Chi squared test

Covers Chi squared goodness of fit and test of association

What is Chi squared

An inferential stats test to determine whether there is significant association between nominal categories

What is chi squared written as

X²

What data does Chi squared compare

Observed data and expected data

What is the test output for Chi squared

a test statistic and a p-value

What happens to the test statistic in chi squared

It is compared against a critical value

What are the two types of Chi-square data

Goodness of fit and Test of association

When is goodness of fit used

If you have one variable and data fits a theoretical distribution

When is Test of association used

if you have 2 variables and they are nominal and related

What do both chi squared tests test

How far from an expectation data deviates

What type of hypothesis is given for a Chi squared test

Non directional alternative

What are the assumptions of Chi-squared

• Types of data

• Mutual exclusivity

• Expected frequencies

• Independence of observations

• Random sampling

What type of data is used in chi squared

Nominal or ordinal

What is the expected frequencies assumption

The expected frequency is +5 per cell

What test is done if the expected frequencies assumption is violated

Fisher’s exact test

What does the goodness of fit test tell you

Whether the overall distribution differs from what is expected

How is expected frequency calculated (goodness of fit)

Split the sample evenly across all outcomes

What does a larger X² score mean

More deviation

How do you know if a score is large or small, chi squared

Compare your value in a critical table

What p-value is typically used for chi-squared

0.05

What is the Df (goodness of fit)

number of categories - 1

What does it mean is the test statistic is greater than the critical value

you reject the null hypothesis

What does it mean is the test statistic is less than or equal to the critical value

Fail to reject the null hypothesis

What type of tables does goodness of fit use

frequency tables

How is chi-squared written up

X² (df, N = XX) = XX.XX, p = .XX

What is the test of association

A chi-squared test used when there are 2 categories

What does the test of association say

Whether there are is a significant association between two categorical variables

How is expected frequencies calculated for test of association

marginal totals in the contingency table are summed for a whole table, and the frequency for each cell is calculated based on (row total x Column total) / grand total

Whats the calculation for test of association expected frequency

(row total x Column total) / grand total

What type of tables does test of assocaition use

contingency tables

What is the max amount of variables a chi squared test can cope with

2