Interviewer Effect on Conversational Interviews: Results from the 2010 & 2011 SIPP-EHC

Rachael Walsh, U.S. Census Bureau

The new conversational interviewing style associated with an Event History Calendar is a departure from the conventional scripted style of Census Bureau surveys. The new Survey of Income and Program Participation include an Event History Calendar (SIPP-EHC) and is currently undergoing annual field tests. The role of the interviewer and differences in interview administration by interviewer is of particular interest. Between 2010 and 2011, the interview length decreased by 17 minutes, but this is not solely attributable to the interviewers because the instrument is still in development. This research uses multilevel modeling to assess the interviewer effect on individual interview length using data from the 2010 and 2011 SIPP-EHC field tests. Interviewer effect decreased 3% from 2010 to 2011. In 2011, 6% of the variation in interview length was attributable to interviewer effect. This makes the interviewer effect on the conversational 2011 SIPP-EHC comparable to the scripted 2008 SIPP interviews.