3003PSY Mini Lecture 3 Chi-Square Assumptions PDF

3003PSY Survey Design and Analysis in Psychology CHI-SQUARE ASSUMPTIONS ASSUMPTIONS OF c2 a participant must fall in one and only one category – they cannot be counted twice independence of observations thu...

3003PSY Survey Design and Analysis in Psychology CHI-SQUARE ASSUMPTIONS ASSUMPTIONS OF c2 a participant must fall in one and only one category – they cannot be counted twice independence of observations thus you cannot use chi-square on repeated measures designs special forms of tests exist to test out such relationships but these are well outside the scope of 3003PSY  usize of expected frequencies uall expected cell frequencies should be at least 5, if they aren’t you shouldn’t use a c2 test ASSUMPTIONS usmall expected frequencies produce few possible values of OF c2 chi-square obtained c2 but we compare to continuous distribution uthe greater the degrees of freedom, the more lenient this requirement ASSUMPTIONS uInclusion of non-occurrences OF c2 ucomputations must be based on all participants in the sample INCLUSION OF NON-OCCURRENCES EXAMPLE u You run a mirror self-recognition study to test gender differences in this important developmental milestone. This is a feline version of the test…. INCLUSION OF NON-OCCURRENCES EXAMPLE You run a mirror self-recognition study with 40 2-year-olds (20 boys and 20 girls). 17 girls pass the test but only 11 boys. Is there a significant gender difference? Girls Boys Total Observed 17 11 28 (Expected) (14) (14) c2 = 1.20; ns BUT!!!! this is a violation of ‘inclusion of non-occurances’ INCLUSION OF NON-OCCURRENCES EXAMPLE You need to include all of the data you collected (including the children that did NOT pass the test). Girls Boys Total Pass 17 11 28 Fail 3 9 12 Total 20 20 40 c2 = 4.29, p