• While the importance of pretesting survey instruments is well-accepted, conventional methods tend to be unsystematic and relatively ineffective (Cannell et. al, 1989). Some researchers have attempted to address weaknesses in the pretest process by using other question development strategies. One such strategy is to do intensive interviewing, usually as part of a "pre-pretest." In intensive interviewing,special techniques are used to uncover information about question comprehension and other response difficulties that may not be elicited without extensive probing. A disadvantage of this technique is that it produces data that is sometimes difficult to analyze objectively. A second strategy for improving the pretest is to conduct behavior coding of interviewer-respondent interactions. This coding systematically identifies potential question problems, but does not always diagnose the sources of the problems it finds. In our paper, we report on a new method that merges the complementary strengths of intensive interviewing and behavior coding. By applying behavior coding techniques to the analysis of intensive interview data, systematic intensive interviewing provides qualitative as well as quantitative information for use in the early stages of questionnaire design. We briefly summarize these two strategies and discuss how they were integrated as part of a larger research project to develop and test questions tbr inclusion in the 1995 redesign of the National Health Interview Survey (NHIS). I ()
  • 6 ()
is ?:cites of
  • 2016-06-24 ()
  • 6 ()
  • 2002-01-01 ()
  • 22309 ()
  • 3 ()
  • Before the pretest: question development strategies ()


expand all