Colombia Ghana Kenya Nepal

Parental Involvement and Academic

Performance in Ghana
By Gina Chowa, David Ansong, & Isaac Osei-Akoto
December 2012 CSD Publication No. 12-42

Introduction children in Sub-Saharan African countries). This brief

reports data from the ongoing YouthSave Ghana
If provided an opportunity to save via formal financial Experiment to offer preliminary findings of the extent
services, will youth participate? This is one of the of parental involvement in childrens schooling and how
fundamental questions being asked by YouthSave, a socio-demographic factors are associated with parental
four-country study targeted for young people ages 12 to involvement. Understanding baseline associations will
18 living predominantly in low-income households. Youth enable investigators to determine if these factors are
do save informally andif given an opportunityalso affected by the YouthSave intervention.
may participate in formal banking services (Save the
Children Federation, Inc., 2012; UNCDF, 2011), but such
opportunities are few. The limited research available Methods
suggests that financial inclusion has important youth The YouthSave Ghana Experiment uses a cluster
development effects and deserves greater study (Chowa randomized design with 100 schools selected randomly
& Ansong, 2010; Deshpande & Zimmerman, 2010; Elliott, from eight of Ghanas ten regions. Fifty schools were
2012; Scanlon & Adams, 2009; Ssewamala & Ismayilova, assigned randomly to treatment condition, and another
2009). 50 were assigned randomly to control condition. Sixty
students were selected randomly from each school for
YouthSave is a pioneering project designed to increase a total of 3,000 youth in treatment condition and 3,000
savings and development among low-income youth in control condition with oversampling for attrition. This
in Colombia, Ghana, Kenya, and Nepal. The goals of process yielded a sample of 6,252 youth who completed
YouthSave research are to measure the uptake, savings a baseline survey. Data from this research brief are from
outcomes, experiences, and impacts of youth savings 4,572 parents and guardians who also completed a
accounts (YSAs) on clients and financial institutions. In baseline survey. Figure 1 presents the gender distribution
Ghana, a rigorous research design that includes a control of parents and guardians interviewed.
group with quantitative and qualitative evidence has
been implemented to assess the impact of savings Figure 1: Gender Distribution among Parents and
accounts on youth development and asset accumulation.
Guardians
Parental Involvement and
Female
Academic Performance
A key question of the YouthSave Ghana Experiment is Male
Guardian 10% 25%
how savings and other factors impact educational
outcomes. Most empirical evidence suggests that
parental or guardian involvement (hereafter referred to
as parental involvement) is associated positively with
students performance in school (Nyarko & Vorgelegt,
Parent 20% 45%
2007; Topor, Keane, Shelton, & Calkins, 2010). Although
research in this area is increasing, attention has not
focused on specific populations (e.g., 12- to 14-year-old
0 10% 20% 30% 40% 50% 60% 70%

Center for Social Development

Data were collected from May through June 2011 by our This is changing as more parents are interacting with the
partners at the Institute of Statistical, Social and school environment by attending school meetings and
Economic Research (ISSER) at the University of Ghana. At recreational events (Elam, Rose, & Gallup, 1994). Other
baseline data collection, parents or guardians (hereafter empirical research consistently has found associations
referred to as parents, mothers, or fathers) were asked between a host of socio-demographic factors (e.g.,
eight questions about their level of involvement. Four economic circumstances, personal priorities, self-
questions assessed parents engagement in their childrens interest, childs and parents level of education, family
education at home, and four questions assessed parents structure, and family size) and the nature, extent, and
engagement with their childrens school environment. Key educational outcomes of parental involvement (Schmitt
variables we examine are multidimensional measures of & Kleine, 2010; Schimpl-Neimanns, 2000).
parental involvement, socio-demographic characteristics
of youth and parents, and math and English scores. Engagement within the School
Analyses are limited to univariate and bivariate methods. Environment
Nature and Level of Parental As depicted in Figure 2, parental involvement in the
school environment appears to be high because a
Involvement majority (87%) report attending PTA meetings, the most
We investigate the nature and level of parental common form of parental involvement at school.
involvementincluding engagement and monitoringby
comparing parents who had never been engaged in their Engagement at Home
childrens education to those who had been involved both
in the school environment and at home. Nyarko (2011) Parents report a high level of involvement with their
observes that while Ghanaian parents often have engaged children at home (Figure 3). The majority of parents
in their childrens schooling in one form or another, their (91%) report discussing expectations with their children,
involvement historically has been limited to school- but nearly 6 out of 10 parents (57%) report they never as-
related activities at home (e.g., ensuring completion of sist their children directly with homework.
homework).

Figure 2. Nature Never

44%
of Engagement Volunteer at school
within the School 56% Yes
Environment
60%
Attend school events
40%

Speak to teachers/ 73%

counselors 28%

87%
Attend PTA meetings
13%

0 20% 40% 60% 80% 100%

Figure 3. Nature Never

of Engagement Discuss 72%
school work Yes
at Home 28%

Talk about 91%

expectations 9%

Assist with 43%

homework 57%

Ensure 83%
homework
is done 17%
0 20% 40% 60% 80% 100%

2
Extent of Engagement at Home and in Parents report attending PTA meetings very often more
than for any other measure of involvement at school.
School Such meetings are often mandatory for parents who may
As Nyarko (2011) and Elam and colleagues (1994) point attend to avoid penalties.
out, Ghanaian parents historically have been more
involved with their children at home than in school. To Parental Involvement and Socio-
assess whether engagement within the school
environment is still lagging behind engagement at home, demographic Characteristics of Parents
we compare parents who had absolutely no involvement Nearly every empirical study on parental involvement
to those who had at least some. We find that average finds that parents engagement in their childrens
involvement at home (M=2.89, SD=0.25) exceeds average education varies by socio-demographic and economic
involvement at school (M=2.77, SD=0.19), but the circumstances, such as marital status, educational level,
difference is minimal. Almost all parents report that they and the childs gender (Georgiou, 2007; Schimpl-
never devote time to assist with childrens homework at Neimanns, 2000; Schmitt & Kleine, 2010). These
home (Figure 4). differences are consistent with findings from the Ghana
Experiment baseline data. In the following sections, we
Regarding engagement in school, involvement in most present differences in parental involvement found for
measures is moderate across the board with the parents marital status, education level, and gender and
exception of attendance at PTA meetings (Figure 5). the childs gender.

3,000 Figure 4. Extent

Discuss of Engagement
school work at Home
2,500 Talk about
expectations
2,000 Assist with
homework
Ensure
1,500 homework
is done
1,000

500

0
1- 2 3 4 5-
Never Very often
Figure 5. Extent
3,000 Volunteer of Engagement
at school within the School
2,500 Attend Environment
school events
Speak to teachers/
2,000
counsellors
Attend PTA
1,500 meetings

1,000

500

0
1- 2 3 4 5-
Never Very often
3
Marital Status a level of education equal to (20%) or greater than (36%)
that of their parents (Figure 7).
Married parents are more likely to be involved in their
childrens education than single parents. Using a scale of A comparison of parents educational level and
1 (never involved) to 5 (very often involved), involvement in their childrens education shows parents
married parents report checking whether their children are more engaged when their own educational level
have done their homework more frequently (M=3.51) exceeds their childrens current level of education
than single parents (M=3.37; p<.001). Among the 1,270 (Figure 8). However, parents less educated than their
single parents, 38.7% assist their children with children are more engaged within their childrens school
homework, but 44.8% of the 3,291 married parents environment than parents who have a level of education
provide more direct assistance with homework equivalent to or greater than their childrens. Less
(Figure 6). educated parents are more involved perhaps because
many would like to see their children attain higher edu-
Education Level cation than they did.

Before assessing the relationship between parents edu- Similarly, parents more educated than their children are
cation and engagement, we compare each childs current more engaged at home than those whose education is
educational level to their parents highest level of edu- equivalent to or lower than that of their children
cation. Almost two thirds of youth already have attained (Figure 9).

Figure 6. Marital Status and Assisting with Homework Figure 7. Childs Current Education Level Compared to
Parents Education
Never assist Sometimes assist
with homework with homework

20% Below child's

Unmarried 61.3% 38.7% education
Equivalent
44% to child's
education
Above child's
36% education
Married 55.2% 44.8%

0 20% 40% 60% 80% 100%

Figure 8. Educational Level of Parents Engaged within the Childs School Environment

Below child's
Volunteer at 21% education
9%
school Equivalent
14%
to child's education
28% Above child's
Attend school
13% education
events
19%

34%
Speak to teachers/
counselors 15%
24%

39%
Attend PTA 18%
meetings 30%
0 5% 10% 15% 20% 25% 30% 35% 40%

4
Using a 15 rating scale, parents with a university-level Childs Gender
education report talking to their children about what
they learn in school more often (M=3.95) than those who Overall, parents involvement is fairly consistent for
have completed senior high school (M=2.91), completed male and female children. However, parents are more
junior high school (M=2.62), or have no formal education involved at home than at school for female children
(M=2.36). Parents with a university-level education also (Figure 10) and more involved at school than at home for
report attending childrens school events more regularly male children (Figure 11). This gender difference may be
(M=3.09) than those with no formal education (M=2.20). a consequence of social norms that long favored active
Likewise, parents who are university graduates say they participation in school for boys and at-home schooling
make sure their children have done their homework for girls. This disparity is changing as Ghana attempts to
(M=4.17) more often than those who have no formal eliminate the gender gap in education (ICF Macro, 2010).
education (M=2.83).

Below child's Figure 9. Education Level

Assist with 26% education of Parents Engaged
10%
homework Equivalent to with Child at Home
7%
child's education
Discuss 35% Above child's
15%
school work education
22%
Ensure 39%
homework 18%
is done 26%

Talk about 41%

19%
expectations 31%

0 10% 20% 30% 40% 50%

Volunteer 50.2% At home Figure 10. Female Child

at school and Parental Engagement
52.3%
At school
Attend
50.3%
school
events 51.9%

Speak to 51.2%
teachers/
counselors 52.2%

Attend PTA 50.9%

meetings 52.1%

49.0% 49.5% 50.0% 50.5% 51.0% 51.5% 52.0% 52.5%

49.8% At home Figure 11. Male Child and

Discuss
school work Parental Engagement
47.7%
At school
Talk about 49.7%
expectations 48.1%

Assist with 48.8%

homework 47.8%
Ensure 49.1%
homework
47.9%
is done
46.5% 47.0% 47.5% 48.0% 48.5% 49.0% 49.5% 50.0%
5
Parents Gender and English. If this trend of non-significant relationships
between parental involvement and educational
We find slightly higher involvement at home among performance changes after the YouthSave intervention,
mothers (72.66% of 2,371 interviewed) than fathers we may learn new insight into how the intervention may
(71.6% of the 2,205 interviewed) (Figure 12). More fathers affect the impact of parental involvement on childrens
talk about their expectations (90.3%) than discuss school education.
work, assist with homework, or ensure homework is done.
Reflecting a contrary trend, 67.18% of fathers engage at Math Performance
school contrasted with 64.56% of mothers (Figure 13).
Overall, engagement at school is low contrasted with Overall, more parental involvement is not associated
engagement at home. significantly with better math performance (Figure 14).
Paradoxically, students whose parents make sure
Parental Involvement and homework is done perform slightly worse in math
(M=53.37, SD=16.87) than students whose parents do not
Academic Performance (M=53.55, SD=16.24).1 Similarly, children whose parents
interact with teachers and school counselors perform
Most empirical studies show children perform better in
worse in math (M=53.36, SD=17.10) than those whose
school when parents are involved (Fantuzzo, McWayne,
parents do not (M=53.45, SD=15.84). It is possible parents
Perry, & Childs, 2004; Nyarko & Vorgelegt, 2007; Toper,
are engaging with school authorities because they are
Keane, Shelton, & Calkins, 2010), but a few show that
concerned about their childrens already poor academic
parental involvement may not always be associated
performance.
significantly and positively with childrens educational
performance (Izzo & colleagues, 1999). For this research
brief, we compare all measures of parental involvement
English Performance
to childrens performance in math and English. Results When parents talk to their children about what they
vary, which supports the mixed findings from existing learned in school, childrens performance in English is
empirical studies. In the YouthSave Ghana Experiment significantly higher (t=2.21, p<.05) (Figure 15).
baseline data, most measures of parental involvement are Otherwise, there is no significant relationship between
not associated statistically with high achievement in math parental involvement and English performance.

Figure 12. Parents Gender and Engagement Figure 13. Parents Gender and Engagement
at Home within the School Environment

Some Some
engagement engagement
Never Never

100% 100%

80% 80%

71.60% 72.66% 67.18% 64.56%

60% 60%

40% 40%

20% 20%
32.82% 35.44%
28.40% 27.34%

0 0
Male Female Male Female

6
 Figure 14. Parental Involvement and Math Grade

53.5% No parent
Attend PTA meetings 52.6% involvement
Speak to teachers/ 53.4% Some parent
counsellors 53.5% involvement
53.7%
Attend school events 53.0%
53.5%
Volunteer at school 53.3%
Ensure homework 53.4%
is done 53.6%
53.5%
Assist with homework
53.3%
53.5%
Talk about expectations
52.9%
53.45%
Discuss school work 53.21%

52.0% 52.5% 53.0% 53.5% 54.0%

Figure 15: Parental Involvement and English Grade

53.8% No parent
Attend PTA meetings 53.1% involvement
Speak to teachers/ 53.6% Some parent
counsellors 54.0% involvement
54.0%
Attend school events 53.3%
54.0%
Volunteer at school 53.5%
Ensure homework 53.5%
is done 53.8%
54.1%
Assist with homework
53.5%

Talk about expectations 53.9%

52.2%
32.0%
Discuss school work 52.8%

30% 35% 40% 45% 50% 55%

Conclusion is still generally low but becoming more prevalent,

This research brief sheds light on the nature of Ghanaian especially regarding engagement within the school
parents engagement in their childrens education and environment.
socio-demographic factors (e.g., marital status,
Married individuals may be able to share the burden of
educational level, childs gender, parents gender) that
engaging with their child, but Donkor (2010) cautions
affect involvement. More Ghanaian parents are
that this relationship is complex and depends on whether
somewhat involved than absolutely detached from their
they are the childs biological parents and whether the
childrens education, but the extent of involvement is
child belongs to a polygamous family. Non-biological and
low overall. This finding supports Nyarko (2011) and Pryor
polygamous parents may have divided attention and be
and Ampiahs (2003a) finding that parental involvement
less likely to be involved in the childs schooling.

7
Findings from this study also reveal that only one Deshpande, R., & Zimmerman, J. (Eds.) (2010). Youth
measure of parental involvementtalking to children savings in developing countries: Trends in practice,
about what they learn in schoolis significantly and gaps in knowledge (A report of the YouthSave
positively associated with academic performance. Prior Consortium). Washington, DC: YouthSave Consortium.
studies have found similarly mixed results in the
relationship between parental involvement and Donkor, A. K. (2010). Parental involvement in education
academic performance. In some cases, childrens in Ghana: The case of a private elementary school.
academic performance is better when parents are International Journal about Parents in Education,
uninvolved. It is possible that behavioral factors not 4(1), 23-38.
controlled for in this analysis mediate the association
between parental involvement and performance in Elam, S., Rose, L., & Gallup, A. (1994). The 26th annual
school. Izzo and colleagues (1999) speculate that non- Phi Delta Kappa/Gallup Poll of the publics attitudes
significant and sometimes negative results could be toward the public schools. Phi Delta Kappan, 75,
caused by parents becoming more involved when 137-152.
children are not performing well in school.
Elliott, W. (2012). Does structural inequality begin with a
Knowing the level of parental involvement at baseline bank account? (Creating a Financial Stake in College,
will allow researchers to understand better the Report II). Washington, DC: New America Founda-
dynamics of YSA take-up and use. Are parents who tion; St. Louis, MO: Washington University, Center for
already are more involved in education at baseline also Social Development.
more receptive and supportive of YSA take-up? What
Fantuzzo, J., McWayne, C., Perry, M., & Childs, S. (2004).
impact could increased involvement of parents have
Multiple dimensions of family involvement and their
on the use of YSAs? These questions might be answered
relations to behavioral and learning competencies for
when we link data regarding parental involvement to
urban, low-income children. The School Psychology
youth savings at endpoint.
Review, 33(4), 467-480.
We know from previous research that parental
Georgiou, S. N. (2007). Parental involvement: Beyond
involvement is predictive of childrens educational
demographics. International Journal about Parents
outcomes (Nyarko & Vorgelegt, 2007; Topor, Keane,
in Education, 1, 5962.
Shelton, & Calkins, 2010). Thus, knowing the extent of
parental involvement at baseline will allow researchers ICF Macro. (2010). Millennium development goals in
to determine the independent impact of YSAs on youth Ghana: A new look at data from the 2008 Ghana De-
educational performance after accounting for factors mographic and Health Survey. Calverton, Maryland,
already known to improve educational outcomes. USA: ICF Macro.
Similarly, insight into the level of parental involvement
at baseline also allows researchers to assess how Izzo, C. V., Weissberg, R. P., Kasprow, W. J., & Fendrich,
take-up and use of YSAs influences parental support of M. (1999). A longitudinal assessment of teacher per-
and involvement in their childrens education. ceptions of parent involvement in childrens educa-
tion and school performance. American Journal of
Endnote Community Psychology, 27, 817-839

1. By Ghana Education Services standards, a 50% score is Miedel, W. T., & Reynolds, A. J. (1999). Parent involve-
average performance. ment in early intervention for disadvantaged chil-
dren: Does it matter? Journal of School Psychology,
References 37(4), 379-402.

Anderson, J. B. (2008). Social capital and student Nyako, K., & Vorgelegt, V. (2007). Parental involvement:
learning: Empirical results from Latin American A sine qua non in adolescents educational achieve-
primary schools. Economics of Education Review, 27, ment. Unpublished doctoral dissertation, Ludwig-
439-449. Maximilians University. Retrieved from http://edoc.
ub.uni-muenchen.de/8384/1/Nyarko_Kingsley.pdf
Catsambis, S. (2001). Expanding knowledge of parental
involvement in childrens secondary education: Nyarko, K. (2011). Parental school involvement: The case
Connections with high school seniors academic of Ghana. Journal of Emerging Trends in Educational
success. Social Psychology of Education, 5, 149-177. Research and Policy Studies, 2(5), 378-381.

Chowa, G., & Ansong, D. (2010). Youth and savings in Pryor, J., & Ampiah, J. G. (2003a). Understanding of Edu-
Assets Africa. Children and Youth Services Review, cation in an African Village: The role of information
32(11), 1591-1596. and communication technologies. London, Depart-
ment for International Development.

8
Scanlon, E. & Adams, D. (2009). Do assets affect well-
being? Perceptions of youth in a matched savings
YouthSave Research Partners
program. Journal of Social Service Research, 35(1), Washington University
33-46. George Warren Brown School of Social Work
Center for Social Development
Schimpl-Neimanns, B. (2000). Social origins and educa- Campus Box 1196
tional participation: Empirical analysis of relation- One Brookings Drive
ship between educational inequality between 1950 St. Louis, Missouri 63130-4899
and 1989. Cologne Journal of Sociology and Social
Psychology, 52(4), 636-669. University of Ghana
Institute for Statistical, Social and Economic Research
Schmitt, M., & Kleine, L. (2010). The influence of family-
(ISSER)
school relations on academic success. Journal for
Legon, Ghana
Educational Research Online, 2(1), 145-167.

Ssewamala, F. M., & Ismayilova, L. (2009). Integrating Kenya Institute for Public Policy Research
children savings accounts in the care and support of and Analysis (KIPPRA)
orphaned adolescents in rural Uganda. Social Service Nairobi, Kenya
Review, 83(3), 453-472.
New ERA
Topor, D. R., Keane, S. P., Shelton, T. L., & Calkins, S. D. Kathmandu, Nepal
(2010). Parental involvement and student academic
performance: A multiple mediational analysis. Jour- Universidad de los Andes
nal of Prev Interv Community, 38(3), 183-197. Bogot, Colombia

United Nations Capital Development Fund (2011). Lis- University of North Carolina
tening to youth: Market research to design financial School of Social Work
and non-financial services for youth in Sub-Saharan Chapel Hill, North Carolina
Africa. Retrieved on June 16, 2012 from http://mas-
tercardfdn.org/what-we-are-learning/publications/ Columbia University
youth-financial-inclusion School of Social Work
New York, New York
Acknowledgments
This brief is a product of the YouthSave project. Suggested Citation
Supported by The MasterCard Foundation, YouthSave Chowa, G. A. N., Ansong, D., & Osei-Akoto, I. (2012).
investigates the potential of savings accounts as a tool Parental involvement and academic performance in
for youth development and financial inclusion in develop- Ghana (YouthSave Research Brief 12-42). St. Louis, MO:
ing countries, by co-creating tailored, sustainable savings Washington University, Center for Social Development.
products with local financial institutions and assessing
their performance and development outcomes with local
researchers. The project is an initiative of the YouthSave
Consortium, coordinated by Save the Children in partner-
ship with the Center for Social Development at Washing-
ton University in St. Louis, the New America Foundation,
and the Consultative Group to Assist the Poor (CGAP).