Question 8: In the SIP table, why should we be cautious in interpreting the significant change in the physical dimension?
The problem here is multiple testing. This is one of 15 tests of significance looking at the changes in SIP. If the null hypothesis that the SIP does not change in any dimension is true, the probability that we would get at least one of the P values less that 0.05 is quite high. We should always beware of one lone significant difference among a group of non-significant ones.
Inspection of the table shows that for some dimensions the score is higher post Asthma School, for some lower, and for the total score the means are the same. There is nothing to suggest that SIP has decreased.
Back to Exercise: Evaluation of a nurse-run asthma school.
To Applied Biostatistics index.
To Martin Bland's M.Sc. index.
To Martin Bland's home page.
This page maintained by Martin Bland.
Last updated: 20 October, 2006.
Back to top.