<html><head></head><body><div style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:13px;"><div><div>Thanks, Mike!</div><div><br></div><div>Could you provide some interpretive guidelines for identifying local dependence using Tables 23.1 and 23.99?</div><div><br></div><div>Thanks!</div><div><br></div><div>Scott </div><div><br></div><div class="ydp8101efb5signature">~~~~~~~~~~~<br>Scott R Millis, PhD, ABPP, CStat, PStat®<br>Professor<div>Dept of Physical Medicine & Rehabilitation</div><div>Dept of Emergency Medicine<br>Wayne State University School of Medicine</div><div>Detroit, MI  48201<br>Email: aa3379@wayne.edu<br>Email: srmillis@yahoo.com<br>Tel: 313-993-8085</div></div></div>
            <div><br></div><div><br></div>
            
            <div id="ydpa33960feyahoo_quoted_2544005622" class="ydpa33960feyahoo_quoted">
                <div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
                    
                    <div>On Monday, March 26, 2018, 10:14:58 PM EDT, Mike Linacre <mike@winsteps.com> wrote:</div>
                    <div><br></div>
                    <div><br></div>
                    <div><div id="ydpa33960feyiv2188463682">
  

    
  
  <div>
    <div class="ydpa33960feyiv2188463682moz-cite-prefix">Trevor:<br>
      <br>Q3 is the correlation between the raw residuals of pairs of items. These are listed, for instance, in Winsteps Table 23.99 with PRCOMP=R<br>
      <br>Summarizing the Q3's across all items with PCA makes the dependency pattern clearer. See, for instance, Winsteps Table 23.1<br>
      <br>Hope this help.<br>
      <br>Mike L.<br>
      <br>On 3/27/2018 10:38 AM, Bond, Trevor wrote:<br>
    </div>
    <blockquote type="cite">
      <div style="color:rgb(0, 0, 0);"><font face="Times"><span style="font-size:16px;">Dear Colleagues,</span></font></div>
      <div style="color:rgb(0, 0, 0);"><font face="Times"><span style="font-size:16px;">It has been claimed that <span lang="EN-US">Yen’s Q3 index </span></span></font><span style="font-family:Times;font-size:16px;"><span lang="EN-US">(1984) </span></span><span style="font-family:Times;font-size:16px;"><span lang="EN-US">can be used to
            identify local dependence in Rasch analyses.</span></span></div>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span lang="EN-US"><font style="color:rgb(0, 0, 0);" face="Times">e.g., Chou, Y.-T., & Wang,
            W.-C. (2010). Checking dimensionality in item response
            models with principal component analysis on standardized
            residuals. Educational and Psychological Measurement, 70(5),
            717-731.</font></span></p>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span style="font-size:12pt;font-family:Times;">Christensen, K. B., Makransky, G., &
          Horton, M. (2017). Critical values for Yen’s Q3:
          Identification of local dependence in the Rasch model using
          residual correlations. Applied Psychological Measurement,
          41(3), 178-194.</span></p>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span style="font-size:12pt;font-family:Times;">Have you used Q3 for that purpose? Do you
          know of papers that have adopted it? Is it helpful/effective?
          Is it available in the software that you use, or know about?</span></p>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span style="font-size:12pt;font-family:Times;">Looking forward to your advice and
          guidance.</span></p>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span style="font-size:12pt;font-family:Times;">Collegially</span></p>
      <p class="ydpa33960feyiv2188463682MsoCommentText"><span style="font-size:12pt;font-family:Times;">Trevor</span></p>
    </blockquote> <br>
    <br>
    <div class="ydpa33960feyiv2188463682moz-signature">-- <br>Mike Linacre, <a class="ydpa33960feyiv2188463682moz-txt-link-abbreviated" href="mailto:mike@winsteps.com" rel="nofollow" target="_blank">mike@winsteps.com</a> or <a class="ydpa33960feyiv2188463682moz-txt-link-abbreviated" href="mailto:winsteps1234@gmail.com" rel="nofollow" target="_blank">winsteps1234@gmail.com</a><br>Winsteps 4.0.1 and Facets 3.80.4 - <a href="http://www.winsteps.com/index.htm" rel="nofollow" target="_blank"></a><a class="ydpa33960feyiv2188463682moz-txt-link-abbreviated" href="http://www.winsteps.com" rel="nofollow" target="_blank">www.winsteps.com</a><br>Concerned about aches, pains, youthfulness? Mike suggests <a href="https://www.modere.com/liquidbiocell?referralCode=790842" rel="nofollow" target="_blank">Liquid Biocell</a></div>
  </div>

</div>________________________________________<br>Rasch mailing list<br>email: <a href="mailto:Rasch@acer.edu.au" rel="nofollow" target="_blank">Rasch@acer.edu.au</a><br>web: <a href="https://mailinglist.acer.edu.au/mailman/options/rasch/srmillis%40yahoo.com" rel="nofollow" target="_blank">https://mailinglist.acer.edu.au/mailman/options/rasch/srmillis%40yahoo.com</a></div>
                </div>
            </div></div></body></html>