Cognitive Interview Results and Recommendations for the National Science Foundation’s Proposed Field of Degree Question for the American Community Survey
Jennifer M. Rothgeb and Jennifer L. Beck
KEY WORDS: cognitive testing, interviewer-administered survey, self-administered survey, American Community Survey, ACS, field of degree, National Science Foundation, bachelor’s degree
ABSTRACT
The Decennial Statistical Studies Division (DSSD) contracted with the Statistical Research Division (SRD) to conduct cognitive testing of a proposed question for the American Community Survey (ACS) on “field of degree” for persons holding a bachelor’s degree or higher. Three versions of the question were proposed by the National Science Foundation (NSF) for testing. The NSF requested that a field of degree question be added to the ACS to enable them to create a data base and sampling frame of scientists and engineers for the National Survey of College Graduates (NSCG.) Four researchers from SRD conducted a total of 42 cognitive interviews using three different versions (two forced-choice formats, one open-ended format) of the field of degree question. The questions were tested using all three modes of interviewing used by the ACS. All interviews were conducted in the Washington DC metropolitan area during July and August 2006. Results indicate that respondents tend to understand the intent of the question, but they are less clear about which category is the best match for their field of degree when presented with the forced-choice format. Also, with the simulated computer-assisted personal interviewing (CAPI) and computer-assisted telephone interviewing (CATI) forced-choice formatted cases, the lack of an instruction telling respondents to answer “yes” or “no” for each field of degree category presented an unknown response task problem for respondents. Details results on these and other findings are provided in the report. Recommendations agreed upon be the NSF and the Census Bureau are also included.
CITATION:
