Вы находитесь на странице: 1из 2

Chi Square Test for Independence

-Chi-square tests are applied when you have 2 categorical variables (is not
measured).
-It is used to determine whether there is a significant association between 2
variables
-eg: in a random sampling of a habitat, if the presence/absence of more
than 1 species is recorded, then it is possible to test for association between
these species
-Positive association: if both species are found at the same time
- Negative association: if both species tend to occur apart
-The distribution of the populations can also be independent from each other
-Can only be used if the expected frequency count is 5 or larger

-There are 2 possible hypotheses


-null (Ho): 2 species are distributed independently
-Alternate (Hl): 2 species are associated

Method
-First, take the raw data and construct a contingency table
Species A Present

Species A Absent

Row Totals

Species B Present
Species B Absent
Column Totals
-Calculate the expected frequencies for each of the combinations (assuming
independent distribution)
Expected Frequency = (row total x column total)/grand total

-Calculate degrees of freedom


df = (m-1)(n-1)
-m = number of rows

-n = number of columns

-Find critical value on a chi-square table (significant level P=5%)

-Calculate the chi-square value for data

-O=observed frequency
-E=expected frequency
-If the calculated value is greater than the critical value, then there is
evidence for an association between species: reject null hypothesis
-If the calculated value is less than the critical value, then accept null
hypothesis

Вам также может понравиться