inferential tests Flashcards
when do you use a binomial sign test?
when there is a difference between the data collected in two conditions, repeated measures design with nominal level data
sign test method?
- calculate the difference between the two conditions (a-b)
a+ for positive differences
a- for negative differences
leave any 0 differences between conditions black and removed from total participants (N) - add up the number of +signs and -signs
whichever group (+ or -) is smaller, is S
in a sign test, the observed value of S must be _________ or ________ than the critical value to be significant
equal to or less than
when do you use mann whitney U test?
when there is a difference between the two conditions, IMD with ordinal level data
mann whitney U test method?
- if one condition has LESS participants, call this condition A
- rank all the scores together, as if they belonged to just one group, but record the ranks for each group separately
- find the sum of the ranks for each condition (R1 and R2)
- calculate U1 and U2 using the formula
- U is the smaller of U1 and U2
in a mann whitney U test, the observed value of U must be _________ or ________ than the critical value to be significant
equal to or less than
when do you use a wilcoxon test?
when there is a difference between the data obtained in two conditions, RMD with ordinal level data
wilcoxon test method?
- work out the difference between the scores in the two conditions (a-b)
- rank the differences from smallest to largest
- ignore any + or -, and ignore any 0 (we take this from N)
- find the sum of the ranks of the positive differences
- find the sum of the ranks of the negative differences
- T is the smallest of these 2 values
in a wilcoxon test, the observed value of T must be _________ or ________ than the critical value to be significant
equal to or less than
when do you use a chi squared test?
when there is a difference between the data collected, data is nominal
chi squared method?
- calculate expected frequencies (if they are not already done for you) by doing E = R x C / T
E = row total x column total / grand total - step by step table
- cell ( a b c d)
- observed frequencies from contingency tables
- expected frequencies
- O - E
- (O - E) squared
- (O - E) squared / E - degrees of freedom (R-1) x (C-1)
in a chi squared, the observed value must be _________ or ________ than the critical value to be significant
equal to or greater than (rule of R)
when do you use spearman’s rho?
when there is a correlation and ordinal level data
spearman’s rho method?
- rank the values of the first variable, then rank the values of the second co-variable
- find the difference between the ranks
- square the differences of the ranks
- add up the square of the differences
- insert this into the equation for rs
in a spearman’s rho, the observed value of rs must be _________ or ________ than the critical value to be significant
equal to or greater than