# Suggested answer to exercise: Serious falls in a home exercise programme

Question: Using SPSS, cross-tabulate experiences of falls by treatment group and carry out Fisher's exact test.

We choose Analyze, Descriptive Statistics, Crosstabs. Put the variable 'Experience of falls' into Row(s): and 'Group' into Column(s):. Click Statistics, Chisquare, Continue. Click OK.

You should get the case processing summary and then the following:

Experience of falls * Group Crosstabulation
Count
Group Total
Control Exercise
Experience of falls No fall 110 119 229
Fall 9 2 11
Total 119 121 240

Chi-Square Tests
Value   df   Asymp. Sig.
(2-sided)
Exact Sig.
(2-sided)
Exact Sig.
(1-sided)
Pearson Chi-Square 4.7926a 1 .029
Continuity Correctionb 3.536    1 .060
Likelihood Ratio 5.155    1 .023
Fisher's Exact Test .033 .028
Linear-by-Linear Association 4.772    1 .029
N of Valid Cases 240
a 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.45.
b Computed only for a 2x2 table

For a two by two table, we automatically get Fisher's exact test and chi-squared with Yates' continuity correction.

We have Fisher's exact test: P = 0.03.

As usual, I have added a leading zero and for P values given only the first significant (i,e, non-zero) figure. When we do this we round as for decimal places. Here we round down from 0.033 to 0.03.

The chi-squared test is valid and gives us the same probablity, P = 0.03, after rounding up from 0.029.