109

attempted to answer this question. It is found that in PISA 2012, differences in students’ socio-

economic backgrounds including family wealth do not account for the performance difference

between Malaysian and Singaporean students. The same conclusion also holds when compared

to South Korean students. In contrast, children from Vietnam perform at the same level as those

from Korea after accounting for socio-economic differences (Asadullah, Pereira and Xiao 2017).

This implies that economic development, poverty reduction, and income growth alone will not

close the learning shortfalls between Malaysia and other high performing East Asian countries.

3.2.6.

Regression Analysis of the Determinants of Learning Outcomes

Educational production function analyses the relationship between inputs and outputs. In this

context, education production function is usually used to determine the relationship between

school level factors that determine students’ performance. School level inputs can be class size,

teacher quality, school resources such as library, computer, teaching and learningmaterials. The

commonly used output is students’ achievement (Pigott, Williams, Pollanin & Wu-Bohanan,

2012). According to Mat Saad, Nik Yusoff & Mohammad Yassin (2001), the classroom

environment plays an important role in providing a convenient and conducive learning

environment. Small classrooms with overcrowded students and inadequate facilities make it

difficult for the learning process (Tanner & Lackney, 2006).

Conducive learning environment, smaller class size, quality and effective teachers have been

commonly cited in policy documents and literature as determinants of students’ performance.

However, recent emphasis has been on improvement on teaching quality, teaching and learning

on higher order thinking skills, promoting school culture as learning organization, school

leadership, parental commitment and encourage private sector involvement. In this section, we

discuss trends in some of these indicators based on available data and education statistics.

Table 3.2.4

presents ordinary least squares (OLS) estimates of the student achievement

function for Malaysia in Reading, Math and Science in PISA 2012 data where achievement is

examined in relation to individual, family and school factors.

The estimationmethods accounts

for multiple plausible values since PISA data does not report a single test score for study

subjects. Among household-specific factors, a number of results are noteworthy. Student

achievement is most sensitive to family wealth in case of mathematics scores – compared to

children from bottom 25% of the wealth distribution, children in the top 25% wealth group

enjoy an extra 44 points in PISA math score, which equivalent to nearly one extra year of

schooling. Equally, children of educated parents perform significantly better compared to those

whose parents have only lower secondary education or below. Second, while there is no

advantage to attending a private school, experience of pre-school attendance is significantly

associated with higher performance in all PISA subjects.

Among individual level factors, one notable finding is the female advantage in science and

language and the absence of any gender gap in mathematics in Malaysia. In other words,

compared to many other parts of the world where girls lag behind boys in educational

achievement, they excel in all domains of learning in Malaysia. However, the girl-boy gap in

Reading is very high and is a concern. If test language is spoken at home, this positively

influences mathematics and science scores though the correlation is negative in case of reading

score. This is important considering the fact that a large proportion of Malaysian students don’t

In the PISA 2012 round, 5197 students from 164 Malaysian schools participated.