Conferences and Workshops
Ask the Grade Doctor
Contact Me

Ask The Grade Doctor

Full Question Detail

Return to All Questions

 : : : :


Hi Ken,
Could you explain to me why it is not good practice to average a student's grade even though you are not using zeros? What about putting more weight on certain assignments and tests?



Sorry for the delay in responding but I have been busy finishing the second edition of the Repair Kit.
If by averaging you mean the mean it is fine if the student's performance is consistent but we teach in math that the mean is an inappropriate measure of central tendency when there are outlier scores and unfortunately students often have outlier scores - and they are usually low outliers. There are two other legitimate measures of central tendency that should be considered - median and mode, but the determination of grades should be based on the student's most consistent level of achievement with emphasis on the more recent. Please note that I said 'determination' of grades, not 'calculation.' For me grading is not simply a mechanical, numerical exercise, it is an exercise in professional judgement.
Teachers are not just bookkeepers, we are - or should be - professionals.


Roger Curtis's


Averaging--be it a mean--makes perfectly good sense in education. This is for a number of reasons not least of which is that the world--parents, employers, teachers and until recently students--understand what it means and how it is arrived at. Furthermore, it (mean) is a measurement that is entirely appropriate for those subjects such as Mathematics that do not repeat the same concepts from strand to strand.

The Grade Doctor's


You suggest that we should use the mean because everyone understands it but that is not
true. Most teachers and parents forget that they were taught in grade 5 or grade 6 that
the mean is skewed by outlier scores and that it is the wrong measure of central tendency
to use when there are outlier scores. Thus when students have outlier scores it is
mathematically wrong to use the mean and mathematically correct to use the median or

Jeffery Riles's


Outliers are only outliers if you do not know the source of the data to be from the same
population. A student's grades, high or low are for the most part valid data entries. In large
samples, a small number of outliers is to be expected (and not due to any anomalous

The Grade Doctor's


I disagree; if a scratch golfer shoots 90 it is an outlier; if I shot a 75 it would be an outlier. In the same way then if a student receives high grades on almost all the assessments on a similar body of knowledge and skills and a low grade on one assessment the low grade is an outlier. The outlier may result from the student having a bad day (ill, upset, etc) or from a bad assessment.