At 10:53 27/11/2000 GMT+0100, you wrote: >Hi All, > >I am new to the list, but I have a question regarding SPSS and the >MCAR-test (missing completely at random). > >I have a dataset that contains some items with instructions when to >use "no answer", namely because a person has not experienced the >situation or it would be to long ago. There was not filter question, >but three questions where realted to the same event (admission >to a nursing home). In many cases all three items where missing >because the admission was to long ago. In some cases one or two items >where missing due to unknown reasons. > >I performed a MCAR test with interviewer and institute as controls. I >expected SPSS to tell me the data would not be MCAR, but to my >surprise this did not happen. Can someone tell me what might have >gone wrong? Or is my thinking flawed in any way? > > >Greetings, > > >Peter Moorer > >ARGO University of Groningen > > I have a method for visually exploring missing data patterns that could be specially appropriate for situations like this. It runs in ViSta, a freeware package available from several places. You might try it or send me the data so I might have a go. A preliminary document on the method is available in:
http://www.uv.es/~valerop/ Look in the page for the link titled Missing data analysis in ViSta. It is experimental. Pedro ========================================================== Dr. Pedro M. Valero-Mora. WWW: http://www.uv.es/~valerop Director M?todos de Investigaci?n Phone: +34 96 3393880 Universitat de Val?ncia INTRAS-Instituto de Tr?fico y Seg. Vial. INTRAS-Institut of Traffic and Road Safety. C/ Hugo de Moncada 4. Entresuelo. Valencia, 46010, Espa?a (Spain). ==========================================================
