A Singapore Government Agency Website

Official website links end with .gov.sg
Secure websites use HTTPS
Look for a lock () or https:// as an added precaution. Share sensitive information only on official, secure websites.
Sign In
Skip Navigation LinksLabour Market Statistics and Publications > Concepts > Mean Pitfalls
Mean Pitfalls
AVERAGES
Switch statistical concept
Using Mean, Median or Mode Mean Median ModeA Measure Enabling Comparison Absolute vs Relative Change Annualized Percentage Growth Distribution Incidence From Sample to PopulationSample Reliability Seasonal Adjustment Percentiles Quantitative vs Qualitative Range Ratios Stock and Flow

Pitfalls

The mean is sensitive to extreme values in the data.

It should therefore be interpreted with caution when the distribution is skewed, i.e. when there are a small number of extreme values in the group. If there are a small number of high earners in a list of observations, the mean monthly wage could be much higher than most of the values in the group. In this case, the mean might no longer be a good summary of the group’s “typical” wage behaviour. A better alternative would be the median.


The mean cannot be computed based on qualitative data.

Let’s say we asked individuals who have changed jobs whether their income in the new job was “higher”, “same as”, or “lower” than that in their old job, but did not ask for the ac​tual difference in income. Will we be able to compute the mean income change of these individuals? Regrettably, the answer is no. However, we can still find the most common option (the mode) and the “middle” response (the median) since “higher” can be ranked above “same as”, and likewise “same as” above “lower”.