Hello,
I have this crosstable and would like to do a Fisher-Freeman-Hamilton test to see if the two (small) cohorts' distribution of performance score (PS) is the same in the two cohorts.
However, I'm very new using SPSS. I've tried to figure out how to put the crosstab into SPSS without succes.
Can you help/do you know a youtube video that explains this very clearly?
Thank you!
Simone
Hi Simone
Thanks for your question.
I’ve seen a similar question on research gate and has some good attachments in this link and advice - https://www.researchgate.net/post/Does_SPSS_provide_the_Fisher-Freeman-Halton_Exact_Test_for_contingency_tables
hope this helps
http://vassarstats.net/fisher2x3.html
https://www.youtube.com/watch?v=h2Mv35pKbIM
I took a chance on opening your Excel file, and found that it showed a contingency table with the observed frequencies. In that case, you can read in the observed frequencies and use the WEIGHT command, as follows:
*** Start of syntax ***.
NEW FILE.
DATASET CLOSE ALL.
DATA LIST LIST / PS Cohort n (3F5.0).
BEGIN DATA
1 1 7
1 2 8
2 1 16
2 2 10
3 1 12
3 2 4
4 1 3
END DATA.
WEIGHT BY n.
CROSSTABS
/TABLES=PS BY Cohort
/STATISTICS=CHISQ
/CELLS=COUNT EXPECTED
/COUNT ROUND CELL
/METHOD=EXACT TIMER(5).
*** End of syntax ***.
Notice that I did not include row of input for the PS=4, Cohort=2. I omitted that row, because the observed count = 0, and weighting by 0 does not work in SPSS. Another option would be to include a row of input and change 0 to a very small value such as 0.000001. If both observed counts in a row were 0, one would have to do that, for example. But as there is only one observed count = 0, we can just omit that row of input. As the output will show, you still see an observed count of 0 in cell 4,2.
See also: https://www.ibm.com/support/pages/fisher-exact-test-rxc-table-fisher-freeman-halton-test
HTH.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS