Page 197 - Contributed Paper Session (CPS) - Volume 4

CPS2173 Felicien Donat Edgar Towenan Accrombessy et al.
total variance of the end-of-school score in reading and mathematics. The
models are globally significant and the variables are introduced in a
progressive process. The main findings are: Repeaters are less efficient than
non-repeaters: Repetition hurts the performance of the education system, in
addition to the fact that it costs for the whole system (parents and authorities).
Repeating students’ progress less quickly than others during the year (PASEC
2004-2005). The results show that pupils who repeat at least once are
significantly less effective at the end of their schooling than those who have
never repeated according to the output of the model estimated on the 2014
data. Older students are less successful in their learning. The results point out
that older students are at a disadvantage in their learning: 2004 and 2014
results indicate that students' ages are significantly and negatively associated
with their academic performance in reading and mathematics. The older the
students, the less they assimilate the concepts. There are also inequalities of
performance by gender of student: The relationship between student gender
and school performance varies by the PASEC year considered. According to
the 2004-2005 PASEC data, under the control of other explanatory variables,
girls are on average less performing than boys in language-reading and
mathematics. In 2014, the trend is reversed: inequalities of performance in
reading and mathematics are in favor of girls at the end of their schooling.
This observation could be explained by the policy in favor of girls set up
throughout the country. In addition, girls in urban area outperform than those
living in rural area. The fact that girls are taught by women has no significant
effect on their acquisitions. Then, the socioeconomic status of families is
positively related to student achievement. The standard of living positively
affects the level of students at the end of the year. The interaction between
the socioeconomic level of families and the location of schools is negatively
linked to school attainment. Also, in relation to school performance, the results
of the linear hierarchical model estimates show significant differences in scores
to the detriment of students who do not speak the language of instruction
(French) in their family environment. As other studies of education in Africa
show, the practice of the language of instruction enables children to acquire
language skills before they enter school and during primary school cycle.
More, students' performance at the beginning of the school year is positively
correlated with the end-of-school scores: Students with one point of
difference in their initial levels are, all other things being equal, with a
difference of 0.55 at the end of the year. Since the coefficient is positive and
less than unity, this indicates that learners who are most successful in their
acquisitions remain those who have a high level at the beginning of the year
and that the gap between students is narrowing towards the end of the year.
Otherwise, the participation of students in the field work, trade and physical
work affects negatively their learning. These out-of-school activities are often

186 | I S I W S C 2 0 1 9