School performance indicator on improving student learning achievement through O-NET scores: Factor analysis on student prior achievement and size of the school

Abstract

     The Ordinary National Educational Test (O-NET) is a mandatory test for K6, K9, and K12 students in Thailand. This ex post facto research conducted quantitative data analysis on all individual O-NET scores and their previous-level O-NET scores, three years earlier (that is, K6 & K9 and K9 & K12 of the same students) to reflect the quality of the schools (during K7–9 and K10–12 respectively); particularly with regard to learning achievement. The analysis included Pearson χ² between K6 and K9 O-NET scores and between K9 and K12 O-NET scores. The K6 & K9 correlation was low (χ² = 0.12, n = 631,770) and the K9 & K12 correlation was medium (χ² = 0.35, n = 411,238), which revealed that there are other factors beyond the previous-level outcome that greatly affect the learning achievement measured by the O-NET scores (88% during K7–9 and 65% during K10–12). Furthermore, the single-factor analysis of variance on the size of the schools showed that the change in the standardized O-NET scores (Z score) in each subject between the consecutive levels differed significantly between schools of different sizes (F > 46, p < .001). However, these changes from eight subjects showed no single trend, that is, extra-large schools were not always the best. As a result, both the student prior ability and the size of the schools should not be used as limitations or excuses for not improving the learning outcome. However, the quality of each school should not be assessed by the O-NET scores of the graduates alone, as currently, it is much fairer for the schools to be assessed by the change of the standardized O-NET scores between the consecutive levels (K6 & K9 or K9 & K12) of each class year.