Dear All,
It seems that the Chi-Square Test for Trend is the Mantel-Haenszel
test. In SPSS output it is the row called "linear by linear
association", which should be reported instead of the Pearson chi-square.
Tamara (thank you very much!) mentioned that it's discussed in Pett,
M. A. (1997) "Non-parametric statistics for health care research" if
anybody is interested!
Best wishes
Ioanna
Ioanna Vrouva
PhD Student
Research Department of Clinical, Educational and Health Psychology
University College London
1-19 Torrington Place, Room 451a
London WC1E 7HB
Tel: 0207 679 8273
Fax: 0207 916 1989
Email: [log in to unmask]
|