pre-posttest parametric

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

pre-posttest parametric
Message posted by margaret on September 18, 2000 at 12:00 AM (ET)

Students have gone to full day and half-day kindergarten. Their pre-and post test scores are abnormally distributed...how can i compare full and half-day k with each other (four tests)?????- thanks

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: pre-posttest parametric
Message posted by JG on September 19, 2000 at 12:00 AM (ET)

Try the wilcoxon rank sum test for differences in two medians, comparing the scores two a time to start with. See page 481 in 'basic business statistics' by berenson and levine.

Re: pre-posttest parametric
Message posted by JG on September 19, 2000 at 12:00 AM (ET)

For multiple comparisons of medians try pages 549 or 574 in berenson and levine.

Re: pre-posttest parametric
Message posted by DL on September 19, 2000 at 12:00 AM (ET)

It looks to me like a paired test for means.

You substract the value after and before and check the difference is equal to 0 (In some statistical packages to it for you).

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.