2). 4.2.2 Discriminant validity. To test the discriminant validity of the construct measurements, the study adopted Fornell and Lacker’s (1981), criterion. The results (as Table 3. Fit indices for the measurement model Fit indices Criteria Initial model Modified model Literature support x 2/df #3 2.35 2.20 Schreber et al. (2006) NFI �0.90 0.93 0.93 Hu and Bentler (1999) TLI �0.90 0.82 0.95 Hu and Bentler (1999) CFI �0.90 0.86 0.96 Hooper et al. (2008) RMSEA #0.08 0.07 0.06 Hu and Bentler (1999) Table 4. CFA results for final measurement model Construct Item code Standardized factor loading t-Value CR AVE CA (a) Self-determined motivation SDM1 0.76*** Fixed 0.85 0.59 0.85 SDM2 0.84*** 14.13 SDM3 0.77*** 13.02 SDM4 0.71*** 11.84 Flood reduction behaviour Beh1 0.89*** Fixed 0.93 0.78 0.93 Beh2 0.85*** 20.15 Beh3 0.89*** 22.53 Beh4 0.89*** 22.55 Autonomy Auto1 0.75*** Fixed 0.88 0.71 0.87 Auto2 0.87*** 15.06 Auto3 0.89*** 15.07 Relatedness Relate1 0.74*** Fixed 0.80 0.57 0.80 Relate2 0.82*** 12.46 Relate3 0.70*** 11.03 Competence Comp1 0.86*** Fixed 0.91 0.77 0.91 Comp2 0.87*** 19.04 Comp3 0.90*** 20.11 Notes: CR = composite reliability; CA = Cronbach’s alpha; x 2[239.874]; df = 109; x 2/df = 2.201; NFI = 0.934; TLI = 0.953, CFI = 0.963, RMSEA = 0.064; ***p< 0.001 Understanding flooding events in Ghana 543 Figure 2. Final CFA measurement model JSOCM 12,4 544 shown in Table 5) demonstrate that the square root of the AVE values for the constructs are greater than the inter-construct correlation coefficients, indicating that the constructs are distinct from each other, hence, providing evidence of good discriminant validity (Hair et al., 2013). 4.2.3 Test for hypothesised structural paths using structural equation modelling. After the measurement model had been proven to be reliable and valid, the structural model was estimated. The results for the structural model (see Table 6) show that the fit indices for the hypothesised model fit the data satisfactorily (x 2/df = 2.84; GFI = 0.97; NFI = 0.95; TLI = 0.94; CFI = 0.97; RMSEA= 0.079). From Table 6, it is revealed that H1 (the path from autonomy to self-determined motivation), H2 (the path from competence to self-determined motivation) as well asH4 (the path between self-determined motivation and behaviour) were all supported in this study as they recorded significant coefficients (b = 0.23, t-values = 4.72, p< 0.001); (b = 0.33, t-values = 5.26; p < 0.001) and (b = 0.50, t-values = 9.89; p < 0.001), respectively. However, H3, which hypothesised a positive significant relationship between need relatedness and self-determined motivation, was not supported in this study, as the results showed a non- significant coefficient value (b = 0.12, t-values = 1.23; p > 0.05). Because the hypothesised path between relatedness and motivation (H3) was statistically unsupported, although not hypothesised, we tested a series of alternative models and compared it to the baseline model to exclude the possibility of alternative explanations. We estimated a baseline model as the full mediation model (see Figure 1), which did not have direct paths from the three psychological need dimensions to flood mitigation behaviour. 4.2.4 Tests for model comparison. Following Anderson and Gerbing’s (1988) approach, the current study tested a series of nested models against our baseline model via sequential Chi-square tests with the parameter constraints of interest in the study. According to Lu et al. (2010), a significant change observed in the Chi-square difference would indicate the importance of the constrained path, and therefore provide support for the baseline model. Table 5. Discriminant validity (correlation matrix) Constructs Mean SD AVE 1 2 3 4 5 Self-determined motivation (1) 3.76 0.67 0.59 0.77 Flood reduction behaviour (2) 3.92 0.96 0.78 0.43 0.88 Autonomy (3) 4.19 0.73 0.71 0.34 0.68 0.84 Relatedness (4) 3.39 0.79 0.57 0.63 0.59 0.53 0.76 Competence (5) 3.43 0.78 0.77 0.75 0.56 0.49 0.70 0.88 Notes: SD = standard deviation; AVE = average variance extracted; square root of the AVEs are on the diagonal, and the inter-construct correlations are off-diagonal Table 6. Analysis of hypothesized structural paths Hypothesized relationships b -estimates t-value p-value Results H1: Autonomy!Motivation 0.23 4.72 *** Supported H2: Competence!Motivation 0.33 5.26 *** Supported H3: Relatedness!Motivation 0.12 1.23 0.52 Not Supported H4: Motivation! Behaviour 0.50 9.89 *** Supported Notes: ***p < 0.001; x 2/df = 2.84; GFI = 0.97; NFI = 0.95; CFI = 0.97; RMSEA = 0.079; R2 (motivation) = 0.42; R2 (BEH) = 0.25 Understanding flooding events in Ghana 545 Consequently, in Model 1, the path linking autonomy, self-determined motivation and behaviour was constrained to zero. Likewise, the paths of the hypothesised relationships were constrained to zero in Models 2 and 3, one at a time, after which a partial mediation model was tested in Model 4, where direct paths from the three psychological need factors to the outcome variable were added to the baseline model. The results of the model comparison test are shown in Table 7. The results of the model comparison test show the Chi-square differences between the baseline model and each of the nested models were significant, indicating that the baseline model fits our model satisfactorily. It can also be observed from the results that the Chi- square difference between Model 4 (partial mediation model) and our baseline model (full mediation model) was significant (Dx 2= 26.66, Ddf= 7, p < 0.001). This, therefore, suggests that adding direct paths from the predictors (autonomy, competence and relatedness) to the outcome variable to the baseline model significantly improved the model fit from x 2/df = 2.84, GFI = 0.97, NFI = 0.95, CFI = 0.97, RMSEA = 0.079 to x 2/df = 1.14, GFI = 0.99, NFI = 0.99, CFI = 0.99, RMSEA = 0.022. Hence, this study concludes that the partial mediation model was superior to the full mediation model; therefore, it is acceptable for theoretical consideration. Figure 3 represents the final structural model for this study. 5. Discussion and conclusion The paper sought to explain individuals’ motivation to participate in flood disaster mitigation behaviours through the theoretical lens of SDT. Specifically, the study sought to examine whether the three psychological needs (autonomy, competence and relatedness) predict self-determined motivation for flood disaster mitigation, and also whether motivation acts as a predictor of flood disaster mitigation behaviour adoption. The findings of the study revealed that the relationship between need autonomy and self-determined motivation (H1) was positive and significant. This finding is consistent with previous studies (Deci and Ryan, 2000; Lavergne et al., 2010; Cooke et al., 2016; Gilal et al., 2019; Trépanier et al., 2015; Ryan and Deci, 2017). For instance, Lavergne et al. (2010) found that need autonomy for pro-environmental behaviour was related to self-determined motivation for environmentally active people. The findings of this current study suggest that individuals’ self-determinedmotivations for flood mitigation are largely driven by their need autonomy for flood mitigation. This implies that individuals with psychological need for autonomy for flood disaster mitigation will develop self-determined motivation for behaviour adoption when they feel that their choices regarding flood disaster mitigation are Table 7. Result of the Chi- square difference model comparison test Model x 2 df Dx 2 Ddf x 2/df GFI NFI CFI RMSEA Baseline (full mediation) 31.23 11 – – 2.84 0.97 0.95 0.97 0.079 Model 1 11.51 8 Dx 2 (b, M1) = 19.72*** 3 1.44 0.99 0.97 0.99 0.039 Model 2 26.06 8 Dx 2 (b, M2) = 5.17*** 3 3.26 0.97 0.93 0.95 0.088 Model 3 23.04 8 Dx 2 (b, M3) = 8.19*** 3 2.87 0.97 0.95 0.96 0.080 Model 4 (partial mediation) 4.57 4 Dx 2 (b, M4) = 26.66*** 7 1.14 0.99 0.99 0.99 0.022 Notes: Baseline model: full mediation (paths between autonomy, competence, relatedness and self- determined motivation; and between self-determined motivation and behaviour). Model 1: (autonomy ! self-determined motivation ! behaviour) was constrained to 0; Model 2: (competence ! self-determined motivation ! behaviour) was restricted to 0; Model 3: (relatedness ! self-determined motivation ! behaviour) was constrained to 0; Model 4: (directs paths from autonomy, competence and relatedness to behaviour were included in the model, resulting in a partial mediation model); ***p< 0.001 JSOCM 12,4 546 based on their true interests and values and not based on coercion (Sheldon et al., 2001; Guay et al., 2010). Similar to H1, our results in H2 also indicate a positive significant relationship between competence for flood disaster mitigation and self-determined motivation. This finding corroborates the results of prior research (Deci and Ryan, 2000; Engström and Elg, 2015; Niemiec and Ryan, 2009). A study by Niemiec and Ryan (2009) in an educational practice setting revealed that competence facilitates students’ self- determined motivation towards academic performance. Similarly, the results of this current research suggest that individuals’ self-determined motivation for flood reduction increases when they feel more capable and believe they have the ability to successfully engage in the behaviour. Therefore, communications that reinforce the target group’s capability and ability for the behaviour will enhance their motivation for flood disaster mitigation. Contrary to our expectations, however, the results for H3 revealed a non-significant relationship between relatedness and self-determined motivation. This finding contradicts existing studies (Qian et al., 2020; Pacewicz et al., 2020; Ryan and Deci, 2017; Lavergne et al., 2010; Standage et al., 2003). For instance, Qian et al. (2020) in their study on fans’ psychological processes of exports consumption revealed relatedness as the most salient predictor of exports consumption. The SDT also posits that because people usually want to feel related and connected to their social groups, they start taking in and internalising the values and practices of such groups and accept them as theirs to meet social the group’s approval (Deci and Ryan, 2000). However, in the current study, our findings suggest that in Figure 3. Final structural model R Notes: Parameters are standardized estimates; ***p < 0.0001; X2/df=1.14, GFI = 0.99, NFI = 0.99, CFI = 0.99, RMSEA = 0.022 Understanding flooding events in Ghana 547 the Ghanaian context, this did not seem to be the case as the need for relatedness regarding flood disaster mitigation does not actually increase participants’ self-determined motivation for flood disaster mitigation. The plausible reason for this difference is that, it is possible that in Ghana participants’ need relatedness for flood disaster mitigation is relevant as they make their decisions regarding flood mitigation based on their own values and self-interest rather than what their social groups prefer they do (Chau and Hu, 2002; Flannery, 2017). With respect toH4, we found that self-determined motivation has a positive influence on flood disaster mitigation behaviour. This finding suggests that individuals who perceive that engaging in flood disaster mitigation activity is a way of contributing to flood disaster mitigation will more likely engage in the behaviour. This finding is similar to previous studies (Weinstein and Ryan, 2010; Hagger and Chatzisarantis, 2009; Ryan and Deci, 2017) that have averred that people with stronger self-determined motivation performed pro- environmental behaviours that were perceived to be challenging because they like the feelings they have when they perform the behaviour. Finally, this study tested a series of nested models and compared it to the baseline model in Figure 1 and, we concluded that the partial mediation model (Figure 3) was significantly better than our full mediation model in Figure 1. This is an important finding because it emphasises the need to further our understanding of the combined role of autonomy, competence and relatedness in motivational processes regarding behaviour change. Hence, future research in this area should fully analyse the mediator relations to explore this relationship in other activity contexts. 5.1 Theoretical implications The current paper makes twomain contributions to the growing body of literature. First, the study adds to the social marketing literature by extending the application of social marketing to the context of flood disaster mitigation, a previously under-researched area. Second, following calls to use additional theories in social marketing research (Dibb, 2014; Lefebvre, 2013), this study provides evidence of the applicability of the SDT to understand a social marketing phenomenon (flooding disaster mitigation behaviour adoption). To the best of the authors’ knowledge, this study is one of the few studies that have attempted to fully test the SDT theory of predictors for motivation and behaviour in a social marketing context, as little (if any) research exists that has examined the potential relationships between autonomy, competence or relatedness and motivation for behaviour adoption, particularly, from a developing country context. In particular, the study found need competence and autonomy to be the most important predictors of higher self-determined motivation and flood reduction behaviour. This study thus provides support for the applicability of the SDT in predicting the motivation for flood disaster mitigation adoption in a developing country context. Social marketers interested in improving flood disaster mitigation behaviours can enhance intervention efficacy by considering the factors proposed by the SDT as a way of changing behaviour. 5.2 Implications for practice The findings of this study also offer some implications for social marketing practitioners, policymakers and city planning officials as successes and failures in many social marketing interventions can be understood from the SDT perspective. In the current study, need competence relating to flood mitigation has proven to be the most important variable that predicts participants’ self-determined motivation for flood disaster mitigation. This was followed by need autonomy relating to the behaviour. To this end, we recommend that implementers of social marketing programs and policymakers aiming to promote the JSOCM 12,4 548 adoption of flood mitigation behaviours among individuals should not focus on relatedness because the study found a non-significant relationship between relatedness and self- determined motivation. They should rather focus more efforts on providing support for the individual’s sense of autonomy and competence. In other words, social marketers can enhance intervention efficacy through the support of individuals’ psychological needs for competence and autonomy. Doing so will not only enhance positive outcomes but also make interventions effective. They can, therefore, come up with communication messages that will support people’s sense of competence and autonomy to make them feel they are in charge or have control over the performance of the behaviour and that they are capable of performing the behaviour successfully without coercion (Deci and Ryan, 2000). According to Russell-Bennett et al. (2009), for consumers or individuals to exert effort to change and adopt a given behaviour, there must be a value proposition that incentivises them. The incentive could be an inherent desire to achieve a desired behavioural outcome which the individuals feel is based on their true interests and value, and although the task or behaviour may appear challenging, they will still perform it out of their own choice and volition, and not because of an external reward (Deci and Ryan, 2000; Ryan and Deci, 2017). Thus, Zinuddin et al. (2013) have suggested that policymakers and intervention planners, when designing social marketing programmes, must understand the motivational mechanisms that underlie the behaviour and what the individuals perceive as value to reduce the barriers to participation, as this will encourage sustainable behaviour change. Finally, we suggest that practitioners aiming to promote flood disaster mitigation behaviour should demonstrate care and interest, by providing an emphatic and warm interpersonal environment and opportunities that support individuals’ relationships and connection with their social groups as this will likely influence their self-determined motivation for the behaviour (Van den Broeck et al., 2016) and intervention effectiveness. 5.3 Limitations and future research directions In spite of the outlined contributions, the study has some limitations that need to be taken into consideration when interpreting the findings. First, the study tested the relationships among the constructs using urban dwellers only. Therefore, it will be interesting for future studies to test our model in non-urban settings for theory validation. Second, the research design used in this study was cross-sectional, which may suggest static relationships between the variables used at a single point in time. Future research may have to adopt longitudinal or experimental designs to determine the pattern of change and the extent to which self-determined motivations that are based on need competence and autonomy influence flood mitigation behaviour over time. Additionally, the current study applied the SDT in a specific pro-environmental context (flood disaster mitigation). Studies in a range of behavioural settings, including, e.g. illegal mining, energy conservation, recycling, would strengthen the theory. Finally, we relied on participants’ self-reports as a single source of information about the study variables, which may suffer social desirability bias. Future research would benefit from using objective measures, or it would be desirable to control for social desirability. In spite of the delineated limitations, the testing of a theoretical model that had all the paths included in the model (partial mediation model) adds to the literature in this area and offers direction for future research. 5.4 Conclusion The current study is the first, to the best of the authors’ knowledge, to attempt to find support for a theoretical model of motivation (specifically, SDT), which combines all three psychological needs dimensions (autonomy, competence and relatedness) and self- Understanding flooding events in Ghana 549 determined motivation in one model in a social marketing context. The study found that competence was the strongest predictor of self-determined motivation for flood disaster mitigation, followed by autonomy. Furthermore, self-determined motivation was found to positively influence individuals’ flood mitigation behaviour. Thus, we suggest that identified variables (i.e. competence and autonomy) that are important in predicting people’s self-determined motivation towards flood disaster mitigation should be prioritised in social marketing programmes and public campaigns that aim to promote effective and sustainable behaviour change. References Abotchie, P.N. and Shokar, N.K. (2009), “Cervical cancer screening among college students in Ghana: knowledge and health beliefs”, International Journal of Gynaecologic Cancer, Vol. 19 No. 3. Addo, K. and Adeyemi, M. (2013), “Assessing the impact of sea-level rise on a vulnerable coastal community in Accra, Ghana”, Jàmb�a: Journal of Disaster Risk Studies, Vol. 5 No. 1, pp. 1-8. Ahadzie, D.K. and Proverbs, D.G. (2011), “Emerging issues in the management of floods in Ghana”, International Journal of Safety and Security Engineering, Vol. 1 No. 2, pp. 182-192. Ajzen, I. (1991), “The theory of planned behaviour”, Organizational Behavior and Human Decision Processes, Vol. 50 No. 2, pp. 179-211. Alhassan, H., Asante, F.A., Oteng-Ababio, M. and Bawakyillenuo, S. (2018), “Application of theory of planned behaviour to households’ source separation behaviour in Ghana”, Management of Environmental Quality: An International Journal, Vol. 29 No. 4, pp. 704-721. AlQahtany, A.M. and Abubakar, I.R. (2020), “Public perception and attitudes to disaster risks in a coastal metropolis of Saudi Arabia”, International Journal of Disaster Risk Reduction, Vol. 44, p. 101422. Amoako, C. and Frimpong Boamah, E. (2015), “The three-dimensional causes of flooding in Accra, Ghana”, International Journal of Urban Sustainable Development, Vol. 7 No. 1, pp. 109-129. Amoako, C. and Inkoom, D.K.B. (2018), “The production of flood vulnerability in Accra, Ghana: re- thinking flooding and informal urbanization”,Urban Studies, Vol. 55 No. 13, pp. 2903-2922. Anderson, J.C. and Gerbing, D.W. (1988), “Structural equation modelling in practice: a review and recommended two-step approach”, Psychological Bulletin, Vol. 103 No. 3, p. 411. Ansah, S.O., Ahiataku, M.A., Yorke, C.K., Otu-Larbi, F., Yahaya, B., Lamptey, P.N.L. and Tanu, M. (2020), “Meteorological analysis of floods in Ghana”,Advances inMeteorology, Vol. 2020. Armstrong, J.S. and Overton, T.S. (1977), “Estimating nonresponse bias in mail surveys”, Journal of Marketing Research, Vol. 14 No. 3, pp. 396-402. Asumadu-Sarkodie, S. Owusu, P.A. and Jayaweera, M. (2015), “Flood risk management in Ghana: a case study in Accra”. Awwad, M.S. and Agti, D.A.M. (2011), “The impact of internal marketing on commercial banks’market orientation”, International Journal of BankMarketing, Vol. 29 No. 4, pp. 308-332. Ayikwa, L.C., De Jager, J.W. and Van Zyl, D. (2020), “Validating the theory of planned behaviour model extended to social marketing behavioural enhancers using structural equation modelling”, Social Marketing Quarterly, Vol. 26 No. 4, pp. 276-296. Bandura, A. (1997), Self-efficacy: The Exercise of Control, Freeman, NewYork. Bagozzi, R.P. and Yi, Y. (2012), “Specification, evaluation, and interpretation of structural equation models”, Journal of the Academy ofMarketing Science, Vol. 40 No. 1, pp. 8-34. Baxter, D. and Pelletier, L.G. (2020), “The roles of motivation and goals on sustainable behaviour in a resource dilemma: a self-determination theory perspective”, Journal of Environmental Psychology, Vol. 69, p. 101437. JSOCM 12,4 550 Binney, W., Hall, J. and Oppenheim, P. (2006), “The nature and influence of motivation within the MOA framework: implications for social marketing”, International Journal of Nonprofit and Voluntary SectorMarketing, Vol. 11 No. 4, pp. 289-301. Carins, J.E. and Rundle-Thiele, S.R. (2014), “Eating for the better: a social marketing review (2000– 2012)”, Public Health Nutrition, Vol. 17 No. 7, pp. 1628-1639. Chau, P.Y. and Hu, P.J. (2002), “Examining a model of information technology acceptance by individual professionals: an exploratory study”, Journal of Management Information Systems, Vol. 18 No. 4, pp. 191-229. Chen, W. (2014), “Psychological needs satisfaction, motivational regulations and physical activity intention among elementary school students”, Educational Psychology, Vol. 34 No. 4, pp. 495-511. Chen, B., Van Assche, J., Vansteenkiste, M., Soenens, B. and Beyers, W. (2015), “Does psychological need satisfaction matter when environmental or financial safety are at risk?”, Journal of Happiness Studies, Vol. 16 No. 3, pp. 745-766. Cooke, A.N., Fielding, K.S. and Louis, W.R. (2016), “Environmentally active people: the role of autonomy, relatedness, competence and self-determined motivation”, Environmental Education Research, Vol. 22 No. 5, pp. 631-657. Couper, M.P. (2000), “Usability evaluation of computer-assisted survey instruments”, Social Science Computer Review, Vol. 18 No. 4, pp. 384-396. Darner, R. (2007), “Supporting students self-determined environmental motivation in a formal environmental education setting”, Doctoral dissertation, University of California, San Diego, San Diego, CA. De Gregorio, F. and Sung, Y. (2010), “Understanding attitudes toward and behaviours in response to product placement”, Journal of Advertising, Vol. 39 No. 1, pp. 83-96. Deci, E.L. and Ryan, R.M. (2000), “The ‘what’ and ‘why’ of goal pursuits: human needs and the self- determination of behavior”, Psychological inquiry, Vol. 11 No. 4, pp. 227-268. Deci, E.L. and Ryan, R.M. (2014), “Autonomy and need satisfaction in close relationships: relationships motivation theory”, Human Motivation and Interpersonal Relationships, Springer, Dordrecht, pp. 53-73. Dholakia, U.M. (2006), “How customer self-determination influences relational marketing outcomes: evidence from longitudinal field studies”, Journal of Marketing Research, Vol. 43 No. 1, pp. 109-120. Dibb, S. (2014), “Up, up and away: social marketing breaks free”, Journal of Marketing Management, Vol. 30 Nos 11/12, pp. 1159-1185. Doku, D. (2012), “Substance use and risky sexual behaviours among sexually experienced Ghanaian youth”, BMCPublic Health, Vol. 12 No. 1, pp. 1-7. Engström, J. and Elg, M. (2015), “A self-determination theory perspective on customer participation in service development”, Journal of ServicesMarketing, Vol. 29 Nos 6/7, pp. 511-521. Erman, A., Motte, E., Goyal, R., Asare, A., Takamatsu, S., Chen, X. and Hallegatte, S. (2018), “The road to recovery: the role of poverty in the exposure, vulnerability and resilience to floods in Accra”, TheWorld Bank. Evans, P. (2015), “Self-determination theory: an approach to motivation in music education”, Musicae Scientiae, Vol. 19 No. 1, pp. 65-83. Fatti, C.E. and Patel, Z. (2013), “Perceptions and responses to urban flood risk: implications for climate governance in the South”,Applied Geography, Vol. 36, pp. 13-22. Flannery, M. (2017), “Self-Determination theory: intrinsic motivation and behavioural change”, Oncology Nursing Forum, Vol. 44 No. 2. Fornell, C. and Lacker, D.F. (1981), “Evaluating structural equationmodels with unobservable variables and measurement error”, Journal of Marketing Research, Vol. 18 No. 1, pp. 39-50. Understanding flooding events in Ghana 551 Fowler, A. (2014),Milton: Paradise Lost, Routledge. Gagne, M. (2003), “Autonomy support and need satisfaction in the motivation and well-being of gymnasts”, Journal of Applied Sport Psychology, Vol. 15 No. 4, pp. 372-390. Ghani, W.A.W.A.K., Rusli, I.F., Biak, D.R.A. and Idris, A. (2013), “An application of the theory of planned behaviour to study the influencing factors of participation in source separation of food waste”,WasteManagement, Vol. 33 No. 5, pp. 1276-1281. Gilal, F.G., Zhang, J., Gilal, N.G. and Gilal, R.G. (2019), “Linking self-determined needs and word of mouth to consumer e-waste disposal behaviour: a test of basic psychological needs theory”, Journal of Consumer Behaviour, Vol. 18 No. 1, pp. 12-24. Glago, F.J. (2019), “Household disaster awareness and preparedness: a case study of flood hazards in Asamankese in the west Akim municipality of Ghana”, Jamba: Journal of Disaster Risk Studies, Vol. 11 No. 1, pp. 1-11. Glago, F.J. (2021), “Flood disaster hazards; causes, impacts and management: a state-of-the-art review”, Natural Hazards-Impacts, Adjustments and Resilience. Graphic Online (2020), “June 3, 2015 flood, fire disaster-have we learnt lessons?”, available at: www. graphic.com.gh/?view=article&id=140617:june-3-2015-flood-fire-disaster-have-we-learnt-lessons& catid=257 (accessed April 2021). Gruneklee, N., Rundle-Thiele, S. and Kubacki, K. (2016), “What can social marketing learn from Dirichlet theory patterns in a physical”, Marketing Intelligence and Planning, Vol. 34 No. 1, pp. 41-60. Guay, F., Ratelle, C.F., Roy, A. and Litalien, D. (2010), “Academic self-concept, autonomous academic motivation, and academic achievement: mediating and additive effects”, Learning and Individual Differences, Vol. 20 No. 6, pp. 644-653. Gyekye, A.K. (2013), “Environment change and flooding in Accra, Ghana”, Sacha J Environ Stud, Vol. 3, pp. 65-80. Hagger, M.S. and Chatzisarantis, N.L. (2009), “Integrating the theory of planned behaviour and self- determination theory in health behaviour: a meta-analysis”, British Journal of Health Psychology, Vol. 14 No. 2, pp. 275-302. Haghighatjoo, S., Tahmasebi, R. and Noroozi, A. (2020), “Application of community-based social marketing (CBSM) to increase recycling behaviour (RB) in primary schools”, Social Marketing Quarterly, Vol. 26 No. 4, pp. 297-308. Hair, J.F., Ringle, C.M. and Sarstedt, M. (2013), “Partial least squares structural equation modelling: rigorous applications, better results and higher acceptance”, Long Range Planning, Vol. 46 Nos 1/2, pp. 1-12. Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006),Multivariate Data Analysis, 6th ed., Pearson Prentice Hall, Upper Saddle River, NJ. Haj-Salem, N. and Al-Hawari, M.A. (2021), “Predictors of recycling behaviour: the role of self-conscious emotions”, Journal of Social Marketing, Vol. 11 No. 3, pp. 204-223. Hiratsuka, V.Y., Beans, J.A., Robinson, R.F., Shaw, J.L., Sylvester, I. and Dillard, D.A. (2017), “Self-determination in health research: an Alaska native example of tribal ownership and research regulation”, International Journal of Environmental Research andPublic Health, Vol. 14No. 11, p. 1324. Hodgkins, S., Rundle-Thiele, S., Knox, K. and Kim, J. (2019), “Utilizing stakeholder theory for social marketing process evaluation in a food waste context”, Journal of Social Marketing, Vol. 9 No. 3, pp. 87-270. Hooper, D., Coughlan, J. and Mullen, M. (2008), “Evaluating model fit: a synthesis of the structural equation modelling literature”, in 7th European Conference on Research Methodology for Business andManagement Studies, pp. 195-200. Huang, T.L., Mathews, S. and Chou, C.Y. (2019), “Enhancing online rapport experience via augmented reality”, Journal of ServicesMarketing, Vol. 33 No. 7, pp. 851-865. JSOCM 12,4 552 http://www.graphic.com.gh/?view=article&id=140617:june-3-2015-flood-fire-disaster-have-we-learnt-lessons&catid=257 http://www.graphic.com.gh/?view=article&id=140617:june-3-2015-flood-fire-disaster-have-we-learnt-lessons&catid=257 http://www.graphic.com.gh/?view=article&id=140617:june-3-2015-flood-fire-disaster-have-we-learnt-lessons&catid=257 Hu, L.T. and Bentler, P.M. (1999), “Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives”, Structural Equation Modeling: A Multidisciplinary Journal, Vol. 6 No. 1, pp. 1-55. Julinawati, S., Cawley, D., Domegan, C., Brenner, M. and Rowan, N.J. (2013), “A review of the perceived barriers within the health belief model on pap smear screening as a cervical cancer prevention measure”, Journal of Asian Scientific Research, Vol. 3 No. 6, pp. 677-692. Kaiser, L.M., Großmann, N. and Wilde, M. (2020), “The relationship between students’ motivation and their perceived amount of basic psychological need satisfaction – a differentiated investigation of students’ quality of motivation regarding biology”, International Journal of Science Education, Vol. 42 No. 17, pp. 2801-2818. Korah, P.I. and Cobbinah, P.B. (2019), “Institutional responses to climate change adaptation: flood management at the metropolitan level in Accra, Ghana”, The Geography of Climate Change Adaptation in Urban Africa, Palgrave Macmillan, Cham, pp. 451-478. Krabbe, P. (2016), The Measurement of Health and Health Status: Concepts, Methods and Applications from aMultidisciplinary Perspective, Academic Press. Lavergne, K.J., Sharp, E.C., Pelletier, L.G. and Holtby, A. (2010), “The role of perceived government style in the facilitation of self-determined and non-self-determined motivation for pro-environmental behaviour”, Journal of Environmental Psychology, Vol. 30 No. 2, pp. 169-177. Lefebvre, R.C. (2013), Social Marketing and Social Change: Strategies and Tools for Improving Health, Well-Being, and the Environment, JohnWiley and Sons. Lin, C.P., Tsai, Y.H. and Chiu, C.K. (2009), “Modelling customer loyalty from an integrative perspective of self-determination theory and expectation–confirmation theory”, Journal of Business and Psychology, Vol. 24 No. 3, pp. 315-326. Logah, F.Y., Kankam-Yeboah, K., Ofori, D. and Gyau-Boakye, P. (2014), “Flood pulse alterations of some river basins in Ghana”,Ghana Journal of Science, Vol. 12, pp. 19-32. Lu, Y., Zhou, L., Bruton, G. and Li, W. (2010), “Capabilities as a mediator linking resources and the international performance of entrepreneurial firms in an emerging economy”, Journal of International Business Studies, Vol. 41 No. 3, pp. 419-436. Luca, M. and Zervas, G. (2016), “Fake it till you make it: reputation, competition, and yelp review fraud”,Management Science, Vol. 62 No. 12, pp. 3412-3427. Mensah, H. and Ahadzie, D.K. (2020), “Causes, impacts and coping strategies of floods in Ghana: a systematic review”, SNApplied Sciences, Vol. 2 No. 5, pp. 1-13. Mitchell, R., Schuster, L. and Drennan, J. (2017), “Understanding how gamification influences behaviour in social marketing”,AustralasianMarketing Journal, Vol. 25 No. 1, pp. 12-19. Modern Ghana (2021), available at: www.modernghana.com/news/1108038/accra-kumasi-flood-again- after-heavy-downpour.html (accessed August 2021). Nair, S. and Abraham, J. (2018), “Hazardous waste management with special reference to biological treatment”,Handbook of Environmental Materials Management, Springer, Cham, pp. 1-27. Ng, J.Y., Ntoumanis, N., Thøgersen-Ntoumani, C., Deci, E.L., Ryan, R.M., Duda, J.L. and Williams, G.C. (2012), “Self-determination theory applied to health contexts: a meta-analysis”, Perspectives on Psychological Science, Vol. 7 No. 4, pp. 325-340. Niemiec, C.P. and Ryan, R.M. (2009), “Autonomy, competence, and relatedness in the classroom: applying self-determination theory to educational practice”, Theory and Research in Education, Vol. 7 No. 2, pp. 133-144. Okyere, C.Y., Yacouba, Y. and Gilgenbach, D. (2013), “The problem of annual occurrences of floods in Accra: an integration of hydrological, economic and political perspectives”, Theoretical and Empirical Researches in UrbanManagement, Vol. 8 No. 2, pp. 45-79. Understanding flooding events in Ghana 553 http://www.modernghana.com/news/1108038/accra-kumasi-flood-again-after-heavy-downpour.html http://www.modernghana.com/news/1108038/accra-kumasi-flood-again-after-heavy-downpour.html Osbaldiston, R. and Sheldon, K.M. (2003), “Promoting internalized motivation for environmentally responsible behaviour: a prospective study of environmental goals”, Journal of Environmental Psychology, Vol. 23 No. 4, pp. 349-357. Owusu-Ansah, J.K. (2016), “The influences of land use and sanitation infrastructure on flooding in Kumasi, Ghana”,GeoJournal, Vol. 81 No. 4, pp. 555-570, doi: 10.1007/s10708-015-9636-4. Pacewicz, C.E., Smith, A.L. and Raedeke, T.D. (2020), “Group cohesion and relatedness as predictors of self-determined motivation and burnout in adolescent female athletes”, Psychology of Sport and Exercise, Vol. 50, p. 101709. Pelletier, L.G., Baxter, D. and Huta, V. (2011), “Personal autonomy and environmental sustainability”, HumanAutonomy in Cross-Cultural Context, Springer, Dordrecht, pp. 257-277. Pelletier, L.G., Tuson, K.M., Green-Demers, I., Noels, K. and Beaton, A.M. (1998), “Why are you doing things for the environment? The motivation toward the environment scale (MTES)”, Journal of Applied Social Psychology, Vol. 8 No. 5, pp. 437-468. Pentecost, R., Arli, D. and Thiele, S. (2017), “It’s my choice! Investigating barriers to pro-social blood donating behaviour”,Marketing Intelligence and Planning, Vol. 35 No. 2, pp. 243-258. Podsakoff, P.M. and Organ, D.W. (1986), “Self-reports in organizational research: problems and prospects”, Journal ofManagement, Vol. 12 No. 4, pp. 531-544. Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y. and Podsakoff, N.P. (2003), “Common method biases in behavioural research: a critical review of the literature and recommended remedies”, Journal of Applied Psychology, Vol. 88 No. 5, p. 879. Prochaska, J.O., Norcross, J.C. and DiClemente, C.C. (2013), “Applying the stages of change”, Psychotherapy in Australia, Vol. 19 No. 2, pp. 10-15. Qian, T.Y., Wang, J.J., Zhang, J.J. and Hulland, J. (2020), “Fulfilling the basic psychological needs of exports fans: a self-determination theory approach”, Communication and Sport, Vol. 10 No. 2, pp. 216-240. Russell-Bennett, R., Previte, J. and Zainuddin, N. (2009), “Conceptualizing value creation for social change management”,AustralasianMarketing Journal, Vol. 17 No. 4, pp. 211-218. Ryan, R.M. and Deci, E.L. (2017), Self-Determination Theory: Basic Psychological Needs in Motivation, Development, andWellness, Guilford Publications. Sanders, D.B., Salvato, M., Aussel, H., Ilbert, O., Scoville, N., Surace, J.A., Frayer, D.T., Sheth, K., Helou, G., Brooke, T. and Bhattacharya, B. (2007), “S-COSMOS: the Spitzer legacy survey of the Hubble space telescope ACS 2 deg2 COSMOS field I: survey strategy and first analysis”, The Astrophysical Journal Supplement Series, Vol. 172 No. 1, p. 86. Sheldon, K.M., Elliot, A.J., Kim, Y. and Kasser, T. (2001), “What is satisfying about satisfying events? Testing 10 candidate psychological needs”, Journal of Personality and Social Psychology, Vol. 80 No. 2, p. 325. Schreiber, J.B., Nora, A., Stage, F.K., Barlow, E.A. and King, J. (2006), “Reporting structural equation modeling and confirmatory factor analysis results: a review”, The Journal of Educational Research, Vol. 99 No. 6, pp. 323-338. Sheldon, K.M., Wineland, A., Venhoeven, L. and Osin, E. (2016), “Understanding the motivation of environmental activists: a comparison of self-determination theory and functional motives theory”, Ecopsychology, Vol. 8 No. 4, pp. 228-238. Slemp, G.R., Kern, M.L., Patrick, K.J. and Ryan, R.M. (2018), “Leader autonomy support in the workplace: a meta-analytic review”,Motivation and Emotion, Vol. 42 No. 5, pp. 706-724. Spotswood, F., French, J., Tapp, A. and Stead, M. (2012), “Some reasonable but uncomfortable questions about social marketing”, Journal of Social Marketing, Vol. 2 No. 3, pp. 163-175. Standage, M., Duda, J.L. and Ntoumanis, N. (2003), “A model of contextual motivation in physical education: using constructs from self-determination and achievement goal theories to predict physical activity intentions”, Journal of Educational Psychology, Vol. 95 No. 1, p. 97. JSOCM 12,4 554 http://dx.doi.org/10.1007/s10708-015-9636-4 Sweeney, J.C., Webb, D., Mazzarol, T. and Soutar, G.N. (2014), “Self-determination theory and word of mouth about energy-saving behaviours: an online experiment”, Psychology and Marketing, Vol. 31 No. 9, pp. 698-716. Tabiri, M.O. (2015), “Perennial problem of floods in Accra: what is the way forward?”, International Journal of Advanced Research and Technology, Vol. 2, pp. 198-201. Trépanier, S.G., Forest, J., Fernet, C. and Austin, S. (2015), “On the psychological and motivational processes linking job characteristics to employee functioning: insights from self-determination theory”,Work and Stress, Vol. 29 No. 3, pp. 286-305. Tweneboah-Koduah, E.Y., Adams, M. and Acheampong, G. (2019), “The role of theories in social marketing in predicting physical activity behaviour among the youth”, Journal of Social Marketing, Vol. 9 No. 4, pp. 398-417, doi: 10.1108/JSOCM-01-2018-0005. Valois, P., Bouchard, D., Talbot, D., Caron, M., Renaud, J.S., Gosselin, P. and Jacob, J. (2020), “Adoption of flood-related preventive behaviours by people having different risks and histories of flooding”,Natural Hazards, Vol. 102 No. 3, pp. 1155-1173. Van den Broeck, A., Ferris, D.L., Chang, C.H. and Rosen, C.C. (2016), “A review of self-determination theory’s basic psychological needs at work”, Journal ofManagement, Vol. 42 No. 5, pp. 1195-1229. Venhoeven, L.A., Bolderdijk, J.W. and Steg, L. (2013), “Explaining the paradox: how pro-environmental behaviour can both thwart and foster well-being”, Sustainability, Vol. 5 No. 4, pp. 1372-1386. Webb, D., Soutar, G.N., Mazzarol, T. and Saldaris, P. (2013), “Self-determination theory and consumer behavioural change: evidence from a household energy-saving behaviour study”, Journal of Environmental Psychology, Vol. 35, pp. 59-66. Weinstein, N. and Ryan, R.M. (2010), “When helping helps: autonomous motivation for prosocial behaviour and its influence on well-being for the helper and recipient”, Journal of Personality and Social Psychology, Vol. 98 No. 2, p. 222. World Health Organization (2019), “Floods”, available at: www.who.int/health-topics/floods#tab=tab (accessed January 2021 andMarch 2021). Further reading Bandura, A. (2006), “Guide for constructing self-efficacy scales”, Self-Efficacy Beliefs of Adolescents, Vol. 5 No. 1, pp. 307-337. Bonfield, P. (2016),The Property Flood Resilience Action Plan, Defra, London. Deci, E.L. and Ryan, R.M. (2008), “Self-determination theory: a macro theory of human motivation, development, and health”,Canadian Psychology/Psychologie Canadienne, Vol. 49 No. 3, pp. 182-185. Deci, E.L. and Ryan, R.M. (2010), “Self-determination”,The Corsini Encyclopaedia of Psychology, pp. 1-2. Zainuddin, N., Russell-Bennett, R. and Previte, J. (2013), “The value of health and wellbeing: an empirical model of value creation in social marketing”, European Journal of Marketing, Vol. 47 No. 9, pp. 1504-1524. Corresponding author Ernest Yaw Tweneboah-Koduah can be contacted at: etkoduah@ug.edu.gh For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com Understanding flooding events in Ghana 555 http://dx.doi.org/10.1108/JSOCM-01-2018-0005 http://www.who.int/health-topics/floods#tab=tab mailto:etkoduah@ug.edu.gh Understanding flooding events in Ghana: a social marketing and self-determination theory perspective 1. Introduction 2. Theoretical context and hypotheses development 2.1 Hypotheses development Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed 3. Methodology 3.1 Sampling and data collection 3.2 Measures 3.3 Common method bias 4. Data analyses and results 4.1 Profile of the respondents 4.2 Confirmatory factor analyses and structural equation modelling Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed Undefined namespace prefix xmlXPathCompOpEval: parameter error xmlXPathEval: evaluation failed 5. Discussion and conclusion 5.1 Theoretical implications 5.2 Implications for practice 5.3 Limitations and future research directions 5.4 Conclusion References