Standardized chi-square statistic for linear-by-linear association

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Standardized chi-square statistic for linear-by-linear association

Margaret MacDougall
Dear all

  I have run a number of chi-square tests of linear trend using the SPSS crosstabs procedure and in the corresponding row of tabulated results for each test entitled 'Linear-by-linear association', I have a reference to a footnote which contains a statement of the form 'The standardized statistic is ...'. Typical cases involve a tabulated chi-square statistic value of about 15 which is referenced to a footnote in which the standardized statistic is of value about -3.  What is happening here? In particular what is the mathematics behind the standardization and why has it been carried out. Is there a relevant publication to which I can refer?

  I look forward to receiving some illumination.

  Many thanks

  Best wishes

  Margaret




---------------------------------
 The all-new Yahoo! Mail goes wherever you go - free your email address from your Internet provider.