The independence test Chi-square allows us to determine whether there is a relationship between two categorical variables. It should be stressed that this test tells us whether there is a relationship between variables, but does not indicate the degree or type of relationship, that is, does not indicate the percentage of influence of one variable on another variable or causing influence (Greenwood & Nikulin, 1996). A chi-square test is used with you have variables that are categorical rather than continuous.
The data I created is on the basis of questions ask by 54 respondents that how they enjoy the sports of tennis. Either by watching on TV, watching at the stadium or by playing themselves. After collecting the data, I used SPSS and run Chi-Square test.
Mode
Number of Respondents
Play
13
Watch on TV
17
Watch at stadium
24
Chi-Square Test
Frequencies
Number who preferred mode of enjoying tennis
Observed N
Expected N
Residual
Play
13
18.0
-5.0
Watch on TV
17
18.0
-1.0
Watch at stadium
24
18.0
6.0
Total
54
Test Statistics
Number who preferred mode of enjoying tennis
Chi-Square
3.444a
df
2
Asymp. Sig.
.179
a. 0 cells (.0%) have expected frequencies less than 5. The minimum expected cell frequency is 18.0.
Weight data file by number of cases
The data in this exercise were already weighted because the number of people preferring the mode of enjoying tennis was provided. However, to assure proper weighting, I selected Number of respondents who prefer different modes of enjoying tennis as 'weight cases by' for the test.
Conduct a one-sample chi-squared test
a) Frequency preferring watch tennis at the stadium is 24
b) p value is p < .179
c) X2 valueX2(2, N = 54) = 3.44
Expected frequency for each category
Since N = 54 and there were 3 categories, watching on TV, watching at the stadium or by playing themselves, then the expected frequency for each category would be 54/3 = 18, also shown in the frequency table.
A one-sample chi-squared test was run to determine if ...