inferential statistics tests

5 tests

• mann whitney u test

• wilcoxon signed ranks test

• chi square test

• binomial sign test

• spearmans rho correlation

if a result is significant then

null hypothesis is rejected / alternate hypothesis is accepted

mann whitney u test criteria

• ordinal data

• independent measures design

mann whitney u test calculation steps

• rank scores (from lowest to highest)

• use formula to find u

• calculate the smaller u

• formula is given in exam

mann whitney u test significance level

• calculated u value is less than the critical u value, then result is significant

• calculated u value is more than critical u value, then result is not significant

wilcoxon signed ranks test criteria

• ordinal data

• repeated measures design

wilcoxon signed ranks test calculation steps

• difference in values of each condition is calculated

• differences are then ranked

• sum of the positive and negative differences is found

• T is the smaller of these values

wilcoxon signed ranks test significance level

• calculated w is less than critical value then result is significant

• calculated w is more than critical value then result is not significant

chi square test criteria

• nominal data

• independent measures design

chi square test calculation steps

• add totals for each column

• calculate observed and expected frequencies (using formula - given in exam)

• chi value is found by adding all cells of expected frequencies

chi square test significance levels

• calculated chi value is less than critical chi value, then result is not significant

• calculated chi value is more than critical chi value, then result is significant

binomial sign test criteria

• nominal data

• repeated measures design

binomial sign test calculation steps

• determine positive and negative values for data

• add each positive and negative assigned direction

• smallest direction score is your s value

binomial sign test significance level

• calculated s value is less than critical s value, result is significant

• calculated s value is more than critical s value, result is not significant

spearmans rho correlation criteria

• variables produce at least ordinal data

• exploring relationship between 2 co variables

• correlation design used

spearmans rho correlation calculation steps

• rank each data set individually

• find different between each data set

• square difference (d²) and find total of d² column

• use formula to find rho - given in exam

spearmans rho correlation significance level

• if calculated rho value is less than critical rho value then result is not significant

• if calculated rho value is higher than critical rho value, result is significant