Self-Esteem and Attitudes of Girls Orphaned to HIV/AIDS towards Education in Kampala, Uganda

Aims: To examine the relationship between self-esteem and attitudes to education among orphaned and non-orphaned adolescent secondary school girls in Kampala. Study Design: Cross-sectional study. Place and Duration of Study: Institute of Psychology, School of Education, Makerere University, Kampala, Uganda between June and October 2005. Methodology: Questionnaire including the socio-economic indicators, Rosenberg General Self-Esteem Scale and the Attitude Scale were administered to 225 students who were selected by simple random sampling in six secondary schools in Kampala, Uganda. Results: Orphaned girls had lower self-esteem and most had a negative attitude to education compared to non-orphans. However, girls orphaned to HIV/AIDS had a higher self-esteem compared to those orphaned by other causes and there was a positive correlationship between self-esteem and attitude towards education among orphaned adolescent girls in Kampala, Uganda. Conclusion: Non-orphaned girls have a higher self-esteem than orphaned girls. Nonorphaned girls have a more positive attitude to education compared to orphaned girls. Girls orphaned to HIV/AIDS have slightly higher self-esteem than those orphaned by other causes, but had the worst attitude to education. Recommendations: A special needs program should be designed to support orphan girls in secondary school so as to enable them perform better in their education. Research Article International Journal of TROPICAL DISEASE & Health, 2(2): 87-99, 2012 88 Additional study should be conducted among secondary school boys to compare the effects of orphan hood to their education and to see whether there are some disparities in the two genders of the same age group.


INTRODUCTION
In the year 2010 there were an estimated 22.9 million people living with HIV in Sub-Saharan Africa (UNAIDS, 2011). This has increased since 2009, when an estimated 22.5 million people were living with HIV, including 2.3 million children (UNAIDS, 2010). The increase in people living with HIV could be partly due to a decrease in AIDS-related deaths in the region. There were 1.2 million deaths due to AIDS in 2010 compared to 1.  Because of the growing number of orphans in some of the sub Saharan African countries as a result of HIV/AIDS due to the experience of generalized epidemics with the prevalence of HIV/AIDS being more than 5% e.g. in countries like Uganda with 6.5%, we found it appropriate to investigate the effect of HIV/AIDS orphan hood to the education of adolescent girls in this part of the globe. Several observation studies conducted in Ugandan secondary schools had indicated that orphaned girl students were shy, timid, lacked confidence and were usually not part of the best ranked performers in examinations at most levels of educational system in Uganda (Opolot et al., 1999;ESA, 2000). We therefore chose to examine the relationship between self-esteem and attitudes to education among orphaned and non-orphaned adolescent secondary school girls in Kampala.
One study though in Uganda had observed also that education increased girls' knowledge, self-confidence and self-esteem (FAWE, 2000). However, it was also observed that school drop-out rates were higher among females students (7.6%) compared to males (6%) in Uganda (Baguma and Muhanguzi, 2000). Most of the reasons for female school drop out were due to family socio-cultural reasons including illness of parents due to HIV/AIDS or orphan hood due to HIV/AIDS or other causes (Baguma and Muhanguzi, 2000). Other sociocultural reasons included the gender inequality accorded to girls (Uganda Bureau of statistics, 2006). Boys were generally provided with the opportunities to continue with their education uninterrupted, while girls were usually requested by their families to stay at home to continue providing household services in the event of illnesses or demise of their parents (Baguma and Muhanguzi, 2000;Uganda Bureau of Statistics, 2006). It should be noted here that, orphan hood due to HIV/AIDS has been one of the greatest effects of HIV/AIDS on school girls' education in Africa leading to school absenteeism during their parents' illnesses and emotional stress manifested in inappropriate behavior (Baggaley and Needham, 1997;Yun, 2001;Wahl, 2001).
Self-esteem is described as a life skill which is a social competence used to cope with academics in order to meet fundamental challenges of forming stable human relationships and maintaining hope about the future (Opolot et al., 1997;Wahl, 2000). High self-esteem is described as a realistic evaluation of the self's characteristics and competencies, coupled with an attitude of self-acceptance and self-respect (Berk, 1997;Yun, 2001).
Most often, loss of a parent was observed as one of the most tragic and stressful events in a child's life (Wahl, 2000). Instead of the child detaching him/herself from the loved ones and resolving the loss, there was a greater attachment to the deceased parent resulting into denial of death (Gunther et al., 1998;Wahl, 2000). Previous studies conducted here in Uganda observed that losing a parent to HIV/AIDS for the adolescents was a more grieving process which was unfortunately shrouded with stigmatization, secrecy and shame (Gunther et al., 1998;Wabwire-Mangen, 2008;Kitara et al., 2011). They experienced a double abandonment; one from the parent and the other from a society that shunned them for being associated with HIV/AIDS (Gunther et al., 1998;Kitara et al., 2011). These negative factors drove teenagers underground as their coping capacities were stretched to the limits (Gunther et al., 1998;Kitara et al., 2011). They had to maintain the precarious balance of continuing the normal routines while acknowledging that within several months or years, their parents were lost forever (Gunther et al., 1998). The stigma attached to HIV/AIDS made the situation worse for girls orphaned to HIV/AIDS (Kitara et al., 2011;Wabwire-Mangen, 2008).
To-date, the number of HIV/AIDS orphans in secondary schools in Uganda is estimated to be about two million people (ESA, 2000;UNAIDS, 2010). The girl child's education was the most affected by the HIV/AIDS orphan hood because they were prevented to go to school and instead asked to remain at home to nurse their other younger siblings and sometimes had to forego schooling completely as their male counterparts continued their learning in schools (ESA, 2000;Uganda Bureau of Statistics, 2006;Wabwire-Mangen, 2008). They were stigmatized for being association with HIV/AIDS and their levels of self-esteem became drastically affected (Opolot et al., 1997, Kitara et al., 2011 and consequently their attitude towards education were expected to be lowered (Opolot et al., 1997;Wabwire-Mangen, 2008).

Rationale for the Study
Several observational studies conducted in Ugandan secondary schools had indicated that orphaned girl students were shy, timid, lacked confidence and were usually not part of the best ranked performers in examinations at most levels of educational system in Uganda (Opolot et al., 1999). It was also observed that there were a growing number of orphans due to HIV/AIDS in Uganda (UNAIDS, 2010). This was thought to occur because the country experiences a generalized epidemic with the prevalence rate of 6.5%. It was on this background that we found it appropriate to investigate the effect of HIV/AIDS orphan hood to the education of adolescent girls in Uganda. We examined the relationship between selfesteem and attitude towards education of girls orphaned to HIV/AIDS, girls orphaned by other causes and non orphaned adolescent secondary school girls in Kampala, Uganda.

METHODOLOGY
Study Site: This study was conducted in six secondary schools in four of the five divisions of Kampala City in Uganda. The schools were selected by a simple random sampling from a list of secondary schools from in each of the divisions of Kampala city. On average, three secondary schools were found eligible in each of the divisions of the city and these three schools were subjected to a random sample by picking a ballot from a basket as we selected a school to be studied per division. For the remaining two schools, all the ballots that were not picked previously picked were returned to the same basket and mixed together and the first two ballots picked consecutively by the researcher were considered part of the study population.

Study Design
A cross-sectional study was conducted between June and October 2005. A cross section was chosen as a quick method of appraisal of orphans to HIV/AIDS towards education in secondary schools so as to prepare for a more comprehensive observation studies on these students.

Study Population
Six Secondary schools (5 Government day school and 1 Private boarding school) were selected for the study, which involved 225 students, all from senior one, two and three classes; 75 students in each of the following three strata: Girls orphaned to HIV/AIDS, girls orphaned by other causes and non-orphaned girls. Girls orphaned to HIV/AIDS were selected by simple random sampling. The list of names of orphans to HIV/AIDS was provided to the researchers by the Head teacher in each school and each name was written on a piece of paper that was folded into a ballot and placed in a basket and shaken to mix properly. Thirteen ballots were consecutively picked from the basket as a sample for orphans to HIV/AIDS in each school. If a student who was picked refused to consent to the study, she was replaced by randomly picking another ballot in that particular stratum. This same technique applied to non orphaned students and orphans due to other causes such as death in wars, diseases, road traffic accidents and old age to represent each school and stratum sampled.

Study Instruments
Questionnaire to the students and interviews of the Head teachers were the methods used for data collection. The questionnaire assessed the socio-economic status of students and the two standard scales included: The Rosenberg General Self Esteem Scale and the researcher constructed Attitude Scale. These two standard scales were modified to suit this particular study. The socio-economic indicators for the students' family were: father and mothers' occupation, their educational levels, their estimated monthly income and their possessions such as Video recorder, Television, cars and the size of the home, home ownership and the number of people in the home which were all used to judge the socioeconomic status of the families.

The Rosenberg General Self-Esteem Scale (Modified)
Whereas the standard scale had reliability coefficient of Cranbach's Alpha 0.92 and a validity of 0.60, this modified scale had a reliability coefficient Cranbach's Alpha set at 0.55. The scale had ten items on a Likert-type scale requiring responses of Strongly Agree (SA), Agree (A), Don't know (DK), Disagree (D) and Strongly Disagree (SD).

The Attitude Scale (Modified)
The Each Head Teacher of the schools studied was interviewed about the performances of the orphaned and non orphaned girls and to confirm the status of their orphan hood. A list of names from the government and NGOs which were in possession of the head teachers confirming orphan hood were provided to the researchers. This helped the researchers to verify and/or confirm the status of orphan hood of the students interviewed. Four undergraduate students from Makerere University were recruited and trained as research assistants by the investigators before undertaking the research. It took each student approximately 30-40 minutes for them to answer all the questions in the study instrument.

Data Collection Procedure
Data for the study was collected from the students using a questionnaire but the information obtained was also confirmed by the Head teachers who provided detailed accounts of the students during their interviews. Before the questionnaire was administered to the selected schools and sampled students, a pilot study was conducted in one secondary school that was not part of research site. Thereafter, modifications were made to the questionnaire until it was considered sufficient to answer the objectives of this study.

Ethical Consideration
Assent was obtained from each student in the presence of the senior woman teacher of the school as the Guardian and confidentiality of information was ensured by the principal investigator during and after the study. The principal investigator collected all the questionnaires, stored them under lock and key and personally supervised the entry of the data in to a computer which was pass-worded. Each student answered her questionnaire in a private classroom and only when clarifications on some questions were required, did the researchers and the senior woman teacher ever moved to her desk to make clarifications on those questions. This phenomenon occurred only once during the whole study period and it was because the questions were pilot tested and modified into simple and direct points. Ethical clearance and approval for study was obtained from the Uganda National Council of Science and Technology (UNCS&T).

Data Analysis
Statistical Package SPSS version 12.0 was used to analyze the data. Levels of self-esteem and attitude to education of adolescent secondary school girls, orphaned and non-orphaned were computed. Descriptive statistics were obtained for the study variables. A students t-test at p<0.05 level of significance was used for comparison of variables. The Pearson correlation coefficient was used to determine the relationships, the direction and magnitude of self-esteem and attitude to education. ANOVA was used to establish the differences between the groups and their levels of self-esteem and attitude towards education. Regression analysis was used to establish the effect of both self-esteem and attitude to education. Table 1 shows socio-demographic characteristics of the study population: 67.9% (150) students were orphans and 32.1% (70) were non-orphans. Also14% were aged 12-13years; 55.7% aged 14-15years and 30.3% aged 16 years and above. Most orphans were Christians (71.0%).

RESULTS
Less than 5% of both orphaned and non-orphaned girls were from rich families. Over half (53%) of the orphaned girls compared to 39% non-orphaned girls were from poor families. Less than half (46%) of orphans and more than half of non-orphans were from average income families. In table 3, Non-orphan secondary school girls had the highest level of self-esteem (mean = 3.91, SD +0.51) than orphans. Girls orphaned to HIV/AIDS had a slightly higher self-esteem (mean = 3.47, SD+ 0.73) compared to orphans because of other causes (mean = 3.39, SD+0.59). Non-Orphan secondary school girls had a more positive attitude towards education (mean = 3.98, SD+0.38) than the orphaned girls. Among the orphans, those orphaned to HIV/AIDS had a lower attitude to education (mean = 3.77, SD +0.50). There was a slight tendency for girls orphaned to HIV/AIDS to score higher on self-esteem than orphans by other causes and this was statistically significant with p= 0.01.  Table 4 shows the means of non-orphaned girls which were higher than those for orphaned girls. Non-orphaned highest mean score was in the attitude to peers (mean=3.98). Orphaned girls highest mean score was in attitude to teachers (mean=3.94). This result confirmed the positive and a statistically significant relationship between self-esteem and orphan hood (t =-5.51, p= 0.00) using a student's t-test. The breakdown in the attitude scale into subcomponents showed that attitude towards school facilities had a negative correlation but a statistically significant (t=-3.13, p=0.00) relationship with the attitude scale. The rest of the sub-components were not statistically significant; Attitude towards peers, class subjects and attitude towards teachers (table 4). This meant that girls in the two groups had different attitudes towards school facilities but the difference was not statistically significant. Causes of Orphan-hood: According to table 3, orphans to HIV/AIDS had a higher selfesteem (mean= 3.47, SD+0.73) than orphans to other causes. Orphans to HIV/AIDS had a less positive attitude to education (mean=3.77, SD+0.50) compared to orphans by other causes (mean=3.92, SD+0.47). This differences in the attitudes among the orphans was statistically significant (p=0.01). Table 5 indicates that the orphaned girls' attitude towards education correlated positively with the level of self-esteem (r = 0.45, p= 0.00). This meant that the higher the self-esteem of the girl the more positive was her attitude towards education.  Table 6 indicates that self-esteem was positively and statistically significant to 3 aspects of attitudes to education: (r=0.49, p=0.00) towards peers; (r=0.35, p=0.00) towards class subjects; (r=0.33, p=0.00) towards teachers, and (r=0.11, p=0.11) towards school facilities. This meant that the higher the self-esteem, the more positive her attitude was towards her peers, teachers and class subjects.

DISCUSSION
The majority of orphaned girls studied were from families with poor socio-economic backgrounds while the majority of non-orphaned girls were from average income families. This situation was confirmed by the Head teachers of these schools at interviews. They acknowledged that, because of many poor orphans who could not afford school fees in their schools, they were unable to run their schools properly because of lack of funds.
They reported that Uganda government had practically ceased its contribution of fees to orphans and only Non Governmental Organizations (NGOs), good Samaritans and charitable organizations were paying school fees for orphans.
Furthermore, most girls orphaned to HIV/AIDS lived with their close relatives as opposed to those orphaned by other causes (death at wars, diseases, road traffic accidents and old age). The fact that most orphans were from poor families could have affected their selfesteem and attitude towards education. Previous studies and experiences from Uganda had evidence to show that female headed households experienced far higher poverty levels because most females were often less educated and usually with less paying jobs (Uganda Bureau of Statistics, 2006).
The majority of the parents of HIV/AIDS orphaned girls died during 1990-1999, perhaps as a result of infections that occurred in the 1980s when the HIV/AIDS awareness levels were low in this country. It had been reported that at the time, HIV/AIDS prevalence in Uganda was about 30% (Uganda AIDS Commission, 2000). Although the majority of orphans to HIV/AIDS stayed with their close relatives, the environment did not provide a conducive atmosphere similar to when their parents were still alive. Price and Cioci (1993) noted that erosion of familial and social supports complicated the lives of adolescents and that family safety nets were weakened as a result of orphan-hood resulting from HIV/AIDS (Foster, 2000). Other studies found that emotional stress was one of the major challenges facing orphans (Baggaley andNeedham, 1997, Wahl, 2000).
Girls orphaned to other causes and non orphans came from families with average socioeconomic status and were therefore, better off in terms of social and family support. This study found that non orphans had the highest level of self-esteem and the best attitudes to education. Girls orphaned by other causes had lower self-esteem compared to girls orphaned to HIV/AIDS, although they had more positive attitudes to education. This therefore brings in a new dimension to this study to find out whether there are other factors that influenced one's attitude to education other than self-esteem. Further studies need to be conducted to assess whether life skills such as assertiveness and socio-economic status of the families could be major players in students' attitudes to education.
Several studies have established that orphan hood due to HIV/AIDS and other causes stressed up students and their self-esteem was often lowered due to the discriminatory manner in which they were treated (Denscombe, 2000;Marcotte, 2002) and this influenced their levels of self-esteem and attitudes to education (Van Heerden, 1999;Kelly, 2000;Marcotte, 2002).
Studies from the western world have shown that girls in racist schools needed higher levels of self-esteem and assertiveness to cope in schools in order to perform well (Van Heerden, 1999;Griffiths, 2002).
From our study, self-esteem of orphans to HIV/AIDS was higher than that of orphans by other causes. It is important to note here that over the last 25 years, Uganda put a lot of human and financial resources in sensitizing the public about the need to overcome HIV/AIDS stigma (Nalwanga, 1998;Tusabe, 1998;Wabwire-Mangen, 2008). Uganda became known internationally for tackling the problem of HIV/AIDS stigmatization with utmost openness (Wabwire-Mangen, 2008). In doing so, the HIV/AIDS victims and those affected had become more accepted by the public and thus, their self-esteem was rising (Nalwanga, 1998;Tusingwire, 1996;Wabwire-Mangen, 2008). Although people affected by HIV/AIDS were accepted by the public, current observation showed that in some communities, stigmatization was still quite intense (Nalwanga, 1998;Baguma, 1997;Wabwire-Mangen, 2008;Kitara et al., 2011).
Individuals and their families still faced HIV stigmatization more especially among the youths (Kitara et al., 2011). Adolescents often faced many challenges which adversely affected their perception of self-worthiness including peer acceptance which was key to self-esteem (Vonk, 2001).
Other studies have shown that self-worth was crucial during adolescence when other pubertal and growth transitional factors were at play particularly friends' intimacy was positively associated with self-esteem and purpose in the life of adolescents (Kee-lee, 2000). Furthermore, studies have shown that any challenges faced by girls during adolescence often resulted in depressive symptoms and subsequently the lowering of their self-esteem (Hart and Thomson, 1996;Wahl, 2000).
Self-esteem had a significant positive relationship with attitude towards education; the higher the level of self-esteem the more positive the attitude to education was. One such study has observed that an internalized self more than classroom requirements, often motivated girls to pursue education (Potter, 2001). Girl-child particularly in Africa had a double disadvantage of poverty and gender inequality that further negated their self-esteem reflecting the influence of stereotyped characteristics associated with her (Tibamwenda, 1994;Wahl, 2000;Wabwire-Mangen, 2008).
The attitude towards education of girls orphaned to HIV/AIDS was lowest in spite of a fairly high self-esteem. As already stated, the relatively higher self-esteem could be attributed to government and non governmental efforts to sensitize the populace on the need to destigmatize HIV/AIDS. This finding had contradicted the notion that high self-esteem necessarily led to positive attitude to education (Kelly, 2000). Some studies have the view that teachers, who besides parents spent a lot of time with students, needed to protect girls in order to promote their attitude towards education (Kelly, 2000;Young and Warrington, 1996). The teachers therefore, needed to provide the necessary support environment to the girl-child especially those orphaned to HIV/AIDS.

CONCLUSION
Non-orphaned girls have a higher self-esteem than orphaned girls. Non orphaned girls have a more positive attitude to education compared to orphaned girls. Girls orphaned to HIV/AIDS have slightly higher self-esteem than those orphaned by other causes (death at wars, diseases, Road traffic accidents and old age), but had the worst attitude towards education.