Examination Standardization explained

The 2016 KCSE results have continued to elicit mixed reactions from different stakeholders in the education sector with the latest being the Kenya National Union of Teachers (Knut) that has demanded the results for all the 574,125 candidates who sat last year’s Kenya Certificate of Secondary Education (KCSE) examinations recalled immediately.

The Kenya National Union of Teachers (Knut) said that due process was not followed in marking and releasing of the 2016 results. The union leadership has added that the results released last month do not reflect the true performance of the candidates, citing clear breaches of marking processes that were overlooked by the Kenya National Examination Council (KNEC).

When the Education CS released the 2016 examinations in Mombasa, he said that the results were credible and honest. In the letter to the Clerk of the National Assembly Justin Bundi and J.M Nyegenye of the Senate, KNUT demanded the immediate recall of KCSE results, and be taken through a due process of ‘moderation and grading’ appropriately by chief examiners at the subject level.

The union argues that the examinations were not moderated during marking, claims the marking scheme was not discussed and adopted by the examiners.

The Union further claims that one uniform grading system was applied in all subjects is a disaster in itself and an action that must be reversed. It also argues that this has disadvantaged the mathematics and science students thereby disadvantaging male candidates who traditionally in many schools opt to take three sciences and mathematics.

Examination Standardization, however, is a process that very few people understand, hence the resultant confusion. Examination standardization is a statistical process that is specifically designed to remove variable elements from test scores and allow the candidates to be compared equally. In other words, it is a way of giving equal value to the results of each test, regardless of the number of questions and the time allocated for completing them. For instance, an A in Mathematics cannot have equal value with an A, say, in Religious Education.

Standardized scores are more useful measures than raw scores (the number of questions answered correctly). This is done in the world for three main reasons;

Examination standardization places the candidate’s scores on a readily understandable scale. Test scores are not made readily understandable by just converting them to percentages because percentage on their own are not related to the average score of all the candidates or on how their scores are spread-. On the other hand.

Tests are usually standardized so that the average, nationally standardized score automatically comes out as 100, irrespective of the difficulty of the test and so it is easy to see whether a candidate is above or below the national average.

The measure of the spread of scores is called the ‘standardized deviation’ and this is usually set to 15 for educational attainment and ability tests, and for many occupational tests.

This means that irrespective of the difficulty of the test, about 68 percent of the candidates in the national sample will have a standardized score within 15 points of the average ( between 85 and 115), and about 96 percent will have a standardized within two standard deviations (30 points) of the average ( between 70 and 130). These examples come from a frequency distribution known as ‘the normal distribution’.

Description: https://www.nfer.ac.uk/images/standardiseddist.gif


Examination standardization is also done so that scores from more than one test can be meaningfully compared or added together. Standardized scores from most educational tests range from 70-140. Hence student’s standing in say, Mathematics and English can be compared directly using standardized scores. It is not, therefore, meaningful to add together raw scores from tests of different length or difficulty.

These, and many other factors that are unknown to the public – the relative difficulty of the exam, the ability of the cohort, among others, allows for the calculation of a standardized score directly from a raw score.

Calculation of a Standardized Score

In order to create a standardized score, a reference table called a ‘look-up table’ is created for each test paper that is written and the table is specific to that test paper because it takes account of the difficulty of the paper.

The minimum standardized score is derived from the look-up table and the actual number will vary, depending on the average score of all those taking the test.