<?xml version="1.0" encoding="utf-8"?><!DOCTYPE article  PUBLIC '-//OASIS//DTD DocBook XML V4.4//EN'  'http://www.docbook.org/xml/4.4/docbookx.dtd'><article><articleinfo><title>FAQ/poly</title><revhistory><revision><revnumber>10</revnumber><date>2013-03-08 10:17:36</date><authorinitials>localhost</authorinitials><revremark>converted to 1.6 markup</revremark></revision><revision><revnumber>9</revnumber><date>2010-03-18 11:52:54</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>8</revnumber><date>2009-06-16 11:19:35</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>7</revnumber><date>2009-06-16 11:16:38</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>6</revnumber><date>2009-06-16 11:16:21</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>5</revnumber><date>2009-06-16 11:11:29</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>4</revnumber><date>2009-06-16 11:09:54</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>3</revnumber><date>2009-06-16 11:07:48</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>2</revnumber><date>2009-06-16 11:05:56</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>1</revnumber><date>2009-06-16 11:04:18</date><authorinitials>PeterWatson</authorinitials></revision></revhistory></articleinfo><section><title>Comparing two trends in proportions using SPSS</title><para>One can quite easily check for differences in linear trend in SPSS using the polynomial subcommand in the logistic procedure for frequencies (TAB5) and reading off the Wald (provided there are no inf MLES which there aren't here!): </para><para>For example suppose I have three genotypes, hsert, increasing in the presence of a particular aspect and two severity levels relating to the presence of absence of childhood maltreatment with two other columns containing the frequencies and the binary response. The first two entries might be something like the below </para><informaltable><tgroup cols="7"><colspec colname="col_0" colwidth="25*"/><colspec colname="col_1"/><colspec colname="col_2"/><colspec colname="col_3"/><colspec colname="col_4" colwidth="25*"/><colspec colname="col_5" colwidth="25*"/><colspec colname="col_6" colwidth="25*"/><tbody><row rowsep="1"><entry colsep="1" nameend="col_3" namest="col_0" rowsep="1"><para> <emphasis role="strong">Tab5</emphasis> </para></entry><entry colsep="1" rowsep="1"><para> <emphasis role="strong">Sert</emphasis> </para></entry><entry colsep="1" rowsep="1"><para> <emphasis role="strong">Maltreatment(Yes/No)</emphasis></para></entry><entry colsep="1" rowsep="1"><para> <emphasis role="strong">Depression(Yes/No)</emphasis> </para></entry></row><row rowsep="1"><entry align="center" colsep="1" nameend="col_3" namest="col_0" rowsep="1"><para> 3 </para></entry><entry colsep="1" rowsep="1"><para> ll </para></entry><entry colsep="1" rowsep="1"><para> Yes </para></entry><entry colsep="1" rowsep="1"><para> <emphasis role="strong">Yes</emphasis> </para></entry></row><row rowsep="1"><entry align="center" colsep="1" nameend="col_3" namest="col_0" rowsep="1"><para> 16 </para></entry><entry colsep="1" rowsep="1"><para> ll </para></entry><entry colsep="1" rowsep="1"><para> Yes </para></entry><entry colsep="1" rowsep="1"><para> <emphasis role="strong">No</emphasis> </para></entry></row></tbody></tgroup></informaltable><screen><![CDATA[WEIGHT
  BY TAB5 .
]]><![CDATA[
LOGISTIC REGRESSION  dep
  /METHOD = ENTER hsert mal hsert*mal  
  /contrast(hsert)=polynomial
  /CRITERIA = PIN(.05) POUT(.10) ITERATE(20) CUT(.5) .]]></screen><para><emphasis role="underline">Reference</emphasis> </para><para>Agresti A. (2002) Categorical Data Analysis. Second Edition. Wiley: New York.  </para></section></article>