# Non Parametric Tests - Methods

notes on non-parametric tests for methods

a2 psychology with aqa

HideShow resource information
• Created by: Cara
• Created on: 29-06-12 13:43

First 245 words of the document:

Non-parametric tests
Inferential statistics can be broadly separated into two groups:
Non-parametric tests ­ the sign test; Wilcoxon signed ranks test; Mann-Whitney; Chi square;
Spearman's correlation
Parametric tests ­ related and independent t-tests; Pearson's moment correlation
Choosing and using
When using a test we need to consider:
The hypothesis
The experimental design
The level of data
We use a non-parametric test of difference when:
It can be a related or independent design
Data is usually nominal or ordinal
There is no homogeneity of variance (this means that data could be interval)
Non-parametric tests ­ tests of difference
Sign test
o At least nominal data
o Used with a related design
o N = the number of participants whose data values changed in either direction
o S is significant if it is equal to or less than the critical table value
Wilcoxon
o At least ordinal data
o Used with a related design
o N = the number of pairs of scores
o W is significant if it is equal to or less than the critical table value
Mann-Whitney
o At least ordinal data
o Used with an unrelated (independent) design
o N1 and N2 = the number of scores in each condition
o U is significant if it is equal to or less than the critical table value