Suchergebnisse

The methodology comprises a computer 'clock', capable of accurately timing responses, and a 'voice-key' that converts sounds emitted by respondents into signals capable of triggering the computer clock. Response times to questions on a range of attitude questions were measured. Results revealed orderly norms in the latencies associated with various types of survey questions. (Abstract amended)

Over the past fifty years, attitude surveys have won general acceptance as means of describing public moods & beliefs. However, the degree to which such surveys can be taken to predict behavior is not clear. Certain attitudinal items, in fact, allow prediction of voting turnout in different groups, making the prediction of voting behavior possible from attitudinal data. These results do not, however, offer any means of explaining such phenomena as declining turnout rates. The Watergate events do not account for this, since turnout has been declining steadily outside the South since 1960. The increasing number of voting-age Americans who do not identify with a political party, which has also risen over that period, offers one area in which such explanation may be sought; such an explanation may be found in the impact of televised campaigning on party structure. 2 Tables. W. H. Stoddard.

A methodology is introduced for accurately measuring the time it takes respondents (Rs) to answer questions in computer-assisted telephone interview (CATI) surveys. The methodology, which is completely invisible to Rs, comprises a computer "clock" capable of timing responses with millisecond accuracy, & a "voice-key" that converts sounds emitted by Rs into signals capable of triggering the computer clock. Response-times to a range of attitude questions, including stable & unstable attitudes, in the CATI survey (N = 246 completed interviews) were measured. The results reveal orderly norms in the latencies associated with various types of survey questions. The latencies associated with the expression of stable & unstable attitudes are discussed in the context of the notion of "nonattitudes" & shed new light on this controversial thesis. Overall, results demonstrate that response latencies can be measured precisely & reliably in CATI surveys & that the data from such measurement open new windows on the cognitive dynamics of survey responses. 2 Figures, 24 References. AA

A shortened version of publication No. A-54 of the Survey Res Center, U of California, Berkeley, on the collection, methodological evaluation, & distribution of existing survey materials collected in developing countries, for comparison purposes. The major problems for the comparative res'er are classified as (A) Sampling errors: (1) poor workmanship, (2) incompleteness of coverage & lack of comparability of sampling frames, (3) problems of relevancy of cluster & quota sampling, (4) obstacles to soc & cultural accessibility (noncompletion rates); (B) Measurement errors: (i) over-sophistication of questions asked by res'ers leading to meaningless evidence, (ii) difficulties of conceptual & linguistic equivalence; (C) Interview bias: (a) biased data because of presence of clinical witnesses, (b) courtesy bias, (c) interviewer-R status congruency. These problems are examined & brought to the attention of comparative res'ers. Some suggestions as to how they can be overcome are made. M. Maxfield.