University of Ghana http://ugspace.ug.edu.gh
REGIONAL INSTITUTE FOR POPULATION STUDIES 
UNIVERSITY OF GHANA 
LEGON 
MULTIPLE SEXUAL PARTNERSHIPS IN GHANA 
EVIDENCE FROM 2003 GHANA DEMOGRAHIC AND HEALTH SURVEY 
(GDHS, 2003) 
BY 
JOHN KP ANNIE TETTEH 
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIMENTS FOR THE M.A. 
(pOPULATION STUDIES), UNIVERSITY OF GHANA, LEGON 
JULY, 2009 
University of Ghana http://ugspace.ug.edu.gh
ACCEPTANCE 
Accepted by the faculty of Social Studies, University of Ghana, Legon, in partial 
fulfillment of the requirements for the degree of M.A. (population Studied). 
~D - f l- 6 ~ 
Date .............................. ) .......... . 
University of Ghana http://ugspace.ug.edu.gh
DECLARATION 
I hereby declare that except for the reference other people's work which has been duly 
acknowledged, this is the result of my own research and it has neither in part or in whole 
been presented for another degree. 
Signed .. d..~).U?: .f ...... . 
JOHN KP ANNIE TETTEH 
(Candidate) 
Date .....~ . 9. ..- ::-: .. !( .. .: -:. ... ?:-.9~ .~. 
11 
University of Ghana http://ugspace.ug.edu.gh
DEDICATION 
I dedicated this work to my wife Rebecca Tetteh who has been very 
supportive and understanding throughout the period of my study_ 
111 
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGMENT 
Thanks be to God Almighty for His marvelous grace and mercies that has seen me through 
this course. My special thanks go to the entire staff of RIPS for their 
co-orperation and assistance throughout the course of my studies. 
I am exceedingly grateful to the industrious and hard working teaching staff of RIPS for 
their immense help and contribution in making my studies here a success. lowe my 
supervisor, Dr Samuel Nii A. Codjoe a particular appreciation and gratitude for using his 
rich experience to guide me from the onset of this work to the final presentation of it. His 
probing and challenging question, criticisms and useful suggestions all contributed to make 
this work possible. I extend my appreciation to the secretary Antie Becky who helped in 
printing some of the manuscripts and Mr. Eric August and his team at the computer 
department for their valuable contributions to this work. 
JOHN KP ANNIE TETTEH 
iv 
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
TITLE PAGE 
ACCEPTANCE .......................... ... ....................................... .......... .i 
DECLARATION ........~  .......................................... , ........................... .ii 
DEDICATION ................................................................................ .iii 
ACKNOWLEGEMENT .................................................................... .iv 
TABLE OF CONTENTS ....................................................................v  
LIST OF FIGURES ......................................................................... vii 
LIST OF TABLES ..........................................................................v iii 
ABSTRACT .................................................................................. .ix 
CHAPTER ONE: INTRODUCTION •....••............•.•...............•...•••••...... 1 
1.1. Background Information ......................................... ; ....................... 1 
1.2. Statement of the Problem ................................................................4  
1.3. Rational For the Study .................................................................. 5 
1.4. Objectives of the Study .................................................................. 6 
1.5. Literature Review ........................................................................ 8 
1.6. Conceptual Framework ..................................................................... 14 
1.7. Hypotheses ................................................................................ 16 
1.8. Methodology ....... ................ ....................................................... 16 
1.8.1 Source of Data. ......................................................................... .17 
v 
University of Ghana http://ugspace.ug.edu.gh
1.8.2 Limitation of Data ...................................................................... 18 
1.8.3 Method of Analysis ..................................................................... 18 
1.9 Definition of Concepts ................................................................... 19 
1.10. Plan of the Study ................................................. . .............. ...... .. 20 
CHAPTER TWO: EVALUATION OF DATA AND ANALYSIS OF DEMOGRAPIDC AND 
SOCIO-ECONOMIC CHARACTERISTICS OF 
RESPONDENTS ................................. .. ........................ ... ..... .. ........ . ..........2 1 
2.0. Introduction .................................. .. ......... ... .... ....... ................ ...... 21 
2.1. Evaluation of Data ................... .. ...... .. .. . ...... ...... .. .. ... . , ... ....... .... .......2 1 
2.2. Demographic and Socio-economic Characteristics of respondents ................... 23 
2.2.1. Age Groups of Respondents .............................................................2 3 
2.2.2. Current Marital Status .................................................................... 24 
2.2.3. Educational Status of Respondents .................................................... .25 
2.2.4 Region and Place of Residence .......................................... '" ............. 26 
2.2.5. Wealth Status of Respondents .......................................................... 28 
2.2.6. Occupational Status ...................................................................... 29. 
2.2.7. Religious Affiliation of Respondents .................................................. .30 
2.2.8. Respondents' Age at First Sex .................................... . .................... .42 
CHAPTER THREE: PATTERN AND DIFFERENTIALS IN SEXUAL 
PARTNERSHIPS ................................................................................3 2 
3.0. Introduction .................................................................................. 32 
3.1. Marital Status and Sexual Partnerships ................................................... 32 
3.2. Age and Sexual Partnerships ...............................................................3 4 
Vi 
University of Ghana http://ugspace.ug.edu.gh
3.4. Educational Status and Sexual Partn~rships ............ .... .. ................... .......... 35 
3.4. Religious Affiliation and Sexual Partnerships ............................................ .37 
3.5. Wealth Status and Sexual Partnerships .....................................................3 8 
3.6. Region of Residence and Sexual Partnerships ..............................................3 9 
3.7. Place of Residence and Sexual Partnerships ............................................... .40 
3.8. Age At First Sex and Sexual Partnerships ............................................... .42 
3.9. Occupational Status and Sexual Partnerships ...............................................4 3 
CHAPTER FOUR: DETERMINANTS OF MULTIPLE SEXUAL P ARTNERSHIPS ••••••• .45 
4.0. Introduction ................................... . ..................................................4 5 
4.1. Determinants of Multiple Sexual partnerships .................... . ........................ .46 
CHAPTER FIVE: SUMMARY, CONCLUSION AND RECOMMENDATIONS •.•••••.•.•• 50 
5.1 Summary ............... .. ....... ... .......................... ........ ........................... 50 
5.2 Conclusion ........................................................................................ 54 
5.3. Recommendations ....................................... .... . ..... .... ........ ................5 5 
BffiLIOGRAPHY ......... ........................ ...... ..... ........ .............................. 58 
LIST OF FIGURES 
Fig 1.1 Conceptual Framework ................................................................ 16 
vii 
University of GLhIaSTna O hFt tTpA:/B/uLgEsSp ace.ug.edu.gh
Table 2.1 Evaluation of Age data, Using United Nations Age Ratios ................... 22 
Table 2.2 Percentage distributions of Respondents by Age group .. ...... ....... .. .......2 4 
Table 2.3 Percentage distributions of Respondents by Current Marital Status ........ .25. 
Table 2.4 Percentage distributions of Respondents by Educationallevel.. .......... .... 26 
Table 2.5a Percentage distribution of Respondents by Region ........................... 27 
Table 2.5b Percentage distribution of Respondents, by type of Residence ............. .28 
Table 2.6 Percentage distribution of Respondents by Wealth Status .................. .28 
Table 2.7 Percentage distribution of Respondents by Type ofOccupation ... .... ...... 29 
Table 2.8 Distribution of Respondents by Religious Affiliation .........................3 0 
Table 2.9 Percentage distribution of Respondents by Age at first sex .................. 31 
Table 3.1 Differentials in Sexual Partnerships by Marital status ........................ 34 
Table 3.2 Differentials in Sexual Partnerships by Age Groups ........................ .35 
Table 3.3 Pattern of sexual partnerships by Educational Attainment ...................3 6 
Table 3.4 Pattern of sexual partnerships by Religious Affiliation .......................3 8 
Table 3.5 Pattern of sexual partnerships by Wealth Status ...............................3 9 
Table 3.6 Regional Pattern in Sexual Partnerships ....................................... .40 
Table 3.7 Pattern of sexual partnerships by Place ofResidence .......................... 41 
Table 3.8 Age at First Sex and Sexual partnerships ...................................... .43 
Table 3.9 Pattern of sexual partnerships by Occupational Status of 
respondent ............................................... .............................4 4 
Table 4 Logistic Regression models of Sexual Partnerships by Demographic 
And Socio- economic Characteristics of men in Ghana Table 2.1.. ... ..... .48 
viii 
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT 
The objective of the study was to examine the magnitude and determinants of 
multiple sexual partnerships with regard to men's sexual behaviour in Ghana and the 
implication for public policy. 
The study employed data from the 2003 Ghana Demographic and Health Survey 
(GDHS, 2003) to examine the relationships between some selected socio-economic and 
demographic factors and sexual partnerships in Ghana. Nine independent variables formed 
the socio-economic characteristics of respondents. These were: age, marital status, 
occupation, religion, place of residence, region, wealth, age at first sex and education. The 
levels of sexual partnerships by background characteristics wee shown in the study. Both 
bivariate and multivariate analyses were employed to identify the most important factors 
determining sexual partnerships in Ghana. For the multivariate analysis, the logistic 
regression model was used. 
The study revealed that multiple sexual partnerships exists, and that about 15% of 
men aged 15 to 59 years are engaging in multiple sexual partnerships in Ghana. This 
sexual behaviour varies with socio-economic status of respondents. The study further 
shows that the most influencing factors of this behaviour are: occupational and educational 
status of respondents, place of residence and age at first sex. For example, bivariate 
analysis indicated that age at first sex has an inverse relationship with multiple sexual 
partnerships. It was also found that men in urban communities and those with high level of 
education are more likely to have more than one sexual partner than men in rural localities 
ix 
University of Ghana http://ugspace.ug.edu.gh
and those without any formal education ~espectively. These results were also substantiated 
by the multivariate or logistic analysis. 
The study made several recommendations including the importance of making sex 
education as part of the school curricular from basic to Junior High School level and the 
setting up of recreational and communities centers across the country. At these centers, 
indoor and outdoor games as well as counseling sections on issues about sex and the 
importance of family planning should form part of the programs for the youth. Older men 
in our communities were also called upon to show excellent examples for the youth to 
follow, especially married men who must realize the importance of faithfulness and respect 
for their wives, the people they love. 
x 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE 
INTRODUCTION 
1.1 Background Information 
"There would be no global AIDS pandemic were it not for multiple sexual 
partnerships and that more should be done to encourage people to reduce their number of 
sexual partners" according to an "education and debate" article written by officials from 
several organizations and published in April 10, 2004 issue ofBMJ, BBC News report 
(BBC News, 4/8). Shelton and co (2004), write that the "key to preventing the spread of 
HN, especially in epidemics driven mainly by heterosexual transmission, is through 
changing sexual behaviour. A study by Adisegon and Wenjuan (2004), reported that multiple 
sexual partnerships (defmed as two or more sexual partnerships that overlap in time) mediate 
the association between early sexual debut and sexually transmitted infection among 
adolescent and young adult males in Africa. 
According to Childs and Williams (2007), at any given time about 10% of men the 
world over are engaging in multiple sexual partnerships with women. Within a culture, men 
and women are not necessarily held to the same standards. For example, a man mayor may 
not be considered promiscuous for engaging in sexual activity with someone he was not 
married to, even in cultures where a woman would be considered promiscuous for the same 
behaviour. According to Mihausen and Harold (1999), men are "allowed" to have multiple 
partners while women are penalized by sexual double standards. Gender roles for 
heterosexual interactions appear to sanction men's sexual risk-taking especially the pursuit 
of multiple sexual partnerships (O'Sullivan, 2006). 
1 
University of Ghana http://ugspace.ug.edu.gh
Demographic and Health surveys in Africa have suggested that there is some 
association between men's socio-economic status and their sexual risk-taking behaviour. For 
example the 2004 Demographic and Health Survey (DHS) in Cameroon revealed a higher 
prevalence of HIV in the richest and most educated people than their poorest and least 
educated compatriots. Studies by Djambain (2004), indicated that socio-economic factors 
play some role in individual sexual risk taking behaviour in sub-Saharan Africa. A socio-
demographic study in Ghana and Kenya by A wusabo-Asare and Annim (2008), revealed 
some relationship between wealth status and sexual risk-taking behaviour in the two 
countries. The 2006 Multiple Indicator Cluster Survey in Ghana (MICS, 2006) reported that 
the proportion of men who had sex with more than one partner 12 months prior to the survey 
was highest (20.2%) within the age group 20-24 years and lowest (7.2%) within the 45-49 
age group. According to this survey, multiple sexual partnership increases with the wealth 
level of men. 
Sexual and reproductive health, especially of the youth has become a topical issue 
and major concern to most governments. There are many issues that undermine the human 
resource and development of young men and this include sound reproductive health in terms 
of safe and responsible sexual behaviour. The sexual behavior of men should be a concern 
because they make reproductive decisions and influence the sexual activity of their female 
counterparts (Ricter, 1996; Jewkes, 1997). Women often cannot control how many children 
they should have, though they are the ones exposed to the risk of pregnancy and childbirth. 
According to Banuako (1975), premarital sex which resulted in pregnancy was 
considered a criminal offense in the past. Offenders were punished and the sanctions ranged 
from ritual cleansing to ostracism of the couple. Premarital sex is not uncommon in Ghana 
2 
University of Ghana http://ugspace.ug.edu.gh
It is high, particularly among males. According to Anarfi (1999), by age 18 years, about 
59.5% of Ghanaian men have experienced premarital sex. About two-thirds of male 
respondents in the study reported to have had two or more premarital sexual partners in their 
life time. According to most of the respondents, they indulged in premarital sex either for 
pleasure or in order to emulate their siblings and peer pressures to fulfill what they termed a 
man's role. 
Extra marital sex is also a common feature in men's sexual behaviour. Some men 
indUlge in extramarital sex when they are dissatisfied with their wives' behaviour, in 
particular their lack of sexual cooperation. Others do it when their wives are practicing 
postpartum abstinence or when they are away (Childs and Williams, 2007). 
Research has shown that multiple sexual partnerships involving Ghanaian men 
vary across the 10 regions of the country as well as the age, education and wealth status of 
men. For example, according to MICS (2006), the percentage of men who had sex with two 
or more sexual partners in the last 12 months prior to the survey was 21.3% (maximum) for 
Central region and 8% (minimum) for Volta region. Ibis multiple sexual partnership was 
higher within the age group 20-24 (20.2%) and lower within the 44-49 age group (7.2%). 
The survey also reported that the proportion of men having more than one sexual partner was 
high for men living in urban areas and for those with at least secondary level of education. 
According to Dudgeon and Inhorn (2004), men's sexual behaviours (including their 
use of barrier contraceptives) have major implications for the transmission ofSTIs, including 
3 
University of Ghana http://ugspace.ug.edu.gh
bacterial, viral, and parasitic agents that can lead to acute and chronic conditions in both men 
and women, as well as pregnancy-associated diseases that affect the well-being of offspring. 
High prevalence of HIV in both West and East Africa has influenced men to seek sex with 
i 
L virgins in an attempt to avoid exposure (Silberschmidt and Rasch, 2001; Smith, 1999). In so 
doing, men (including HIV-positive men unaware of their HIV-status) may expose 
adolescent and even prel?ubescent girls to STIs through unprotected sex that may have 
damaging reproductive health consequences on their victims. 
Multiple sexual partnerships involving Ghanaian young men are associated with a 
high risk of exposure to sexually transmitted infections (STIs) especially HIV/AIDS, 
unwanted pregnancy with female counterparts, abortions among other social problems. 
However, the use of condoms, which could help to check the above problems, is not 
encouraging (MOH, 2001). 
1.2 Statement of the Problem 
In Sub-Saharan Africa, hetero-sexual intercourse with more than one partner is 
associated with higher risk of exposure to sexually transmitted infections (STIs) and acts as a 
major transmission route ofHIV/AIDS infection (Kalichman, et al, 2007). 
In Ghana, a considerable number of people suffer from sexually transmitted infections 
(STIs), and many have multiple sexual partners but do not use condoms to protect 
themselves (MOH, 2001). According to the National Sentinel survey in 2006, the prevalence 
of syphilis among sexually active men, 15-24 years was 5.5% and HIV prevalence 2.5%. 
Men's risky sexual behaviour poses a growing health burden to the population (GSS and 
Macro International Inc, 2004). The use of contraception to help fight the above problems is 
4 
University of Ghana http://ugspace.ug.edu.gh
not encouraging. In Ghana, despite almost ,universal knowledge and benefits of condom use, 
less than half of young men (46%) aged 15-24 years used condoms during their first sexual 
intercourse (GSS et aI, 2003). Potential consequences of men's uncontrolled and high risk 
sexual behaviour include impregnating school girls with possible result of tenninating girl's 
education, failed abortion or complications in abortion, single parenting, child neglect, street 
children, and child delin~ency among others. 
Not much has been achieved in terms of goals of the Ghana Family Planning Program 
(GFPP) in getting the young people especially those who are sexually active to practice safe 
sex life, control fertility and avoid contracting reproductive and sexually transmitted 
infections. Awareness has been created in the practice of safe sex, benefit and use of 
condoms, young men avoiding premarital and unprotected sex etc., but these efforts have 
not been complemented with equal involvements of men in family planning and this limits 
the effectiveness of Family Planning Programme in Ghana. 
Most reproductive health programmes aimed at ensuring people have satisfying and 
safe sex life and to overcome such problems as mentioned above are predominantly focused 
on women leaving the men who however, normatively controlled reproductive decisions in 
Ghana (Ampofo, 2000), and this is a problem. Men's dominance over reproductive decisions 
has been hypothesized to be the force delaying the onset of fertility decline in Africa 
(Caldwell et al. 1992; Knitz, 1999). 
5 
University of Ghana http://ugspace.ug.edu.gh
1.3 Rationale for the Study 
In recent times, sexuality, which includes sexual behaviour of men, has become an 
important area of research and development interest in response to concern about 
reproductive health, notably the spread ofHIV/AIDS and concern about fertility control and 
global population growth (Ulim, 1992). The study of multiple sexual partnerships by men in 
Ghana is important, timely and reasonable for a number of reasons. The youth of today are 
exposed to sex earlier than formally through the internet, where strip and sex shows, 
marketing of women and children in commercial sex have captured the attention of young 
men. The problems connected with men's risky or unsafe sexual behaviour especially that of 
the youth are of national concern. 
Male sexual behaviour have been less studied but holds the key to effective fertility 
control programs since it has implications on their female partners' sexual and reproductive 
health. Men are active players in the reproductive health context, influencing female 
decisions. Thus, they play a pivotal role in achieving fertility goals, including sound 
reproductive and sexual health and reduction in HIV transmission. Women's ability or 
inability to protect themselves against sexually transmitted infections (STIs), unwanted 
pregnancy and unwelcome sexual acts usually depend on the decisions of their male partners. 
Hence, understanding what influences men exploration of multiple sexual partnerships and 
how these affect their socio-economic and demographic development would add to the 
existing knowledge that will be useful to the youth in understanding themselves better and to 
policy makers in effective policy planning. 
The study will contribute to the general knowledge and understanding concerning 
multiple sexual partnerships and help to improve research efforts in the study of men's 
6 
University of Ghana http://ugspace.ug.edu.gh
sexual behavior in Ghana. Since the data was collected countrywide, it will provide a wide 
spectrum of base line information for the nation. Social researchers could utilize the findings 
of the study as baseline information to further delve into other related areas that will furnish 
current information on the subject which will be of great significance. 
It is also expected that this study will aid heads of institutions, community leaders, 
government agencies etc. in finding relevant answers and solutions to several questions and 
problems that involve the sexual behaviour of men in Ghanaian communities. It will also 
help the government and non-governmental organizations (NGOs) in embarking on program 
interventions on men's sexual and reproductive health issues in the communities. 
1.4 Objectives of the Study 
Main Objectives: 
The general objective of the study is to investigate the magnitude and determinants of 
multiple sexual partnerships with particular reference to men's sexual behaviour in Ghanaian 
communities. 
Specific Objectives: 
i. To estimate the levels of multiple sexual partnerships in Ghana. 
ii. To identify the relationship between selected demographic and socio-economic 
characteristics of men that influences multiple sexual partnerships. 
iii. To make suggestions and recommendation based on the findings of the study to the 
public and policy makers for informed policy formulation. 
7 
University of Ghana http://ugspace.ug.edu.gh
1.4 Literature Review 
This section reviews related studies, which contribute to the general knowledge and 
understanding of men's sexual behaviour with particular reference to multiple sexual 
partnerships in Ghanaian communities. 
A national survey in the United States of America in 2005 reported that 90% of men 
aged 18-44 years considered themselves to be heterosexual, 2.3% homosexual 1.8 bisexual 
and 3.9% others (Mosher, Chandra and Jones, 2005). 
Socio-behavioral studies in Africa have indicated that multiple sexual partnerships are 
a common occurrence in the region. A study on mUltiple sexual partners and the spread of 
IDV/AIDS in Africa by Timberg in 2007 indicates that nearly one in three or 33% of 
sexually active men in Botswana reported having multiple concurrent sex partners, as did 14 
percent of women, in a 2003 survey paid for by the U.S. government. Among men younger 
than 25 years, the rate was 44 percent. 
According to Treas and Giesen, ( 2005), infidelity by men is common. Thousands of 
women around Africa are in relationships with men who are unfaithful, and they (men) do so 
because the majority have been raised to believe that infidelity is acceptable, and in fact 
often encouraged. Infidelity, they reported, is influenced by many social and demographic 
factors. A study on 'Sexual Risk Behaviour among men with multiple, concurrent female 
sexual partner' by Chopra et al, in 2008 reported that People living in urban, informal 
settings are particularly at risk ofHIV infection (Shisana et al., 2005) as a result of men 
having high levels of sexual partner turnover and concurrency (i.e. more than one sexual 
partner during the same period of time), especially in situations where males are five or more 
years older than their female sexual partners, (Jewkes et al., 2006; WCDOH, 2001). 
According to this study, men reported a range of2-39 sexual partners in the three months 
prior to the survey, with an average of six and a median of five. 98% reported having 
concurrent sexual relationships during the same period. Most men (83.3%) indicated that 
8 
University of Ghana http://ugspace.ug.edu.gh
their friends would approve if they had sex -with women who were not their steady partners 
or wives and 86.1 % indicated that their friends would approve if they changed girlfriends 
often. 
While there are societal and cultural norms for what is considered to be appropriate in 
sexual behaviour, a man's sexual behaviour is associated with a number of socio-economic 
and demographic factors including age, marital status, education, wealth or occupation, place 
and region of residence. These socio-demographic and economic factors play some role in 
influencing men's sexual behaviour including the number of sexual partners a man has. 
Findings from a study of "Men's Exploration of Multiple Sexual Partners" by Kumar Das et 
al ( 2008) in the United States of America show that male resource attainment positively 
influences their preference for multiple partners, 
A research in Cameroon indicated that wealthy men in Cameroon are more likely to 
start sexual activity early and have both multiple concurrent and lifetime sex partners, and 
are less likely to (consistently) use a condom in sex with a non-spousal non-cohabiting 
partner than their poorest and least educated compatriots (Kongnyuy et al, 2006). These 
unsafe sexual behaviours may explain the higher HIV prevalence among wealthier men in 
the country. The work by Chopra et al stated in the opening chapter of this section reported 
that 46% of participants thought that their main partners had sex with them because they 
expected or had received any form of material goods. 
As already mentioned, most multiple sexual partnerships, many of which are 
probably concurrent, are not uncommon among sexually active men in Africa. The World 
Health Organization (WHO) in the 1990s, observed that in Africa men and women often 
have more than one-typically two or perhaps three-concurrent partnerships that can 
9 
University of Ghana http://ugspace.ug.edu.gh
overlap for months or years. For example, according to the WHO study, 18%,22% and 55% 
of men in Tanzania, Lusaka (Zambia) and Lesotho, respectively, reported having two or 
more regular, ongoing (lasting at least a year) sexual partnerships in the previous year 
According to a UNAIDS report, in 2007, on the average one man has nine premarital 
partners in rural Kenya before marriage (Abrahams, 2007). A study on multiple sexual 
partners and HIV transmission risks among people living with HIV/AIDS in Botswana 
showed that 62% of the participants were currently sexually active, of whom 80% reported 
only one sex partner in the previous 3 months and 20% reported two or more partners during 
that time (Kalichman et al, 2007). 
A recent Kenyan study in April, 2009 by Njue, et aI, discusses how the tradition of 
"disco matanga", or "disco funerals", put people at a significant risk of becoming HIV -
infected. These parties are held by the relatives of a person recently deceased, in order to 
raise funds for the funeral. The "disco funerals" are characterized by loud music, singing, 
dancing, bidding games and risky sexual behaviour. These studies are just two examples, 
drawn from an abundant amount of research that has confirmed the fact that high-risk sexual 
behaviour is prevalent in Africa, and is often supported by beliefs and traditions. These 
behaviours are subsequently far more difficult to change than if they were instead 
inconsistent with the beliefs and moral standards of the people. 
Men's access to resources (such as, higher education and prestigious occupations) is 
one of the determining factors in exploring multiple partners for a short time-span. Thus, 
while the belief system pertaining sexuality and stereotypes of women's behaviour 
encourage men to explore temporary partners, the lack of resources constrains some men's 
exploration multiple sexual partnership. 
10 
University of Ghana http://ugspace.ug.edu.gh
Men's exploration of multiple female sex partners can be measured by their 
frequency of buying sex services from prostitutes and keeping sexual relationships with 
multiple women within a short span of time. According to Farley and Hughes, buying sex 
and having multiple sex partners are closely dependent on a belief system that is operative in 
the minds of the actors, and thus, the analysis of these beliefs also reflects a social system. 
For men, both taking services from prostitutes and keeping mUltiple partners are socially 
acceptable and tolerable in several societies where the consumers of prostitution are invisible 
(Farley 2004 & 2005; Hughes 2004 & 2005; Hughes & Denisova 2002; Marianne 2004), and 
for the same reason, men's search for multiple partners is "naturalized" or "essentialized" in 
societies. Men are "allowed" to have multiple partners while women are penalized by sexual 
double standards (Mihausen and Harold, 1999). The 2003 Demographic and Health Survey 
undertaken in Ghana (GDHS, 2003) reported that 1.5% of men engaged the services of 
prostitutes twelve months before the Survey. 
Religion and culture are important factors that play some role in men's sexual 
practices. A recent work on various African cultural beliefs and traditions encourage risky 
sexual practices, which in turn increase the risk of exposure to HIV. A recent study in 
Tanzania, for example, describes the unfortunate "peer pressure" that is regularly placed on 
the youth to have multiple concurrent sexual partners, with men who limit themselves to just 
one partner being ridiculed, and called "domo zenge", meaning "slow to move". On the 
contrary, men who succeed in having concurrent sexual relationships with several women 
are commended, and referred to as either "mshua" (the connoisseur) or "kichwa kikali" 
(the gifted) (Rweyemamu, 2007). 
11 
University of Ghana http://ugspace.ug.edu.gh
In another study on multiple sexual 'partners in Ghana, Anarfi, (1999) reported that 
about 80 per cent of the males have had more than one sexual partner in their lifetime. 
According to this study, premarital sex is a common practice in all communities surveyed. 
About 25 per cent of the males respondents have multiple partners currently (Anarfi, 1999). 
Analysis of 1998 Ghana demographic and Health Survey (GDHS) showed that men's sexual 
practices vary with age, in a normal distribution format, increasing from 7.8% in the 15-19 age 
group, peaking at 40-44 age group and decreasing to 49.0% in the 55-59 age group, for sexually 
active in last four weeks before the survey. 
Both national and local area studies have shown that men's sexual behaviour varies 
across the regions of Ghana In a study of 1,782 unmarried young people aged 15-19 years 
in the Greater Accra and Eastern Regions, it was found that 67% of males had ever had sex, 
and the mean age at sexual debut among those who had ever had sexual intercourse was 
15.5 years for males (Nabila et al ,1996). In a study of 1,415 males and females aged 10--19 
years in Ketu South, upper Denkyira and Offmso electoral constituencies, it was observed 
that the median age of fIrst sexual intercourse for males aged 10--19 years in these three areas 
in Ghana was 16years. In a survey of 1,038 students (567males and 471 females) aged 13-
18 years in nine senior secondary schools, 50% of the adolescents considered chastity as an 
ideal thing to attain and thought that it was realistically attainable. Yet 42% of the male 
students surveyed had had sexual intercourse. The average age at fIrst sexual experience was 
15years. Half of the sexually experienced students had their fIrst sexual experience between 
ages 14 and 17years, and 25% had their fIrst sexual experience at age 13 or younger. 
Reasons given for engaging in sexual intercourse included pressure from peers, deception by 
partners, experimentation, and satisfaction of sexual desires (Awusabo-Asare et al, 2004). 
12 
University of Ghana http://ugspace.ug.edu.gh
By age 15years, 47% of the males had ever had sex. Among adolescents aged 
12- 20years, studied in Kumasi and Accra, it was observed that the median age at first sexual 
experience was 16 for both boys and girls in the sample (Nabila et al ,1996). However, a 
ccording Djamba, 2004 evidence from studies in Africa suggest a declining age of first 
sexual debut, but increasing numbers of sexually active adolescents, and high-risk sexual 
behaviour (use of multiple sexual partners, commercial sex, and poor use of contraceptives) 
(Djamba, 2004). For example, the median age at fIrst sex and median age at fIrst marriage, 
for Ghanaian men according to GDHS, 2003 were respectively 20.2 years and 24.7 years. 
The period between age at fIrst sex and age at fIrst marriage can be a period of sex 
experimentations 
In an era ofHIV infection, the use of the condom as protection against STIs in 
addition to its use as a family planning method has become important. As with other modem 
contraceptives, adolescents' awareness of the male condom is high, but the vast majority of 
sexually active adolescents are not using condoms. Results from the 1998 Ghana Youth 
Reproductive Health Survey (GYRHS) indicated that 88% of males aged 15-19 years were 
aware of at least one method of contraception, but the number of males who were currently 
sexually active in this group, only 37% used modem methods. 
According to a Multiple Indicator Cluster survey, 2006, the percentage of men who 
used a condom at last sex with non-marital, non-cohabiting partner was 53.5%. However, 
this varied across the wealth quintals and education levels of respondents with 33.9% for the 
poorest wealth quintiles and 57.7% for the richest while condom use was 30.9% for men 
with no education and 60.2% for men with at least secondary education (MICS, 2006). 
13 
University of Ghana http://ugspace.ug.edu.gh
One reason for the low level of condom use is that young people do not feel 
confident in insisting on condom use in a relationship. In the 1998 GYRHS, among those 
who had heard of condoms, 27% of males said they could not insist on using a condom if 
their girlfriend did not want to use one. Moreover, about one-third of male adolescents said 
they could not refuse to have sex if their girlfriend did not want to use a condom. The 
confidence that young people have to use condoms, an effective means of preventing the 
spread ofHIV, is still far too low (Adih, and Alexander, 1999). 
1.6 Conceptual Framework 
The Conceptual Framework is shown in figure 1.1. It links the respondent's socio-
demographic characteristics and sexual behaviour, operationalized here as men with two or 
more sexual partners and those with single sexual partners in the last 12 months. In this 
model, socio-demographic characteristics like age, marital status, religion, place of 
residence, region, occupation, wealth, age at first sex and education that are considered to 
influence sexual behaviour are categorized as independent variable. 
Each of the background characteristics listed below as independent variable in the 
conceptual framework exerts an influence on sexual partnerships. For example it is expected, 
considering review of literature that men's occupation, wealth and education as well as urban 
residence will positively affects multiple sexual partnerships. It is also anticipated that age at 
first sexual intercourse will have an inverse relationship with multiple sexual partnerships, 
while currently married men or living together will keep single sexual partners compared to 
unmarried men. The Conceptual Framework will seek to explore how religion and region of 
residence will affect sexual partnerships in Ghana. 
14 
University of Ghana http://ugspace.ug.edu.gh
The categories of independent variables are, 
Age Group: 15-19,20-24,25-29,30-34,35-39,40-44,45-49,50-54,55-59. 
Marital status: Never married, Married/living together, Divorced or Separated and 
widowed. 
Place of residence: Urban or Rural. 
Region: All the ten regions of Ghana. 
Education level: No education, Primary, Secondary, and Higher education 
Religion: Christianity, Moslem, Traditional, and other Religions 
Occupational Status: Professional/Technical & Managerial, Clerical, Sales and services, 
agriculture and self employed, skilled manual, unskilled manual and no work. 
Age at First Sex: < 15, 15-19,20-24, and 25+ 
Wealth Index: lowest (poorest), Second, Middle, Fourth and highest quintile. 
The dependent variables is 
Multiple sexual partners (the percentage of sexually active men who had two or more 
sexual partners) or Single sexual partners (percentage of sexually active men who had one 
sexual partner in last 12 months before the survey). 
15 
University of Ghana http://ugspace.ug.edu.gh
Figure 1.1: CONCEPTUAL FRAMEWORK 
Indeperd91t 
Varicbles 
Oeperd91t 
Age Varicble 
Marit al stat us 
PlaceofR eside nee 
Religion 
Multi pie 
Level 0 f E due ati on ... sexu al p art ne IS 
Age at Ii rst sex .? Single 
sexu aI P art ne rs 
Occupati on forthepast 12 months 
Region 
Wealth st atus 
Source: Author's construct 
1.7 Hypothesis 
The study will test the following hypothesis: 
i. Men in urban residence are more likely to engage in multiple sexual partnerships than men 
residing in rural communities. 
ii. Men with professional/technical/managerial occupations are more likely to engage in 
multiple sexual partnerships than those with no occupation. 
iii. Men who had their first sex at earlier ages, e.g. before attaining age 15 years are more 
likely to have multiple sexual partners than those who have their first sex at older ages, e.g. 
when they are 25 years or more. 
16 
University of Ghana http://ugspace.ug.edu.gh
1.8 Methodology 
1.8.1 Source of Data 
The main source of data used for the study is the 2003 Ghana Demographic and 
Health Survey (GDHS, 2003). This was undertaken by the Ghana Statistical Service (GSS) 
in collaboration with Noguchi Memorial for Medical research (NMMR) and under the 
sponsorship of, USAID and Ghana Health service (GHS). A national representative 
household survey with a sample size of 5,015 men within the age group 15-59 years from a 
totai of6,251 households were selected in all the ten regions of Ghana 77.3% or 3,777 out 
of this sample of men were sexually active 12 months before the survey. This latter sample 
size (3,777) is the focus of this study. 
The primary objectives were to provide current and reliable data on fertility, 
marriage, sexual activity, family planning behaviour, mortality patterns among others. 
1.8.2 Limitation of Data 
The 2003 Ghana Demographic and Health Survey (GDHS, 2003) did not collect data 
purposely on men's sexual behaviour. Hence this study is limited in scope. Another problem 
is that the data is almost out of date that it will less provide the most current evidence of 
men's sexual behaviour in Ghanaian communities. These limitations may affect the final 
interpretation of the results of this study. 
17 
University of Ghana http://ugspace.ug.edu.gh
1.8.3 Method of Analysis 
Both descriptive and analytical methods will be applied in this study. First, bivariate 
analysis using simple frequency, ratios & rates, cross tabulation of variables will be 
employed to explain the basic relationships between these independent variables and sexual 
behaviour. Secondly, Logistic regression technique will be used to obtain degree of 
relationships between each of the selected demographic and socio-economic variables and 
the dependent variable. 
In the logistic regression model, the log of the odds ratio or exponential f3 (which in 
this study is the ratio of the proportion of men engaging in multiple sexual partnership to 
those having single sexual partners) is expressed as a function of the various independent 
variables (i.e. predictors). A positive f3 shows the likelihood increase in multiple sexual 
partnerships in relation to the reference category (RC) while a negative f3 indicates the 
likelihood reduction in multiple sexual partnerships in relation to the reference category. 
A zero f3 coefficient on the other hand indicates no change in relation to the reference 
category. 
Mathematically, it is expressed as follows: 
Or ~=ez 
1- p 
Where f3n are the logistics coefficients and Xn are the explanatory variables. 
P = probability of success and e = base of natural logarithm, approximately 2.7183. 
18 
University of Ghana http://ugspace.ug.edu.gh
The logistic regression model is fitted for all men between 15 and 59 years to determine the 
factors that influence their sexual behaviour (single and multiple sexual partnerships). The 
model estimates the probability that a man will have at a given time one or more than one 
sexual partners. 
In the logistic regression analysis, the demographic and socio-economic factors are 
the independent variables while the dependent variable is sexual partnerships (i.e. single or 
multiple sexual partners). The statistical Package for Social Sciences (SPSS) was used for 
the imalysis. 
1.9 Definition of Concepts 
Multiple Sexual Partnerships 
This term will be used to define a situation where male-female partnerships overlap in time, 
either where two more partnerships continue over the same time period, or where one 
partnership begins before the other terminates. A sexual partnership is considered to be 
concurrent in surveys if man reports having two or more sexual partners in a month. 
Male sexual behaviour: 
In this study, male sexual behaviour is simply used to refer to how men express themselves 
sexually as human males. 
19 
University of Ghana http://ugspace.ug.edu.gh
SexuaUy Active men: 
In this study, sexually active men will be used to refer to all men age 15 to 59 years who had 
sex twelve months before the survey. These men are conceived to have the potential to 
engage in sex at some regular time intervals. 
Risky Sexual behaviour: 
In this study, risky sexual behaviours will be used to refer to unprotected vaginal, oral, or 
anal intercourse with non-marital, non-cohabiting partner which takes forms ranging from 
two sexual partners to a large number of sexual partners. 
1.10 Plan of the Study 
The study will be organized into five chapters. The first chapter which is the 
introductory chapter focuses on the Background to the Study, Problem Statement, Rationale 
of the Study, Objectives, and Literature review, Conceptual Framework, Hypothesis and 
Methodology. Chapter 2 examines the Demographic and Social characteristics of the study 
population, while chapter 3 looks at the Patterns and Differentials in men's sexual 
partnerships. 
Chapter 4 takes a look at the determinants of multiple sexual partnerships through the 
analyses of the relationships between selected socio-economic and demographic 
characteristics of men and sexual partnerships while Chapter 5 will comprise the summary, 
conclusion and recommendation. 
20 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER TWO 
EVALUATION OF DATA AND ANALYSIS OF DEMOGRAPmC AND SOCIAL 
CHARACTERISTICS OF RESPONDENTS 
2.0 Introduction 
The purpose of this chapter is to evaluate data and give a descriptive analysis of some 
demographic characteristics of the respondents. Socio-demographic characteristics of a 
popUlation have a bearing on sexual behaviour. Hence, background characteristics that may 
playa key role in the analysis of men's sexual behaviour are described in this chapter. 
2.1 Evaluation of Data 
Age is an important demographic variable. The age structure is considered basic to 
the study of population problems. However age data are not free from errors especially in 
sub- Saharan African countries. The 2003 GDHS defines age in terms of completed years 
that is age at last birthday. The errors in age data are of two types: coverage error and content 
error. Coverage error occurs as a result of omission or erroneous inclusion of individual at 
certain ages. Content errors are errors, which result from either the erroneous reporting of 
age by the respondent or erroneous estimation of age by the interviewer. 
Table 2.1 evaluates age data of respondents using United Nations Age Ratios. Age 
Ratio (AR) is the ratio of the population in a given age group to the average of the population 
in the two adjacent age groups. In the absence of any disturbing factor, the age ratio will be 
approximately 100%. Deviations from 100 imply that data is subject to errors. 
21 
University of Ghana http://ugspace.ug.edu.gh
AR == Px x 100 
The Age Ratio is expressed as 7I2I  (p + ) x-5 Px+5 
Where 
Px is the population at age group x 
Px-5 and Px+5 are the populations at the two adjacent age groups x-5 and x+5 before and after 
age group x respectively. 
Table 2.1: Evaluation of Age data, Using United Nations Age Ratios 
Age Population Age Ratio Deviation from 100 
15-19 1,095 - -
20-24 692 75.96 -24.04 
25-29 727 109.74 9.74 
30-34 633 101.69 1.69 
35-39 518 99.23 -0.77 
40-44 411 85.71 -14.29 
45-49 441 124.05 24.05 
50-54 300 93.90 -6.10 
55-59 198 7.45 -92.55 
Total 5,015 697.73 1173.221 
Source: Computed from GDHS data set 
The table indicates major age misreporting at age groups 20-24, 45-49 and 55-59. 
The errors are also measured by Age Ratio Score (ARS) which is an index derived from Age 
Ratios given by the average of the absolute deviations as follows: 
IIARi -100 II 
ASR == ='---N-~ 
22 
University of Ghana http://ugspace.ug.edu.gh
Where N is the number of deviations from toO. 
If AFR = 0 then the age data are free from errors. However, from Table 2.1, 
ASR = 173.22 = 21.65 
8 
Since 21.65 is far from 0, GDHS data are subject to errors and must be used bearing this 
situation in mind. 
2.2 Demographic and Socio-economic Characteristics of Respondents 
2.2.1 Age Distribution 
The study population is made up of 3,777 sexually active men aged 15-59 years .. 
The distribution of the sexually active respondents by five- year age groups is shown in 
Table 2.2. 
Conventional five- year age groups of respondents can minimize errors in single 
year's ages. Five years age groups tend to smoothen age shifting and this enables the 
researcher make a fair analysis of data. The distribution shows that about half (51.2%) of the 
respondents are under age 35 years while the proportions in each of the youngest and oldest 
age groups are about 5%. 
23 
University of Ghana http://ugspace.ug.edu.gh
Table 2.2: Percentage distributions of Respondents 
b) y A~ e:e e:roup 
Age 
Group Frequency Percentage 
15-19 192 5.1 
20-24 479 12.7 
25-29 654 17.3 
30-34 619 16.4 
35-39 499 13.2 
40-44 407 10.8 
45-49 437 11.6 
50-54 295 7.8 
55-59 195 5.2 
Total 3,777 100.0 
Source: Generated from 2003 GDHS data set 
2.2.2 Current Marital Status 
Marital status is generally considered to be indicators of exposure to sexual activity. 
The 2003 GDHS classified mar ital status as: Never married, Married, Living Together, 
Divorced, Separated or Widowed. For the purpose of this study, marital status is grouped 
into three categories as shown in Table 2.2. 
The distribution of Respondents by Current Marital Status shows that about 52.6% 
are married or living together, while 40.0% are never married. The rest (7.4%) are divorced, 
separated, or widowed. 
24 
University of Ghana http://ugspace.ug.edu.gh
Table 2.3: Percentage distributions of Respondents by Current 
Marital Status 
Marital Status Frequency Percentage 
Never Married 1,510 40.0 
Married! Living Together 1,986 52.6 
Separated! Divorced! Widowed 281 7.4 
Total 3,777 100.0 
Source: Generated from GDHS, 2003 
2.2.3 Educational Status of Respondents 
Table 2.3 shows the Percentage Distribution of educational status of respondents by 
age group. Education affects man's behaviour and for that matter the sexual activity of men. 
The formal education system has undergone several restructuring over the past 25 years 
(Sedjwick, 2000). The 2003 GDHS grouped the current educational system as: No education, 
primary, secondary and Higher education. 
j 
The distribution shows that about 53% of respondents had completed secondary 
school, while 7.4% attained higher education and 25.4% had no formal education. 
25 
University of Ghana http://ugspace.ug.edu.gh
Table 2.4 Percentage distributions of Respondents 
b)y  Ed ucat.i onallevel 
Educational level Frequency Percent 
No education 958 25.4 
Primary 531 14.1 
Secondary 2,008 53.2 
Higher 280 7.4 
Total 3,777 100.0 
Source: Generated from 2003 GDHS data set 
2.2.4 Region and Place of Residence 
Table 2.5a displays the percentage distribution of Respondents by Region. 
Ghana covers an area of238,537 knlwith an average of 79.3 people per km2• The country is 
divided into ten major regions, which have different popUlation densities. Almost every 
region has different ethnic groups and differs also in infrastructure, communication and 
occupational activities among others. These variations may have some influence on the 
general behaviour of the people including their sexual activity. Regional distribution shows 
that majority of the respondents are in Ashanti (15 .9%) followed by Northern (12.9) and 
Greater Accra (12.7%), while the least proportions of respondents are in Upper West, Central 
and Upper East regions. 
26 
University of Ghana http://ugspace.ug.edu.gh
Table 2.5a: Percentage distribution of Respondents 
b) y R e2i. on 
Region Frequency Percentage 
Western 344 9.1 
Central 220 5.8 
Greater Accra 481 12.7 
Volta 294 7.8 
Eastern 364 9.6 
. Ashanti 600 15.9 
Brong Ahafo 434 11 .5 
Northern 489 12.9 
Upper West 268 7.1 
Upper East 283 7.5 
Total 3,777 100.0 
Source: Generated from 2003 GDHS data set. 
Also differences in people's lifestyles and behaviour depend upon whether one is 
brought up in rural or urban environment. The GDHS, 2003 categorized towns with more 
than 5,000 people as urban and localities with less than 5,000 people as rural. 
Table 2.5b indicates that majority of respondents (about 63%) are in rural 
communities while 37% resides in urban localities. 
27 
University of Ghana http://ugspace.ug.edu.gh
Table 2.5b: Percentage distribution of Respondents, 
By type of Residence 
Type of Residence Population Percentage 
Urban 1,402 37.1 
Rural 2,375 62.9 
Total 3,777 100.0 
Source: Generated from 2003 GDHS data set 
2.2.5 Wealth Status of Respondents 
Research has shown that a man's level of wealth has some influence on his sexual 
behaviour (Kongnyuy, 2006). The 2003 GDHS classified wealth into five categories, using an 
index based on wealth scores of households. The wealth index is assumed to capture the 
underlying long term wealth index through information on household assets and it intended to 
produce a ranking of households by wealth from poorest to richest (Rutstein and Johnson, 2004). 
Table 2.5 shows that the proportion of respondents by wealth status varies from 
17.8% for the middle wealth quintal to 23.6% for the poorest wealth quintal. 
Table 2.6: Percentage distribution of Respondents 
b, y Weiat h Status  
Wealth Status Population Percentage 
Poorest 893 23.6 
poorer 727 19.2 
middle 670 17.8 
richer 676 17.9 
richest 811 21.5 
Total 3,777 100.00 
Source: Generated from GDHS, 2003 data file 
28 
University of Ghana http://ugspace.ug.edu.gh
2.2.6 Occupational Status of Respondents 
Table 2.6 shows the percentage distribution of respondents by type of occupation. 
Men engaging in agricultural activities constitutes the majority of respondents (52.5%) 
followed by skilled and unskilled manual workers (19.7%). A proportion of 10.9% and 9.4% 
are in the Professionalffechnicall Managerial& Clerical and sales & services respectively 
while about 7.5% are made up of men without any specific work. 
Table 2.7: Percentage distribution of Respondents 
by Type of Occupation 
Type of Occupation Frequency Percentage 
Not working 285 7.5 
Professionalffe chnicall Managerial 
& Clerical 413 10.9 
Sales and services 350 9.4 
Agriculture 1,984 52.5 
Skilled manual & Unskilled manual 745 19.7 
Total 3,777 100.0 
Source: Generated from GDHS, 2003 data file 
2.2.7 Religious AOlliation of Respondents 
People's religious beliefs and norms have some impact on their general behaviour, 
depending on the specific doctrine of their religious affiliation which is likely to influence 
their sexual behaviour including multiple sexual partnerships. Table 2.7 shows the 
distribution of respondents by their religious affiliation. 
29 
University of Ghana http://ugspace.ug.edu.gh
It is evident from Table 2.7 that the Christian faith is the major religion of respondents. 
The survey grouped the Christian religion into: Roman Catholic, Anglican, Methodist, 
Presbyterian and other Christians which together form about 64.5% of the various religious 
groups. About 21% of the respondents were Moslems while 6.9% are from traditional and 
religion. 
Table 2.8: Distribution of Respondents by 
Re rIg·lO US Aft. .. rIatl·O n 
Type of 
Religion Frequency Percent 
No religion 7.8 
296 
Moslem 20.7 
782 
Christian 64.5 
2,439 
Other Christians 0.1 
2 
Traditionalist 6.9 
260 
Total 100.0 3,777 
Source: Generated from GDHS, 2003 data fi le 
30 
University of Ghana http://ugspace.ug.edu.gh
2.2.8 Age at First Sex 
The onset of sexual activity is an important determinant of subsequent sexual 
behaviour and the risk of exposure to STIs including HIV/AIDS. 
Table 2.8 shows the percentage distributions of respondents by age at first sexual 
intercourse. 
The distribution shows that by age 19 years, more than half (52.9%) of the 
respOndents has had their first sexual intercourse, about 13 % initiated sex at 25 years and 
above while 5.1% had their first sexual intercourse before age 15 years. 
Table 2.8: Percentage distribution of Respondents. 
B5 y A !!e a tfiI rs t sex 
Age at first sex Frequency Percent 
< 15 191 5.1 
15-19 1,806 47.8 
20-24 1,280 33.9 
25+ 500 13.2 
Total 3,777 100.0 
Source: Generated from GDHS, 2003 data file 
31 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE 
PATTERNS AND DIFFERENTIALS IN SEXUAL PARTNERSIllPS 
3.0 Introduction 
Men's sexual behaviour varies with demographic and socio-economic characteristics. 
The study of such variations among sub groups of a population is an important aspect of 
demographic research. Socio-economic factors influencing the number of female sexual 
part:ners a man has, according to Kumar Das et al (2008) include male difference in resource 
attainment, views about sexual behaviour, and beliefs in stereotypes about "expected" for 
women. Findings from their studies indicated that male resource attainment positively 
influences the men's preference for multiple female sexual part:ners. 
The fmdings from such studies provide the bases for formulation of policies or 
drawing of efficient strategies aim at urging adolescents to make responsible choices that 
will not compromise their sexual and reproductive health. This section focuses on some 
selected background characteristics of respondents and their influence on the number of 
sexual partners they have. 
3.1 Marital Status and Sexual Partnerships 
Marriage is an important variable that determines the age at which a man is expected 
to first experience sexual intercourse. Multiple sexual part:nerships are a common 
phenomenon in African country like Ghana ( Wiederman, 1997). Whether a man is currently 
married, formally married or never married affects his sexual behaviour and for that matter 
32 
University of Ghana http://ugspace.ug.edu.gh
the number of sexual partners he has. Multiple sexual partnerships exist among currently 
married, never married or formally married men in communities in Ghana (Anarfi, 1999). 
Table 3.1 summarizes the pattern of sexual partnerships according to respondents' 
marital status. For the purpose ofthls analysis, widowed, divorced and separated or not 
living together were put under one category while currently married and living together were 
put under one group. The proportion of men having more than two sexual partners ranges 
from 11.4% for men currently married/living together to 19.7% for the never married. 
According to Table 3.1, the widowed, divorced or separated have 18.5 percent likelihood of 
engaging in multiple sexual partnerships. 
On the other hand, men married or living together with partners are the most likely 
to maintain single partner relationships as compared to all other categories of men while men 
widowed, divorced or separated have the least likelihood of keeping single partners. This is 
expected for married men since they are always close to their normal sexual partners and as 
such are not exposed to the temptation of exploring multiple sexual partnerships. On 
average, about 85% of men are having single sexual partners. 
33 
University of Ghana http://ugspace.ug.edu.gh
Table 3 .1.· Differentials in Sex uaI  P a rtn ersh ·Ip S b) , M an.t aI  status 
Total of 
Single Partner Multiple Partners sexually 
Marital Status active men 
Number Percent Number Percent Number Percent 
% % % 
Never married 1,221 80.3 298 19.7 1,510 100 
Married! Living together 1,751 88.6 228 11.4 1,986 100 
Widowed! Divorced 
Or separated 227 81.5 52 18.5 281 100 
Total 3,199 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
3.2 Age and Sexual Partnerships 
Age is one of the important demographic variables that are associated with sexual 
behaviour. People sexual preferences typically change over time, reflecting sexual 
experience. Age of an individual influences his behaviour and for that matter his sexual 
activity. Table 3.2 enumerates the pattern and differences in sexual partnerships by the age of 
respondents. Multiple sexual partnerships vary in a decreasing pattern from age 20-24 years 
(18.6%) to oldest age group 45-49 years (about 12%). Men within the age group 20-24 years 
have the highest proportions of men engaging in multiple sexual partnerships compared to 
those in other age groups. This situation is expected since men of this group are mostly 
unmarried and sexually active. However, 16.6% of men in the youngest age group and about 
13% of those in the oldest age groups have more than one sexual partner. 
For single partner relationships, men in older age groups (e.g. 35 years and above) 
are more likely to maintain one sexual partner than those in younger age groups. For 
34 
University of Ghana http://ugspace.ug.edu.gh
example about 87.8% of men within the 45 .. 49 age group have single sexual partners 
compared with 81.4% for men in the 20-24 age group. 
Table 3 .2.' Differentials in S exuaI Pa  rtne rsh I' pS b)Y A,geG r oups 
Total of 
Single Partner Multiple Partners sexually 
Age Groups active men 
Number Percent Number Percent Number Percent 
% % % 
15-19 160 83.4 32 16.6 192 100 
20-24 390 81.4 89 18.6 479 100 
25-29 538 82.3 116 17.7 654 100 
30-34 521 84.1 98 15.9 619 100 
35-39 425 85.2 74 14.8 499 100 
40-44 356 87.4 51 12.6 407 100 
45-49 384 87.8 53 12.2 437 100 
50-54 255 86.3 40 13.7 295 100 
55-59 170 87.1 25 12.9 195 100 
Total 3,199 84.8 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
3.3 Educational status and Sexual partnerships. 
Education contributes to men's socio-economic position. Educational status is an 
important socio-economic variable that play important role in the lives of men in regard to 
acquisition of skills and knowledge, enhancement of economic and secular values as well as 
morals that transforms norm and beliefs that shapes attitudes in life. 
35 
University of Ghana http://ugspace.ug.edu.gh
Table 3.3 displays the pattern of sexual partnerships by educational level of 
respondents. It is evident that men with a higher level of education has the highest proportion 
of their group engaging in multiple sexual partnerships compared to men without any formal 
education. The tendency of men to have more than one sexual partner increases from 14% 
for those with no formal education to about 22% for men with higher level of education. 
Furthermore, about 15% of men with primary education and those who have attained 
secondary level of education are engaged in multiple sexual partnerships. However, the 
reverse pattern exists in the case of single partner relationships where men with no formal 
education have a higher proportion (86%) having single sexual partner, compared with about 
78% for men who has completed higher education. 
Table 3.3: Pattern of sexual partnerships by Educational Attainment 
Total of 
Single Partner Multiple Partners sexually 
Educational Attainment active men 
Number Percent Number Percent Number Percent 
% % % 
No education 824 86.0 134 14.0 958 100 
Primary 451 84.9 80 15.1 531 100 
Secondary 1,706 85.0 302 15.0 2,008 100 
Higher 218 77.9 62 22.1 280 100 
Total 3,199 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
36 
University of Ghana http://ugspace.ug.edu.gh
3.4 Religious AffUiation and Sexual p,artnerships 
People's religious beliefs may have some considerable impact on their sexual and 
reproductive behaviour, depending on the specific doctrine of their religion which is likely to 
influence their involvement in multiple sexual partnerships. For the purpose of this study, 
Roman Catholic, Anglican, Methodist, Presbyterian and other Christian sects from the 
GDHS, 2003 raw data were put under Christianity. It is worth noting that, data were not 
obtained on the religiosity of respondents (i.e. how religious or how they strongly or 
fervently practice their faith) and this can affect the outcome and interpretation of results as 
presented. 
Table 3.4 indicates that men affiliated to the traditional religion are more likely to 
engage in multiple sexual partnerships than members from any other religious group. 
However, Moslems with a proportion of 14.6%, are less likely to involve themselves in this 
practice compared with Christians (15.5%). For some reasons, men with no specific religion 
have the least proportion (14.2%) of its members having more than one sexual partner. It 
appears the pattern of sexual partnerships does not vary too much among the various 
religious groups. As regards single partner relationships, the proportion ranges from 83% for 
men affiliated to traditional religion to 85.8% for those with no specific religion. The highest 
proportion of multiple sexual partnerships for traditionalist could be due to their likelihood 
of adhering to traditional beliefs and norms that promote large family sizes. 
37 
University of Ghana http://ugspace.ug.edu.gh
Table 3.4: Pattern of sexual partnerships by Religious Aftlliation 
Total of 
Single Partner Multiple Partners sexuaUy 
Religious AftUiation active men 
Number Percent Number Percent Number Percent 
% % % 
No religion 254 85.8 42 14.2 296 100 
Christian 2,061 84.5 378 15.5 2,439 100 
Moslem 668 85.4 114 14.6 782 100 
Traditionalist 216 83.0 44 17.0 260 100 
Total 3,199 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
3.5 Wealth Status and Sexual partnerships 
Researchers in Africa hold the view that men's resources or roughly their socio-
economic positions (e.g. wealth and education), for example, may provide more avenues for 
exploring multiple sexual partners, because they have resources useful for such exploration 
(Kimuna and Djamba, 2005). According to Kongnyuy et ai, 2006, the socio-economic status 
of a man has an association with the number of sexual partners he has. 
Table 3.5 summarizes the pattern of sexual partnerships by the wealth status of 
respondents. With the exception of those in the lowest level of the wealth rankings where 
about 14% of men have more than one sexual partner, the proportion of men having multiple 
sexual partners increases from 11.4%  (the poorer) to about 19% in the richest. 
38 
University of Ghana http://ugspace.ug.edu.gh
Table 3 .5 .' Pattern o f sexuaI.  par tn ersh I' PS b, y W ealth Status 
Total of 
Single Partner Multiple Partners sexually 
Wealth Status active men 
Number Percent Number Percent Number Percent 
0/0 % % 
Poorest 767 86.0 126 14.1 893 100 
Poorer 644 88.7 83 11.4 727 100 
Middle 563 84.3 107 16.0 670 100 
Richer 56 84.4 108 16.0 676 100 
Richest 657 81.3 154 19.0 811 100 
Total 3,199 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
3.6 Region of Residence and Sexual Partnerships 
As mentioned in the previous chapter, Ghana is divided into ten major regions, which 
have different population densities and also differ in resources. Some of the regions also 
share borders with neighbouring countries that may influence the culture of the people in 
these regions. These differentials have some bearing on the general attitude and behaviour of 
the people including whether or not a man has multiple sexual partners. For example, the 
2003 GDHS reported that men who engaged the services of prostitutes 12 months before the 
survey was highest in the Western Region (3.6%), which shares border with Cote d'Ivoire, 
compared with Greater Accra (1.2%) and Upper East (0.2%). 
Table 3.6 displays the differences in sexual partnerships for the ten regions. The 
region with the highest density of people, Greater Accra has the greatest proportion (21.4%) 
of men engaging in multiple sexual partnerships followed by Volta (19.4%) while Upper 
West, one of the less dense regions in the country has the least proportion (8.6%) of men 
39 
University of Ghana http://ugspace.ug.edu.gh
engaging in this practice. The high proportion of multiple sexual partnerships in Greater 
Accra can be attributed to the fact that the region is the most urbanized in the country, an 
area of high immigration and as such predisposed to high sexual networking. The low 
proportions as regard this sexual behaviour in the Upper West and Upper East is 
understandable since these regions are among the least urbanized in the country and also 
areas of high emigration of young men. 
However, men residing in the Brong Ahafo or Northern regions have similar 
proportions (16.6% and 16.7% respectively) of men having more than one sexual partner. 
Table 3.6: Regional Pattern in Sexual Partnerships 
Total of 
Single Partner Multiple Partners sexually 
active men 
Region Number Percent Number Percent Number Percent 
0/0 % % 
Western 303 88.1 41 11.9 344 100 
Central 181 82.3 39 17.7 220 100 
Greater Accra 378 78.6 103 21.4 481 100 
Volta 237 80.6 57 19.4 294 100 
Eastern 310 85.2 54 14.8 364 100 
Ashanti 519 86.5 81 13.5 600 100 
BrongAhafo 362 83.4 72 16.6 434 100 
Northern 407 83.3 82 16.7 489 100 
Upper west 245 91.4 23 8.6 268 100 
Upper east 257 90.8 26 9.2 283 100 
Total 3,199 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
40 
University of Ghana http://ugspace.ug.edu.gh
3.7 Sexual Partnerships by Place of Residence 
The type of environment, (rural or urban) in which a person is brought up has some 
influence on his moral behaviour and attitude towards sex. Review of literature shows that 
men who live in urban, informal settings who have concurrent, female sexual partners 
constitute a high-risk population that could easily contract HIV/AIDS. 
Table 3.7 summarizes multiple sexual partnerships by type oflocality. It is evident 
that men in urban communities are more likely to engage in multiple sexual partnerships than 
rural dwellers. For sexually active men living in urban communities there are about 18% of 
them having more than one sexual partner. On the other hand, men residing in rural 
communities most likely to maintain single partner relationships. 
Table 3.7: Pattern of sexual partnerships by Place of Residence 
Total of 
Single Partner Multiple Partners sexually 
Type of residence active men 
Number Percent Number Percent Number Percent 
% 010 0/0 
Urban 1,155 82.0 252 18.0 1,402 100 
Rural 2,054 86.3 326 13.7 2,375 100 
Total 3,199 84.8 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
41 
University of Ghana http://ugspace.ug.edu.gh
3.8 Age at First Sex and Sexual partnerships 
The onset of sexual activity typically takes place during adolescence, a period of 
growth, experimentation and identity search during which individuals are particularly 
vulnerable and in many cases ill informed with respect to making responsible choices that 
will compromise their sexual and reproductive health (Zabin and Kiragin, 1998). Age at First 
Sexual intercourse influences the number of sexual partners a man has (Zaba et al, 2004). 
The median age at first sexual intercourse for men generated from the 2003 survey was 18 
years. 
Table 3.7 outline the pattern of multiple sexual partnerships by age at fIrst sex. A 
signifIcant proportion (31 %) of men who initiates sexual activity by age 14 years are 
engaging in multiple sexual partnerships. It is apparent from table 3.7 that men who started 
sexual intercourse at earlier ages e.g. less than 15 years are more likely to have multiple 
sexual partners than men who initiates sexual activity at later ages e.g. 25 years and above, 
where less than ten percent of men within that category have more than one partner. On the 
other, men who initiate sexual activity at ages 15-19 years have 17.5% of men engaging in 
this sexual behaviour. 
42 
University of Ghana http://ugspace.ug.edu.gh
Table 3 .8 .' Patter n 0 f sexuaI  par tne rsh ·I PS b'Y  A1 ge at FI· rst Se x 
Total of 
Single Partner Multiple Partners sexually 
Age at First Sex active men 
Number Percent Number Percent Number Percent 
0/0 % 0/0 
< 15 132 69.1 59 30.9 191 100 
15-19 1,490 82.5 316 17.5 1,806 100 
20-24 1,124 87.8 156 12.2 1,280 100 
25+ 453 90.6 47 9.4 500 100 
Total 3,194 84.7 578 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
3.9 Occupational Status and Sexual partnerships 
Employment status enhances man's economic value such that the higher a man is in 
gainful employment the higher it may have an influence on his behaviour. Occupation is 
considered to exert some influence on sexual network. Foe example a study conducted in 
Ghana in 1991 showed that long distant itinerant soldiers, policemen, teachers and drivers 
are at risk of contracting STIs including HIV / AIDS due to high rate of sexual partner change 
(Anarfi,1991). 
Work or economic status is an important variable that may affect a man's sexual 
behaviour. The type of occupation a man engages in has a bearing on his sexual activity. For 
the purpose of this study occupation has been classified into five categories and the pattern 
of sexual partnerships by each type of group is shown in table 3.8. 
43 
University of Ghana http://ugspace.ug.edu.gh
Table 3.9: Pattern of sexual partnerships by Occupational Status of respondent 
Total of 
Single Partner Multiple Partners sexually 
Occupational Status active men 
Number Percent Number Percent Number Percent 
% % % 
Not working 228 79.9 57 20.1 285 100 
Professional, tech., 
managerial & clerical. 337 81.5 76 18.5 413 100 
Sales and services 292 83 .5 58 16.5 350 100 
Agriculture 1,728 87.1 256 12.9 1,964 100 
Skilled & unskilled manual 621 83.4 124 16.6 745 100 
Total 3,206 84.7 571 15.3 3,777 100 
Source: Generated from GDHS, 2003 data file 
The proportion of men in various work categories who have at least two sexual 
partner ranges from about 13% in the agriculture sector to 20.1 % for those without any 
specific employment. Skilled & unskilled manual workers are equally as likely as sales and 
service workers to engage in multiple sexual partnerships, while professionals, technical, 
clerical & managerial workers have 18.5% probability of engaging in multiple sexual 
partnerships. According to table 3.9, men without any employment have the greatest 
likelihood of having at least two female partners. For every ten men who have no specific 
work, there are two of them having more than one sexual partner. 
44 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR 
DETERMINANTS OF MULTIPLE SEXUAL PARTNERSmpS 
4.0 Introduction 
Multiple sexual partnerships are influenced by a number of demographic and socio-
economic factors. In chapter three, examination was made of the relationship between 
multiple sexual partnership and the various demographic and socio-economic background 
characteristics (age, marital status, age at first sex, educational level, work status, religion, 
wealth status, region and place of residence) of respondents using bivariate methods of 
analysis. However, it may be difficult to identify the relative importance of the variables 
considered using only the result obtained from the bivariate methods of analysis. As a result, 
the logistic regression method described in chapter one under "Methodology" has been 
applied to show the net effect and strength of each of the independent variable on the 
dependent variable (sexual partnerships).This section therefore seeks to look at the results of 
the logistic regression model on the determinant of multiple sexual partnerships for men in 
Ghanaian communities. 
The summary of the logistic regression analysis are presented in the Table 4. In the 
table, three statistics are shown - the beta or logistic coefficient (fJ), the standard error 
(S.E.), the Exp (fJ) and the significance (Sig.) of the relationship between independent and 
dependent variables. 
45 
University of Ghana http://ugspace.ug.edu.gh
4.1 Determinants of Multiple Sexual Partnerships 
The logistic regression showed that men in older age groups are less likely to have 
multiple sexual partners than men within younger age groups. This confirms the result of the 
bivariate analysis that the likelihood of men engaging on multiple sexual partnerships 
decreases from younger ages to older age groups. For example men within the age group 
20-24 and 55-59 years are 25.5% and 76.4% less likely to have two or more female sexual 
partner respectively compared to men within the age group 15-19 years. The regression 
analysis also indicates that marital status of men is not a strong determinant of their having 
more than one female sexual partner. For example, never married men are only 6.7% likely 
to engage in multiple sexual partnerships compared with married/living together. 
Men's education emerges as a strong determinant of sexual partnerships. The model 
shows that education level of men significantly increases the probability of them having 
more than one sexual partner. For example men with primary or secondary education are 
about twice more likely to engage in multiple sexual partnerships than their counterparts 
with no formal education. Also those who completed higher education are 2.6 times more 
likely to have more than one female sexual partner than men without any formal education. 
Results from the logistic regression also indicate that men's occupation has a significant 
influence on the number of sexual partners they have. The likelihood for men in good 
employment positions to have more than one sexual partner is higher than men without any 
work. For example Professionalffechnical/Clerical and managerial workers are about 2.6 
times more likely to have multiple sexual partners than men who have no work. This result 
confIrm the truth of the hypothesis in chapter one that men with professional/technical and 
managerial work are more likely to have multiple sexual partners than men without any 
46 
University of Ghana http://ugspace.ug.edu.gh
work. In the same manner, men working in the sales and service sectors are three times more 
likely to have more than one partner than their counterpart without any work. 
Logistic analysis shows that men residing in Greater Accra and Volta regions are 
1.4 times and 1.5 times more likely to have multiple sexual partners than those in Western 
region respectively. This result implies that multiple sexual partners in Western region have 
little relationship with whether or not a man engages the services of prostitutes in that region. 
However, living in Upper East decreases the likelihood of having multiple sexual partners by 
58%, confmning the results of the bivariate analysis in chapter three. Also men in urban 
communities are 1.5 times more likely to have multiple sexual partners than those living in 
rural localities, substantiating assertions in literature review. 
Result also shows that Age at first sex has some influence on the number of sexual 
partners a man has. For instance, a man who initiates sex at age 15-19 and 24-29 years are 
1.5 and 1.3 respectively more likely to engage in multiple sexual partnerships than men who 
begins having sexual intercourse at age 25 years and above, thus confuming the results of the 
bivariate analysis and hypothesis iii of this study. 
Logistic analysis did not show religion as an important determinant of sexual 
partnerships. Results indicates that Moslems, Christians and men affiliated to the 
traditional/spiritualist sects are 40%, 26% and 43% respectively less likely to have more than 
one sexual partner compared to men with no particular religious affiliation. The results also 
indicate that the influence of wealth status on sexual partnerships were not as significant as 
expected. Men within the five categories of wealth status are equally as likely to have one 
sexual partner. The logistic model indicated the independent variables employed for the 
study was able to explain about 30% of the dependent variable (multiple sexual partnership). 
47 
University of Ghana http://ugspace.ug.edu.gh
Table 4: Logistic Regression models of Multiple Sexual Partnerships by Demographic 
And Socio- economic Characteristics of men in Ghana 
Background Logistic Standard 
Characteristics coefficient error(S.E. ) SIG. Exp(f3) 
(/3) 
Age group 
15-19 (RC) - - - 1 
20-24 -0.281 0.218 0.192 0.755 
25-29 -0.481 0.236 0.042 0.618 
30-34 -0.916 0.267 0.001 0.400 
35-39 -0.869 0.295 0.003 0.419 
40-44 -1.120 0.314 0.000 0.326 
45-49 -1.051 0.306 0.001 0.347 
50-54 -0.915 0.356 0.010 0.400 
55-59 -1.446 0.351 0.000 0.236 
Marital Status 
Never Married -2.702 0.178 0.000 0.067 
MarriedlLiving together (RC) - - - 1 
WidowedlDivorced/separated -2.883 0.167 0.000 0.056 
Religion 
None (RC) - - - 1 
Moslem -0.528 0.257 0.040 0.590 
Christian -0.299 0.236 0.204 0.742 
Traditional/Spiritualist -0.568 0.320 0.076 0.567 
Wealth status 
Poorest (RC) - - - 1 
Poorer -0.079 0.203 0.697 0.924 
Middle 0.004 0.211 0.985 1.004 
Richer -0.219 0.247 0.375 0.804 
Richest -0.065 0.275 0.813 0.937 
Education 
No education (RC) - - - 1 
Primary 0.749 0.202 0.000 2.114 
Secondary 0.7787 0.178 0.000 2.177 
Higher 0.957 0.291 0.001 2.605 
Place of residence 
Rural (RC) - - - 1 
Urban 0.436 0.167 0.003 1.546 
48 
University of Ghana http://ugspace.ug.edu.gh
Table 4 Continued: Logistic Regression models of Multiple Sexual Partnerships by 
D emog~Jh i cand SOC'l O- economic Characteristics of men in Ghana. 
Background Logistic Standard 
Characteristics coefficient error(S.E.) SIG. Exp(fJ) 
(/3) 
Occul!ational status 
Not working (RC) - - - 1 
Profess., Clerical, technical 
& managerial 0.968 0.230 0.000 2.633 
Sales and services 1.106 0.324 0.000 3.021 
Agriculture 0.945 0.214 0.000 2.574 
Skilled & Unskilled manual 0.945 0.191 0.000 2.596 
Region 
Western (RC) - - - 1 
Central 0.005 0.298 0.987 0.774 
Greater Accra -0.256 0.245 0.296 1.447 
Volta 0.370 0.293 0.207 1.508 
Eastern 0.411 0.276 0.137 0.883 
Ashanti -0.125 0.230 0.588 1.007 
BrongAhafo 0.007 0.268 0.980 1.005 
Northern -0.781 0.278 0.005 0.986 
Upper West -0.857 0.382 0.025 0.438 
Upper East -1.121 0.291 0.000 0.424 
Age At First sex 
< 15 0.136 0.266 0.609 1.145 
15-19 0.392 0.183 0.032 1.481 
20-24 0.258 0.183 0.1 59 1.294 
25+ (RC) - - - 1 
Constant 2.751 0.496 0.000 0.733 
(RC) =Reference category 
Number of Cases : 3,777 
R2 = 0.299 
Source: Computed by Author from GDHS, 2003 data file 
49 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FIVE 
SUMMARY, CONCLUSION AND RECOMMENDATION 
5.1 Summary 
The study was undertaken to frod out the factors that influence multiple sexual 
partnerships with respect to men's sexual behaviour in Ghana. Specifically, the study aimed 
at examining the levels of multiple sexual partnerships and the effects of selected 
demographic and socio-economic characteristics such as education, marital status, and 
religion among other factors on this important aspect of men's sexual behaviour. The data set 
used for the study was the Ghana Demographic and Health Survey, 2003 (GDHS, 2003). The 
study population comprised 3,777 sexually active men aged 15 to 59 years selected from all 
the ten regions in the country. 
Methodologies employed for the study included simple frequency, ratios, percentages 
and cross tabulation to examine the pattern and differences in sexual partnerships and use of 
logistic regression analysis to determine the contribution of each of the background 
characteristics to multiple sexual partnerships by men in Ghana. 
Analysis of background characteristics shows that more than half (about 52%) of the 
study population were below 30 years. About 40% percent are never married, 52.6 percent 
are married or living together with partners while 7.4% are divorced, separated or widowed. 
Fourteen percent of the respondents had primary education, while 25.4% had no formal 
education. However, about 53% had completed secondary school while only 7.4% completed 
higher education. Out of the country's ten regions, the highest proportion of respondents 
50 
University of Ghana http://ugspace.ug.edu.gh
(about 16%) was from Ashanti, while the Central region had least percentage (about 6%) of 
respondents. 
The proportion of respondents according to wealth status ranged from 17.8% 
(the middle quintile) to 23.6% (the poorest quintile). 21.5% of the respondents are in the 
richest quintile. Moslems accounted for about 21 percent of the population while 65% of the 
respondents were from the Christian faith. Seven percent have no religion while the 
remaining proportions are from the traditional and other religions. Most of the respondents 
(about 63%) were from rural areas. About 5% of the respondents had their first sex before 
age 14 years while 87% experienced sexual intercourse before age 25. About 13% initiated 
sex at age 25 years and above. 7.5% of the respondents were not working, 52.5% were 
engaged in the agricultural sector, and nearly 20% worked as manual labourers. About 9% 
were in the sales and services sectors while about 11 % were in the professional, managerial, 
clerical and technical fields. 
Bivariate analysis revealed some differences and levels in the number of sexual 
partners that men of various socio-economic and demographic backgrounds tend to have. 
Currently married men were more likely to maintain single sexual partners and avoid 
multiple partners than their Widowed! Divorced or separated counterparts. About 88.6% of 
currently married or men living together with partners have single sexual partner while 
11.4% are engaging in multiple sexual partnerships compared with about 80% and 20% 
respectively for never married men. This situation is expected because the currently 
marriedlliving together are constantly with their regular sex partners and thus will not be 
exposed to the temptation of exploring multiple sexual partners. Age pattern revealed that 
men within the age group 20-24 years have the highest probability (18.6%) of engaging in 
51 
University of Ghana http://ugspace.ug.edu.gh
multiple sexual partners as against 12% for men in the 45-49 age group. With the exception 
at age groups 15-19 and 50-54, multiple sexual partnerships follows a decreasing pattern 
from men within younger groups to older age group. 
Bivariate analysis further shows that men's educational attainment follows an 
increasing pattern on the number of sexual partners they have. About 14% of the sexually 
active men with no formal education have more than one sexual partner compared to 22% 
for those who had completed higher education. The study revealed that men affiliated to the 
traditional or spiritualist sects (with a proportion of 17%) are most likely to practice multiple 
sexual partnerships than those with no specific religion which has about 14% of its group 
engaging in this practice. However, Moslems, with a proportion of 14.6% are less likely 
compared to Christians, with a proportion of 15.5% to engage in multiple sexual 
partnerships. 
Analysis also revealed that wealth status has a positive influence on the number of 
sexual partners a man has. With the exception of men within the poorest category which has 
14% of its members having more than one partner, the tendency of men to engage in 
multiple sexual partnerships increases from those within the poorer quintile (11 %) to about 
19% for men within the richest quintile. Regional pattern shows that Greater Accra has the 
highest proportion of men, 21.4% who have more than one sexual partner while the Upper 
West region has the lowest percentage of men, 8.6% engaging in this practice. 
The practice of multiple sexual partnerships with respect to men's sexual behaviour 
is more pronounced in urban localities than in rural communities, 17.6% for urban as against 
13.5% for rural localities. Results further indicated that age at first sex has a bearing on 
multiple sexual partnerships. Men who initiate sexual intercourse at older ages are less likely 
52 
University of Ghana http://ugspace.ug.edu.gh
to engage in multiple sexual partnerships Gompared to men who start having sex at lower 
ages. Bivariate analysis shows that about 31 % of men who initiates sexual activity at ages 
below 15 years have two or more sexual partners while those who started at ages 25 years 
and above have 9.4% of their group involving themselves in this behaviour. Concerning 
occupational status, men having no specific work have 20% of their group having multiple 
sexual partners followed by men holding positions in the managerial/ professional/technical 
field (18.5%) while men in sales and services sectors as well as those engaged as skilled 
and unskilled manual labour have equal proportions, about 16.5% of men engaging in this 
sexual practice. 
Analysis of the logistic regression showed some differences with respect to important 
predictors of multiple sexual partnerships. However the significant predictors of this aspect 
of men's sexual behaviour were education, occupational status, place of residence and age at 
first sex. The logistic model indicates that men's economic positions significantly influence 
their exploration of multiple sexual partnerships. For example men engaged in various 
activities (e.g. sales & services, professional/managerial/technical and agriculture) are two or 
three times more likely to have multiple sexual partners than their counterparts without any 
work. The results revealed that men with at least primary or secondary education are more 
than twice likely to engage in multiple sexual partnerships than their counterparts with no 
formal education. Logistic results further indicate that men in urban communities are 1.5 
times more likely to engage in multiple sexual partnerships than their rural counterparts. 
However this model did not indicate wealth status and religious affiliation as significant 
determinant of multiple sexual partnerships in Ghana. 
53 
University of Ghana http://ugspace.ug.edu.gh
S.2 Conclusion 
This study has confirmed that multiple sexual partnerships exist among men of 
different background characteristics in Ghanaian communities. The findings from this study 
have shown that demographic and socio-economic factors have some influence on the 
tendency of men to engage in multiple sexual partnerships, a very important aspect of men's 
sexual behaviour. The results of this study, at least shows that education, occupational status, 
place of residence and age at fIrst sex are important socio-economic factors determining 
mUltiple sexual partnerships in Ghana. 
Notwithstanding the variations in magnitudes of this sexual practice among men of 
different socio-economic standings, this work has revealed that on the average 15.3 percent 
of sexually active men in Ghana have two or more female sexual partners while about 85% 
have single sexual partners. 
However, only about 30% of the independent variables considered for this study 
could explain this aspect of sexual behaviour of men in Ghana. This implies that other 
factors might be involved in the tendency of men to engage in multiple sexual partnerships 
with women. These other factors could be socio-cultural that influences man's attitude or 
behaviour, e.g. moral teaching learned from parents, peers or society. Others include man's 
own beliefs, traditions, practices (e.g. alcohol consumption), morals and personal motives. 
Also spirituality, which is described as a search process in an attempt to discover that which 
is sacred (e.g. God) through different pathways (pargoment and Mahoney, 2002), culture 
and societal norms and religiosity of an individual, which according to Hendrick and 
Hendrick (1987) is signifIcantly related to predicting or shaping one's sexual behaviour, 
54 
University of Ghana http://ugspace.ug.edu.gh
attitudes and experiences. A combination of these might bring about a man understanding of 
what he believes about sex and how he will approach sexual behavior. 
5.3 Recommendations 
The findings of this study have important policy implications for reproductive health. 
Results from the study have shown that men enter relationships at young ages some before 
10 years. The results further revealed that men who initiate sex at early ages tend to have 
multiple female sexual partners. The youth of today are exposed to sex earlier than formally 
through the internet, where strip and sex shows, marketing of women and children in 
commercial sex (Hughes, 2003). Hence the earlier they are taught the right things concerning 
their sexuality and the dangers of risky sex the better it will help develop and shape their 
attitude and behaviour towards sex before they are fully matured. There is no other way to 
achieve this other than through education. Sex education must form part of the school 
curricular and made a compulsory core subject from primary level through to Junior High 
school. Parents must also take up the challenge of discharging parental responsibilities of 
caring and proper monitoring of children developments, teaching them the right things about 
their sexuality before they learn the wrong things about sex from their peers. To help 
consolidate the efforts of parents and teachers in educating the youth on sex, virgin clubs 
should be formed in schools and in our communities. 
This study and literature review indicates that the number of young men having multiple 
female sexual partners is increasing. Meanwhile this aspect of sexual behaviour (multiple 
55 
University of Ghana http://ugspace.ug.edu.gh
heterosexual relationships) account for most HIV transmissions in Africa 
(Timberg, 2 0 0 0 ) . 
Recreation, an important mode of relaxation and enjoyment has the potentials to draw 
the youth attention from sex. However, recreational centers is lacking in our communities. 
There is therefore the need to establish youth centers at the districts and revamp the existing 
community centers across the country. The Districts Assemblies should take up this 
responsibility with technical and financial assistance from government. At these centers 
indoor and outdoor games as well as counseling sections on issues about sex and the 
importance of family planning should form part of the programs for young men in the 
communities. Recreational centers can offer some potentially good programs, with 
appropriate support, they could become an essential part of community wellbeing, providing 
a safe and positive place that promotes a healthy lifestyle to community members of all ages. 
Trained counselors specifically equipped with knowledge on sexual and reproductive health 
issues could act as important resource persons at these centers. Here, participants would 
benefit from specifically-targeted HIV prevention interventionS which address partner 
concurrency, inconsistent condom use, excessive alcohol consumption, and intimate partner 
violence. At a time when the world is still trying to find a lasting remedy for HIV I AIDS 
epidemic, this program when properly implemented and efficiently monitored could help 
control the spread of the deadly virus in Ghana as well as check the resultant social problems 
that accompanied multiple sexual partnerships, like unwanted pregnancies, abortions 
fatherless and street children in our societies. 
While focusing on the youth to develop and follow a healthy and responsible sexual 
behaviour, attention should also be drawn to older men in our societies. Older men in our 
56 
University of Ghana http://ugspace.ug.edu.gh
communities must be called upon to show excellent examples for the youth to follow, 
especially married men who must realize the importance of faithfulness and respect for their 
wives, the people they love. Men's superior socio-economic position as against their female 
counterparts make the latter to contend with the allurement of financial gratification and 
sexual overtures by relatively richer peers and older males (the "sugar daddy" syndrome) 
whereby school girls for example enter into sexual relationships with older, wealthy men 
who can assist them with school related expenses or the purchase of material goods 
(DeBrugn, 1992; Gorgen et al, 1993) These men should appreciate the importance of keeping 
their love under one roof by being faithful to their wives and know that having several sexual 
partners in addition to their wives/partners is dangerous for their health and the well-being of 
their families. 
It is also recommended that any future research in the area of men's sexual behaviour 
should explore other plausible factors, including research on local nonns, religiosity and 
spirituality that affect individual and group attitudes and behaviours in relation to sexual 
activity. 
57 
University of Ghana http://ugspace.ug.edu.gh
Bibliograpby 
Adimora,AA, Schoenbach VJS, Martinson FEA, et al., 2003 Concurrent partnerships 
among rural African Americans with recently reported heterosexually transmitted HN 
infection. Epidemiol Soc Sci 
Adih, WK and Alexander C.S. (1999), Detenninants of condom use to prevent HN 
infection among youth in Ghana, Journal ofA dolescent Health" 24(1): 63-72 
Addo VN; Tagoe-Darko E.D, (2009), Knowledge, practices, and attitudes regarding 
emergency contraception among students at a university in Ghana. International Journal of 
Gynaecology and Obstetrics. (3):206-209 
Anarfi, John (1999), Sexuality, Migration and AIDS in Ghana - A socio-behavioral study' , 
Institute for Statistical, Social and Economic Research, University of Ghana, Legon, Accra, 
Ghana. 
Anderson RM., 1992. Some aspects of sexual behaviour and potential demographic impact 
of AIDS in developing countries. Social Science and Medicine 34(3): 271-80. 
Ankomah A; Ford, N. 1994, Sexual exchange: understanding pre-marital heterosexual 
relationships in urban Ghana. AIDS: foundations for the future,123-35 
Awusabo-Asare, et al, 2008, Wealth Status and Risky Sexual Behaviour in Ghana and 
Kenya, Applied Health Economics and Health Policy, Volume 6, Number., pp. 27-39 
Awusabo-Asare, Kofi ; Albert M. Abane and Akwasi Kumi-Kyereme, 2004 Adolescent 
Sexual and Reproductive Health in Ghana: A Synthesis of Research Evidence. 
Cooper, A., Scherer, C., Boies, S., Gordon, B. Sexuality on the Internet: From Sexual 
Exploration to Pathological Expression. 1999. Vol. 30(2), pp. 154-164. 
www.apa.org/joumals/features/pr0302154.pdf 
Dixon-Mueller R: The sexuality connection in reproductive health, Studies in Flamily 
Planning 1993 Sep-Oct;24(5):269-82. 
Djamba, Yanyi K.(2004), Sexual Behaviour of Adolescents in Contemporary Sub-Saharan 
Africa 
Dodoo, F.N.A 1995. Contraceptive behaviour in Ghana: a two- sex model, International 
Journal of Sociology oft he Family 
Dodoo, F. N-A, and L. F. DeRose. 2005. Structural and Cultural Influences on Women's 
Bodies: Relative Influences of Education and Men's Dominance on Reproductive Decisions. 
Unpublished Manuscript. 
58 
University of Ghana http://ugspace.ug.edu.gh
Dodoo, F.N.A 1998. Men's matter: Additive and interactive gendered preferences and · 
reproductive behaviour in Kenya, Demography 35: 229-243 
Dudgeon, M.R. and Inborn Marcia, C., 2004, Men's influences on women's reproductive 
health: medical anthropological perspectives 
Edlin B, Irwin K, Ludwig D, et al. High-risk sex behavior among young street-recruited 
crack cocaine smokers in three American cities: an interim report. J Psychoactive Drugs 
1992 24: 363-71. [Web of Science [Medline] 
Ezeh AC ; Seroussi M ; Raggers H: Men's fertility, contraceptive use, and reproductive 
preferences. Calverton. Maryland, Macro International, 1996 Mar. viii, 45 p. (Demographic 
and Health Surveys Comparative Studies No. 18) 
Fatusi A and Wang W2 2008 Predictors of early sexual initiation among a nationally 
representative sample of Nigerian adolescents. Department of Community Health, College of 
Health Sciences, Obafemi Awolowo University, He-Ife, Nigeria. 
Ghana Statistical Service, Ghana Demographic and Health Survey, (DHS) 1998 
Hall, Peter A., Holmqvist, B.A.,Sherry, Simon B 2004, Risky Adolescent Sexual Behavior: 
A Psychological Perspective for Primary Care Clinicians 
Hulton L ; Falkingham J. Male contraceptive knowledge and practice: what do we know? 
Reproductive Health Matters 1996 May;(7):90-100. 
Kalichman, Seth C et al, 2007, Recent Multiple Sexual Partners and HIV transmission risks 
among people living with HIV/AIDS in Botswana 
Karra MV ; Stark NN ; Wolf J, Male involvement in family plimning: a case study spanning 
five generations of a South Indian family. Studies in Family Planning. 1997 Mar;28(1) 
Kimuna, Sitawa R.; Djamba Yanyi K., 2005, Wealth and Extramarital Sex Among Men in 
Zambia. International Family Planning Perspectives,Volume 31, Number 2 
Kongnyuy, Eugene; Wiysonge, Charles; Mbu Robinson; Nana Philip; Kouam Luc, (2006,) 
Wealth and sexual behaviour among men in Cameroon. . 
Kwankye Stephen O. (2003) Female Adolescent Sexuality, Contraceptive Health in Ghana: 
A case study of Cape Coast and Mankrong- Theses presented at the Regional Institutes for 
Population Studies (RIPS). 
Laumann, E., Gagnon, J.H., Michael, R.T., and Michaels, S. The Social Organization of 
Sexuality: Sexual Practices in the United States. 1994. Chicago: University of Chicago Press 
(Also reported in the companion volume, Michael et al, Sex in America: A Definitive 
Survey, 1994). 
59 
University of Ghana http://ugspace.ug.edu.gh
Mackay, Judith ,How does the United States compare with the rest of the world in human 
sexual behavior? Copyright BMJ Publishing GroupWest J Med. 2001 June; 174(6): 429-
433. 
Mbizvo MT ; Bassett MT ..R eproductive health and AIDS prevention in Sub-Saharan 
Africa: the case for increased male participation. Health Policy and Planning. 1996 
Mar;II(I):84-92. 
Morris L., Sexual Behavior of Young Adults in Latin America.: Advances In Population: 
Psychological Perspectives.1994;2:231-52. 
Mosher WD, Chandra A, Jones J. Sexual behavior and selected health measures: Men and 
women 15-44 years of age, United States, 2002. Advance data from vital and health 
statistics; no 362. Hyattsville, MD: National Center for Health Statistics. 2005. 
Nabila J .S. and Fayorsey C, Youth and Reproductive Health in Africa: Assessment of 
Adolescent Reproductive Health Needs in Ghana, Accra, Ghana: United Nations Population 
Fund (UNFPA), 1996. 
Lansky A, Nakashima AK, Jones J. Risk behaviors related to heterosexual transmission from 
HIV-infected persons. Sex Transm Dis 2000 27:483-9.Web of Science Medline 
Njue, Carolyne; Voeten, Helene; Remes, Pieter, 2009, Disco Funerals: A Risk Situation for 
HIV Infection Among Youth in Kisumu, Kenya 
Ngom P: Men's unmet need for family planning: implications for African fertility 
transitions. Studies in Family Planning ,1997 Sep;28(3):192-202. 
Pelletier A, Messiah A, International Conference on AIDS, Sexual practices of heterosexual 
men with multiple partners in France: results from ACSF, the French National Survey., Int 
ConfA IDS. 1992 JulI9-24; 8: D504 (abstract no. PoD 5694). 
Robey B; Thomas E ; Baro S ; Kone S ; Kpakpo G: Men: Key partners in reproductive 
health. A report on the First Conference of French-Speaking African Countries on Men's 
Participation in Reproductive Health, March 30 -April 3, 1998, Ouagadougou, Burkina Faso. 
Rweyemamu, Datius, 2007, Multiple and Concurrent Sexual Partnerships Among youth in 
Tansani5!,. University of Dar es Salaam 
Schehl M ; Green C. Men as Partners initiative: summary report ofliterature review and case 
studies.: New York, New York, AVSC International, 1997. v, 45 p. (USAID Cooperative 
Agreement No. CCP-3068-A-00-3017-0) 
Wegner MN; Landry E; Wilkinson D; Tzanis J. Men as partners in reproductive health: from 
issues to action. International Flamily Planning Perspectives. 1998 Mar;24(1):38-42. 
60