Stop the ‘mean score’ mania and let middle candidate rank school

Standard Eight candidates sit for the mathematics paper on the start of KCPE exams on October 31, 2017. The 2018 candidates will be briefed on Monday, October 29, 2018 by examiners on what will be expected of them when they start their Kenya Certificate of Primary Education examinations Tuesday. PHOTO | ONDARI OGEGA | NATION MEDIA GROUP

What you need to know:

  • So, their mean scores should not be compared with other schools admitting students who scored less marks.
  • It is expected that fewer candidates attain very high scores with a similar proportion on the other extreme.

The recent rejection of newly posted principals perceived to be unqualified to head ‘good’ schools was based on the historical ‘mean scores’ in the national examinations at their previous stations.

This should jolt us to the reality that we may have placed the mean score on a pedestal over the years while overlooking how it may have been attained.

But is a mean score the best criterion to rank a school or candidate? What, really, is a ‘good grade’ or ‘good school’?

RESULTS

Exam results are usually ranked from the highest- to lowest-scoring candidate but an average score may be measured as a mean, median or mode. The mean score is obtained after the total number of marks scored by all the students are put together and divided by the total number of candidates.

The mean is easily manipulable by tipping the scale either way. To achieve a high mean score, some schools take advantage of the variability of the mean as a measure of central tendency to have as many candidates as possible passing as few or none fail.

In this scenario, the most common score, which is the mode, is also as high as the mean because majority of the students will have done well and the middle student, who scores the median score, will also have as high a score.

For decades, national schools commonly referred to as the “academic giants” have selected students with the highest CPE or KCPE scores in their districts and set a high cut-off point for admission limited by the number of places available.

FORM ONE

So, their mean scores should not be compared with other schools admitting students who scored less marks.

Some principals in the ‘good’ schools have gradually segregated high scorers into an alpha or elite class, sometimes on entry at kindergarten or Form One.

These would get preferential attention by the teachers while they ignore the other, ‘average’ students.

The ostensibly high sense of achievement can only be authenticated by delinking pupils’ KCPE mean scores with their admitting schools during Form One selection and doing away with the school categorisation as “national” or “county” and classify them as boarding, day or special, regardless of State support.

Let all KCPE candidates be admitted to the schools near their homes to lighten households’ education cost burden and dampen the motivation to cheat in exams while allowing county administrations to objectively improve the quality of education in their schools.

We need to reflect on the hidden effect of this mean score mania on our candidates in the next three decades, not on schools or head teachers.

CANDIDATES

Most candidates may suffer depression or loss of self-esteem when the results are released publicly and sensationalised by the media, creating a sense of hopelessness that may trigger high-risk behaviour.

Parents and teachers associations should objectively review the trend in their school’s exam results over the past few decades. If the middle (median) and mean (average) score differ, then the results may not be a true reflection of the school’s performance.

The results should confirm a normal distribution. It is expected that fewer candidates attain very high scores with a similar proportion on the other extreme.

The majority will attain average scores.

If the marks scored by the middle candidate on each list is close to or the same as the mean score for each year, and if there are as few students scoring extremely high scores as have scored extremely lowly, then the results are authentic.

Scoring a school based on the median would be the true reflection of the efforts of the candidates and how effectively the teachers have delivered the curriculum for that level.

Dr Mwendwa, a clinical epidemiologist, is a lecturer in the Department of Medical Physiology at the University of Nairobi and the interim vice-chair, the Public Health Society of Kenya (PHSK). [email protected].