• Version
• Download 195
• File Size 567.36 KB
• File Count 1
• Create Date August 2, 2018
• Last Updated August 2, 2018

New Parameterization of the T Score Values: A Case Study of the Standardization of the Kenya Certificate of Primary Education (KCPE) Examination

Standardization of raw examination marks is a vital procedure in ensuring the validity of examinations results that is undertaken by assessment bodies. In Kenya, the Kenya Certificate of Primary Education examination (sat after the first eight (8) years of basic education) uses eight tests/papers that are standardized before they are combined to get the overall results. The current standardize score computed using fixed parameters of test mean of 50 and standard deviation of 15 and 12 for the objective/multiple choice papers and composition/essay papers respectively. The mean and standard deviation of raw marks in humanities, mathematics and science subjects are quite distinct and thus subjecting the marks to the same fixed parameters when standardizing raw marks leads to biasness that disadvantages candidates in mathematics and sciences subjects.The purpose of this study is to address the limitation of application of a fixed mean and standard deviation on the standardization of tests that have different performance characteristics that leads to disadvantaging examines in some subjects. Using data from the Kenya National Examinations Council, new parameters have been generated and applied in the computation of the standardized marks for different tests/papers in the years 2012 to 2014. The new standardized scores distribution are compared to that of the current standard scores distribution and applied on standard normal distribution curve.The study establish that the distribution of marks produces a better and valid standardize scores than the one that is currently used in Kenya with a fixed mean and standard deviation for all tests. The method is transparent and ensures that the standardize scores of high and low abilities candidates at both tails of the curve are not distorted to disadvantage/advantage one group of examinees.Keywords: Standardization, Standard Normal Distribution

Attached Files