RESPONSE-TIME MEASUREMENT IN SURVEY RESEARCH A METHOD FOR CATI AND A NEW LOOK AT NONATTITUDES

The paper introduces a methodology for measuring accuratelythe time it takes respondents to answer questions in computer-assistedtelephone surveys. The methodology, which is completely invisibleto respondents, comprises a computer "clock," capable of timingresponses with millisecond accuracy, and a "voice-key" thatconverts sounds emitted by respondents into signals capableof triggering the computer clock. Response times to questionson a range of attitude questions, including stable and unstableattitudes, were measured. The results revealed orderly normsin the latencies associated with various types of survey questions.The latencies associated with the expression of stable and unstableattitudes are discussed in the context of the notion of "nonattitudes"and shed new light on this controversial thesis. Overall, ourresults demonstrate that response latencies can be measuredprecisely and reliably in telephone surveys and that the datafrom such measurement open new windows on the cognitive dynamicsof survey responses.     

RESPONSE LATENCY VERSUS CERTAINTY AS INDEXES OF THE STRENGTH OF VOTING INTENTIONS IN A CATI SURVEY

This CATI study compares two measures of attitude strength,accessibility and certainty, for the purpose of predicting discrepanciesbetween voting intentions and voting behavior. Accessibilitywas indexed by response latency to a voting intention questionand certainty by a question about the finality of the votingintention. Logistic regression showed response latency to bea better predictor of discrepancies between voting intentionsand voting behavior than certainty. The theoretical implicationsof this finding as well as the practical implications of a methodfor measuring response latencies in CATI election studies arediscussed.     

RESPONSE LATENCY AS A SIGNAL TO QUESTION PROBLEMS IN SURVEY RESEARCH

This research explores the potential utility of response latencyas an index of question problems in survey research. The timerespondents took to answer three types of bad questions wascompared to the time they took to answer the repaired versionsof the questions. Questions containing a superfluous negativeand double-barreled questions took longer to answer than nearlyidentical questions without these problems. Repaired versionsof questions soliciting frequency estimates, however, took longerto answer than their problematic counter parts. The resultsare discussed in the context of a model of question answering,and their implications for survey methodology are explored.