Chi Square test

Chi Square test

• difference between observed and expected data— a statistical procedure

• whether it correlates to our data’s categorical variables

• The difference between 2 categorical variables is due to the relationship between them

• One sample or Goodness-of-fit test (person’s chi-square goodness-of-fit test)

• Chi-square test of association

types of chi square test

• One sample or Goodness-of-fit test (person’s chi-square goodness-of-fit test)

• determine whether the distribution of cases in a single categorical variable follows a known or hypothesized distribution

• 1 sample only

• Chi-square test of association

relationship of 2 categorical variables

eg: gender and coffee intake (effect)

• variables should be measured at a nominal or ordinal level

• 5 expected frequencies in each group of your categorical variable

• Groups of a categorical variable must be mutually exclusive (color is chosen, data should also be color)

• All observations are independent (one categorical variable, will yellow affect red? no)

assumptions

the sample data follow the hypothesized distribution

hypothesis of one-sample goodness fit

there’s no significant association between the 2 variables

chi-square test of association hypothesis

reject the null hypothesis if the p-value is less than a

decision rule