Apologies but this is a very basic, probably naive question from someone with little to no comprehension of statistical ideas.

I need to interpret some results that use statements about the SD as a measure. These are from a couple of statistical surveys on the effects on test scores of laptop use in classrooms

What does it mean to say:

Example (1)

"...student test scores improve by 6.41% of a standard deviation"

Sample size: 130,482

Example (2)

" [The results show that the use of laptops] ... reduces final exam scores by 18% of a standard deviation"

Sample: size 726

Any guidance appreciated. Thanks


They are using "standard deviations" the way you might use another unit, like "inches" or "years".

They don't specify what they mean by "student test scores", but in this context they are likely referring to "average student test scores".

Statement 1 might be read as (my additions in bold):

[average] student test scores improves by 6.41% of the standard deviation of student test scores

And statement 2:

reduces [average] final exam score by 18% of the standard deviation of exam scores.

| cite | improve this answer | |
  • 1
    $\begingroup$ I agree with this guess. I'd add that the implication appears to be that the original standard deviation is a benchmark. Statements like that are likely to be poor ways to think about the data if the standard deviation did not itself remain about the same. $\endgroup$ – Nick Cox Jun 26 '16 at 13:31
  • $\begingroup$ Thanks Nick. I'm a real novice when it comes to stats so this is another dumb question: in general, is it safe to say that 6.41% of a SD is a weaker effect than 18%? $\endgroup$ – DJL Jun 26 '16 at 13:40
  • $\begingroup$ @DJL I would say "smaller" instead of "weaker." It's just a size comparison; the idea of effect strength tends to imply different things to different people. $\endgroup$ – shadowtalker Jun 26 '16 at 14:28

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.