I'm looking a single adjective or a short adjective phrase that refers to values that are either above or below an average. Consider this sentence:

Participants with scores above or below the average will receive special instructions.

I'd like to replace the that clause by a modifier structure EXPR:

Participants with EXPR scores will receive special instructions.

Roughly speaking, the expression should refer to values that fall within the ranges shaded blue in the following figure:

Target ranges

As you can see, the overall distribution of values may follow a bell curve, with an average of about 50 (dashed line). EXPR should apply to values in the blue-shaded areas, i.e. scores that are not between 40 and 60. Statistically speaking, more than 40 percent of all participant scores are in that area, so apparently I'm not really talking about participants with an unusual or unexpected score, or even an extreme or excessive score – these are participant scores that are just not average.

Any suggestions what I could use as EXPR?


Average responses refer to a "typical" response. For that reason, I would use the term "atypical."

"Values falling outside one standard deviation demonstrate atypical performance in our experiments."

  • 'Typically' values are recorded between 10% and 90%, with the median at 50%. If the Bell curve were skewed then that would be 'atypical'. – Nigel J Mar 13 '18 at 15:09
  • @NigelJ Yes, when referring to the shape of the curve. I was referring to the individual responses, per se. – Stan Mar 13 '18 at 15:40


As in:

"Participants with outlying merriam scores will receive special instructions."



an observation that is well outside of the expected range of values >in a study or experiment, and which is often discarded from

the data set: Experience with a variety of data-reduction problems has led to several strategies for dealing with outliers in data sets. a person whose abilities, achievements, etc., lie outside the range of statistical probability. dictionary.com

  • An outlier has a more specific meaning, referring to something that is statistically unexpected. Results that merely lie on the narrower portions of the normal curve are not outliers. – Robusto Mar 13 '18 at 15:21
  • @Robusto as in: 'This unit uses data analytics to find outliers and the tell-tale signs of crime—like who is prescribing the most drugs, who is dispensing the most drugs, and whose patients are dying from overdoses.' In statistics one looks for the expected ... and unexpected. – lbf Mar 13 '18 at 15:34
  • I understand that. My point is that the OP is asking about things that are expected, and predicted by a normal curve, so outlier is simply not the word to use here. – Robusto Mar 13 '18 at 16:02
  • @Robusto those in blue outlie those in white. – lbf Mar 13 '18 at 16:11
  • Now you're indulging in sophistry. – Robusto Mar 13 '18 at 16:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.