Thu | Aug 16, 2018

More analysis needed on academic scores

Published:Friday | April 3, 2015 | 3:26 AM

I write in response to the article printed in the March 31, 2015 edition of The Gleaner titled 'The CSEC grade: time for a change'. The article builds itself on analysis of recent CSEC results that identifies all-girl schools as the top performers, with the recommendation that Jamaica's boys need help.
According to the Ministry of Education document: 'School Profiles 2012-2013', there were some 15 all-girl schools across Jamaica, and seven all-boy and 148 co-ed ones. This means that, on average, girls' schools will have twice the chance of outperforming boys' schools because there are simply more than twice as many. But we must be careful of such casual analysis as the relevant ceteris paribus, or all other things equal, condition may not obtain.
Where the CSEC grade report falls critically short of, in order to validate this cry for help for our boys, is in providing value-added scores of the schools, ie, accounting for the movement from GSAT grades to CSEC grades. Such statistics will be vital in order to gauge which schools are more efficient at progressing rather than just maintaining pupil attainment. 
We demand, too, much of the data if this information is not present; and the result is that we may end up comparing apples and oranges since the boys' and girls' schools may not be suitably matched as it relates to grade profiles of the intakes from GSAT. Statistically, the best thing would be to look at the entire population of Jamaican students taking the CSEC exams and to determine if male performance is statistically different from female performance; and then to investigate whether males in single-sex schools do better than those of similar ability elsewhere. This would then provide valuable information to determine whether teaching styles and learning styles for boys may be a meaningful talking point in the discussion. I humbly submit that while the CSEC grade report is moving us in that direction, the analysis is not yet sufficiently detailed to facilitate the 'call to action', even though such a call may indeed be borne out by further research.
Ryan Palmer
Chislehurst, Kent