Chi squared test Flashcards Preview

Epidemiology and biostatistics > Chi squared test > Flashcards

Flashcards in Chi squared test Deck (6)
Loading flashcards...

What does the Chi squared test check for

association between 2 categorical variables. Calculates the frequencies that would be expected if no association->compares observed frequencies with expected values->if observed frequencies are very different=association


how to calculate the degrees of freedom for chi squared

(no. of rows-1)x(no. of columns-1)


Should you use frequency or % for chi squared test



Which estimates to use for 2x2 table in chi squared test

risk difference: p1-p2, use if actual size of difference is of interest
relative risk: p1/p2, use if relative difference is of interest, easier to interpret than OR but do not use for case control studies
Odds ratio: (p/(1-p))//(p2/(1-p2)) use for case control studies, approximately equal to relative risk when outcome is rare.Can be misinterpreted when outcome more common


Null hypothesis for chi squared

In the population where the samples come, there is no association between the 2 variables


Assumption of chi squared

large sample test, at least 80% of expected frequencies must be greater than 5. For 2x2 test, all frequencies>5