PLACE:  1126 Sweeney

SPEAKER:
Christopher Bilder
Kansas State University

TITLE:
Testing for Marginal Independence with Pick Any/c Variables

ABSTRACT:
 

Many survey questions allow respondents to pick any number out of c possible categorical responses or "items".  These kinds of survey questions often use the terminology "choose all that apply" or "pick any".  Often of interest is determining if the marginal response distributions of each item differ among r different groups of respondents.  Agresti and Liu (1998, 1999) call this a test for multiple marginal independence (MMI).  If respondents are allowed to pick only 1 out of c responses, the hypothesis test may be performed using the Pearson chi-square test of independence.  However, since respondents may pick more or less than 1 response, the test's assumptions that responses are made independently of each other is violated.  Recently, a few MMI testing methods have been proposed by Loughin and Scherer (1998),  Agresti and Liu (1998, 1999), and Decady and Thomas (1999).  The first part of the talk will evaluate the proposed methods and discuss a few new methods.  The second part of the talk will examine the case of determining if the marginal response
probabilities of each item differ among r different groups of respondents within q strata.  This is a hypothesis test for conditional multiple marginal independence (CMMI).  Extensions of the Cochran-Mantel-Haenszel test are proposed to test for CMMI.
 
 
 
 

COFFEE: 3:40p.m., 104 Snedecor Hall