Chi Square Flashcards
What constitutes a chi square goodness of fit
If you have one independent variable then a chi squared goodness of fit is used.
What constitutes a chi squared test of association
When you have more than one independent variable.
What are the assumptions of chi square
The data must be nominal
The data must consist of frequencies
The data must be unrelated
No more than 20% of cells should have a expected frequency of <5.
How would you work out the expected frequencies
Number of times/people divided by Number of cells
How do you determine if the null hypothesis is correct
The total proportionate differences between the observed and expected frequencies will be small
What is the equation for Chi-Sqaured
X2 = (o - e)2 divided by e o = observed frequencies e = expected frequencies
What is meant by degrees of freedom
How many items in a data set that need to be specified before all items are known.
DF = k-1 where k = number of cells.
What can be determined if the results from a Chi square test of association are significant.
If the results are significant than you can say that there is a relationship between the variables.
What is the formula for the expected values
E = RC divided by T R = row total for cell C = column total for cell T = overall total e.g. Number of participants
