hypothesis tests conditions

binomial distribution conditions

• a fixed number of trials

• only two possible outcomes - either success or failure

• the probability of success (p) is the same for each trial

• each trial is independent

normal distribution conditions

• data should be symmetrical

• mean = median = mode

• data should produce a bell shaped curve

two sample propportion test conditions

• both populations follow a binomial distribution

product moment correlation coefficient conditions

• the data needs to be bivariate normally distributed i.e. both variables are normally dstributed

spearmans rank test conditions

• unlike PMCC, the data is not required to be bivariate normally distributed

contingency table - χ² test conditions

• data sample must be randomly selected

• if any of the expected values are less than 5 then you must pool that whole row/column with another sensible row/column

z test conditions

• the data sample must be 30 or bigger

• the probabiity must be close to 0.5

• the population variance is known

t test conditions

• used when population variance is unknown

• the population must be normally distributed

sign test conditions

• data sample must be randomly selected

wilcoxon signed rank test conditions

• the data must be symmetrical

wilcoxon rank sum test conditions

two random samples where the populations cannot be assumed to be normally distributed