Showing papers in "Public Opinion Quarterly in 1993"

Estimating the effect of incentives on mail survey response rates: a meta-analysis

Abstract: This article reports the results of a meta-analysis of 38 experimental and quasi-experimental studies that implemented some form of mail survey incentive in order to increase response rates. A total of 74 observations or cases were classified into one of four types of incentive groups: those using prepaid monetary or nonmonetary rewards included with the initial survey mailing and those using monetary or nonmonetary rewards as conditional upon the return of the survey. Results were generated using an analysis of variance approach. The overall effect size across the 74 observations was reported as low to moderate at d = .241. When compared across incentive types, only those surveys that included rewards (both monetary and nonmonetary) in the initial mailing yielded statistically significant estimates of effect size (d = .347, d = .136). The average increase in response rates over control conditions for these types of incentives was 19.1 percent and 7.9 percent, respectively. There was no evidence of any impact for those incentive types offering rewards contingent upon the return of the survey.

Who gets the news? alternative measures of news reception and their implications for research

Abstract: This article investigates patterns in audience recep- tion of 16 news stories that received prominent media coverage in the summer and fall of 1989. Using a national sample of American adults, it compares education, self-reported rates of media use, interpersonal communication, and prior levels of general political knowledge as predictors of individual differences in recall of cur- rent news events. Results indicate that respondents' background level of political knowledge is the strongest and most consistent predictor of current news story recall across a wide range of topics, suggesting that there is indeed a general audience for news and that this audience is quite sharply stratified by preexisting levels of background knowledge. Thus, in survey research appli- cations that require estimates of individual differences in the re- ception of potentially influential political communications, a mea- sure of general prior knowledge-not a measure of news media use-is likely to be the most effective indicator. The article fur- ther concludes that the tendency of individuals to acquire news and information on a domain- or topic-specific basis fails to un- dermine the value of political knowledge as a general measure of propensity for news recall.

Effects of questionnaire length, respondent-friendly design, and a difficult question on response rates for occupant-addressed census mail surveys

Abstract: An experimental study of alternatives to the current U.S. decennial census questionnaire shows that shortening the questionnaire and respondent-friendly questionnaire design improve response, whereas asking a potentially difficult and/or objectionable question, that is, social security number, lowers response. This national study of 17000 household addresses also desmonstrates that relatively high mail survey response can be achieved without addressing correspondence to individual names of residents

Interviewer gender and gender attitudes

Abstract: L'article souleve la question des effets du genre de l'interviewer (masculin-feminin) sur les reponses apportees aux interviews, et plus generalement sur les attitudes des personnes interviewees ; le genre respectif de ces personnes affecte a son tour le contenu des reponses et les attitudes adoptees.

Response latency versus certainty as indexes of the strength of voting intentions in a cati survey

Abstract: Comparaison de deux comportements politiques : l'intention de vote et le comportement electoral. Etude des notions d'accessibilite (intention de vote) et de certitude (comportement electoral), qui stucturent le champ du comportement politique.

Innumeracy about minority populations

Abstract: We use a small but nationally representative sample to investigate the sources of innumeracy regarding the proportion of blacks Hispanics and Jews in the U.S. population. In addition to a number of standard demographic differences we find that overestimates are closely related to region as well as to the density of the local black/Hispanic population. The extent to which minority populations are perceived as a kind of threat is also related to perceived proportions though the direction of causality cannot be determined. (EXCERPT)

The impact of privacy and confidentiality concerns on survey participation the case of the 1990 u.s. census

Abstract: This study investigates the relationship between concerns about confidentiality and privacy, on the one hand, and mail returns to the 1990 census, on the other. Such concerns significantly affect mail returns, even when demographic variables known to be related to both concerns and survey participation are controlled. However, their impact is not very large, and the effects of confidentiality and privacy concerns vary for black and white respondents

Should we take don't know for an answer?

Abstract: Much attention has been given to the problem of non-attitudes, that is, people expressing opinions while lacking an underlying attitude. In comparison, the potential problem of false negatives has been neglected. Using a survey on nuclear power from Sweden, we examine whether people who answer « don't know » but are induced subsequently to give an opinion really have attitudes. The attitudes these people express on follow-up questions predict behavior to a significant extent. This implies that the usual don't know category includes some false negatives, that is, people who really have attitudes but refrain from expressing them

Collective memories in britain and the united states

Abstract: In this article we investigate the events and changes that the British and American public regard as important. We argue that national differences in perceptions of past events and changes might lead to different evaluations of recent history. Our data are from a 1990 probability sample of British households in which respondents were asked to report "the national or world events or changes over the past 60 years" that seemed to them especially important and, then, to explain the reasons for their choices. These questions replicated items from an earlier American survey on the Intersection of Personal and National History. The data are used both qualitatively and quantitatively to compare British and American views of recent history. Overall, the two nations have remarkably similar views on which events are important, and there are also striking similarities in the way British and American choices are structured by cohort. However, interesting national differences emerge in the meaning associated with World War II, the most frequently mentioned event. We discuss the implications of such distinctive memories for public opinion. The English and Americans have been described as two peoples divided by a common language. This somewhat facetious characterization understates the immense differences in culture, systems of government, and national character of the two nations, to say nothing of their divergent histories and different roles in global politics. Given these JACQUELINE SCOTT is research director at the Economic and Social Research Council (ESRC) Research Centre on Micro-social Change. LILIAN ZAC is a Ph.D. student in the Department of Government at the University of Essex. The support of ESRC and the University of Essex is gratefully acknowledged. The work was part of the scientific program of the ESRC Research Centre on Micro-social Change in Britain. The authors are grateful for comments on an earlier draft by Howard Schuman, Cheryl Rieger, and colleagues at the British Household Panel Study (BHPS). The authors also acknowledge the help of Jane Rooney and Ann Farncombe in data entry and Graham Upton and Alan Taylor for statistical advice. Public Opinion Quarterly Volume 57:315-331 ? 1993 by the American Association for Public Opinion Research All rights reserved. 0033-362X/93/5703-0001$02.50 This content downloaded from 157.55.39.180 on Mon, 25 Apr 2016 06:57:53 UTC All use subject to http://about.jstor.org/terms 316 Jacqueline Scott and Lilian Zac differences, one might expect Britain and America each to have a peculiarly distinctive view of recent history. In this article we investigate systematically the events and changes that the British public regard as important in recent history and examine how the British view of history differs from that of Americans.1 The American data come from a 1985 survey on the Intersection of Personal and National History, conducted by the Survey Research Center at the University of Michigan. The core questions asked people what national or world events or changes were especially important in the last 50 years, and the reasons why they chose a particular event. These questions were replicated in Britain in October 1990. Thus, with the important exception of the 5-year time lapse between the two surveys, we can directly compare the historical perceptions and memories of the two countries. Of course, this 5-year period was one of immense historical change, with the fall of the Berlin Wall and the more general collapse of communist regimes, within both Europe and the Soviet Union itself. Thus it is not a period that is likely to go unmentioned in terms of important world events and changes. Despite this major difference in the historical timing of the two surveys, we expect Americans and the British to share a broadly similar view of recent history. For example, we expect that both countries will have a common view of World War II as an event of outstanding importance. This is, in part, because the two countries have been allies throughout the period in question. Also, the American and British mass media, through which both historical and current events are portrayed, have much in common. Yet, we also expect there to be some clear national differences. For example, we predict that Americans will be more likely to mention domestic events and changes than the British because America is a superpower and because its own internal events and changes play a part in shaping world history in a way that is not true of Britain. In the American study, strong cohort effects were found in people's choices of historical events, with people's memories referring back disproportionately to events that occurred in their youth. This finding provided empirical support for the fundamental premise behind Mannheim's idea of a political generation: that historical events that happen in people's formative years leave a permanent imprint on people's memories (Mannheim [1928] 1952). If the same pattern is true of the British data, then mentions of recent events in Europe should be con1. This article owes much to the ideas that informed the American project-the Intersection of Personal and National History-for which Howard Schuman and Philip Converse were the principal investigators. These original investigators, however, bear no responsibility for the British replication of the study or the opinions and interpretations expressed herein. This content downloaded from 157.55.39.180 on Mon, 25 Apr 2016 06:57:53 UTC All use subject to http://about.jstor.org/terms Collective Memories 317 fined to younger cohorts, with older people tending to overlook such events in favor of earlier events from their teens or youth. The British data, however, because they were collected in 1990, less than a year after events such as the fall of the Berlin Wall, might give rise to what could be called a "period effect," with virtually everyone mentioning the recent events in Europe. Of course, if this happens, then the period effect would effectively wipe out any cohort effect. Our prediction is that, despite the historical importance of 1989, older cohorts will select events from their youth. This is not only because of the fact that for older people earlier memories are most readily recalled (e.g., McCormack 1979; Rubin, Wetzler, and Nebes 1986) but also because these earlier events have the advantage of primacy and later events rarely have the same impact (Halbwachs [1950] 1980; Mannheim [1928] 1952, p. 296; Schuman and Scott 1989). Thus we expect World War II to dominate the memories of older people in Britain, just as it dominated older people's memories in the United States.2 An interesting question is whether World War II will be significantly more memorable for the British and whether the wartime memories of the two nations differ significantly, not just in terms of individual memories but also in terms of the way in which the war has been represented in the collective memory of each nation. Past events may both influence how recent events and changes are evaluated and influence political action (Connerton 1989; Middleton and Edwards 1990). The "Munich model," for instance, had great potency as an example of how not to conduct foreign policy, on both sides of the Atlantic. However, if the two nations differ in their representation of past events, then this could have important ramifications for public opinion. For example, in America George Bush successfully used the analogy to Hitler to condemn Saddam Hussein and justify a military offensive in Iraq (Schuman and Rieger 1992). Yet, this analogy may not have been as effective in Britain, because, as we shall see, the British do not wholeheartedly share the American view of World War II as "The Good War" (Terkel 1984).

“my wife knows best” a comparison of event dating accuracy between the wife, the husband, the couple, and the belgium population register

Abstract: This study offers a new approach to the problem of memory bias in retrospective data by using a sample survey conducted in Belgium that allows the comparison of recall accuracy not only between the husband and the wife but also the couple together. Respondents' errors in event dating is generally in years, and not in months, and furthermore this pattern is not general and is related to the event itself. Design of retrospective questionnaires should take into account tha fact that time units are not recalled equally for differing autobiographical events, and that dating accuracy could be improve through use of parallel event questioning.

Newspaper effects on policy preferences

Abstract: This research offers clear evidence that what is presented in the media influences the policy preferences of the American public More important, it reconciles some earlier problems that led to an inability to determine the approximate relationships between television and newspaper information concerning policy By correcting some methodological problems in earlier attempts to examine the impact of New York Times policy news, the A is able to conclude that different actors or news sources do indeed have differential impacts on public opinion

Using paid advertising to modify racial stereotype beliefs

Abstract: This is a pilot study to determine the feasability of using mass media to modify beliefs underlying discrimination against Aborigines in a small country town in Australia. A mass media campaign was developed utilizing primarily paid advertising. A major aim of the campaign was an attempt to neutralize beliefs about Aborigines and employment. The results showed significant changes in beliefs about the proportion of Aborigines in paid employment and in the proportion of employed Aborigines remaining in a job for an extended period of time.

The social bases of feminism in the european community

Abstract: Etude des relations entre les objectifs des mouvements feministes americains et europeens et les attitudes et mesures individuelles rencontrees en Europe et aux Etats-Unis Identification des attitudes de base et du cadre social de base propices ou en coherence avec l'ideologie feministe

Multiple-measure assessment of party identification

Abstract: Recent research has focused attention on the impor- tance of accounting for measurement error in party identification when modeling the stability of partisanship and the determinants of the vote. Measurement error estimates have in the past been based on a single measure of partisanship observed at multiple points in time, a test-retest methodology that requires fairly strong assumptions about the character of change over time. This article assesses the reliability of the Michigan party identification scale using multiple measures of partisanship at a single point in time. Our data not only corroborate previous test-retest results but also suggest that the accuracy with which partisanship pre- dicts candidate preferences can be enhanced using multiple mea- sures. One measure in particular, a labeled 7-point self-placement continuum, is found to hold significant potential to supplement and illuminate the Michigan scale.

Survey evidence on excessive public pessimism about the future behavior of unemployment

Abstract: Two of the questions in the Surveys of Consumer At- titudes undertaken by the Survey Research Center of the Univer- sity of Michigan deal with households' expectations about infla- tion and the change in unemployment. We compare quarterly time series of the mean responses with the actual behavior of inflation and unemployment to see whether households are overly optimistic or pessimistic about the future behavior of inflation and unemployment. We find an asymmetry. Over the 21 years of our analysis, on average the public is neither overly optimistic nor overly pessimistic about future inflation. However, the public has been significantly too pessimistic about the future behavior of unemployment. These results have an important implication for macroeconomic policy. If politicians respond to complaints from the public about inflation and unemployment they will target policy instruments at unemployment to a greater extent than if the public was not so pessimistic about the behavior of unemploy- ment. The result may be a higher rate of inflation than if the public were not mistakenly pessimistic about the behavior of un- employment.

Direct and indirect routes to politicizing personal experience does knowledge make a difference

Abstract: This study examines the effects of issue-specific knowledge on the extent to which personal unemployment experiences influence presidential approval. The well informed are found to be more likely to connect personal experiences directly to political preferences, yet less likely to generalize from their own personal experiences in assessing the state of the nation as a whole. Since perceptions of the state of national conditions have a well-documented influence on presidential approval, simultaneously considering direct and indirect paths helps to resolve contradictory implications in past findings on the role of information in conditioning the politicization of personal economic experience. The weak overall impact of self-interest on political preferences is widely known and well documented (for a review, see Sears and Funk 1990). Subsequent research in this vein has shifted toward locating conditions under which personal experiences are most likely to be politicized (see, e.g., Feldman 1985). The concepts used to locate groups most likely to rely on personal experiences-have run the gamut from distinctions made on the basis of socioeconomic position (Weatherford 1983), habitual versus occasional voting (Cohen and Uhlaner 1991), levels of issue-specific knowledge (Conover, Feldman, and Knight 1986), and levels of news media use (Mutz 1992; Weatherford 1983). Scholars generally concur about the distinction relevant to these comparisons: those with high levels of political knowledge, heavy use of mass media, high levels of political interest, and regular involvement in the political process are expected to differ in their decision-making DIANA C. MUTZ is an assistant professor in the Department of Political Science and the School of Journalism and Mass Communication at the University of WisconsinMadison. The author would like to thank the Letters and Sciences Survey Center at the University of Wisconsin-Madison for making these data available and Dick Brody for his comments on an earlier draft of this study. Public Opinion Quarterly Volume 57:483-502 ? 1993 by the American Association for Public Opinion Research All rights reserved. 0033-362X/93/5704-0006$02.50 This content downloaded from 207.46.13.103 on Thu, 20 Oct 2016 04:02:11 UTC All use subject to http://about.jstor.org/terms

Argumentative complexity of abortion discourse

Abstract: Using integrative complexity theory and its associated coding scheme, this article expores the structure of arguments on abortion articulated by single and multi-issue « pro-choice » and « pro-life » groups between July 1989 and May 1991. A simple random sample of 13 paragraphsized statements representative of each organization's position was rated by two trained coders on a 7-point scale measuring conceptual differentiation and integration. The debate was conducted at a low level of integrative complexity. Both pro-choice and pro-life arguments were characterized by similarly low levels of integrative complexity. Supporting an ideologue hypothesis, the arguments of multi-issue organizations were more integratively complex

Measuring satisfaction with organizations predictions from information accessibility

Abstract: The study explored the process through which customers assess their satisfaction with service organizations. Our theoretical analysis suggests that when a general satisfaction question appears after questions about specific domains were asked (SG order), the earlier questions increase the accessibility of both positive and negative information. In contrast, when a general satisfaction question appears prior to any other question (GS order), negative information is more accessible than positive information. On the basis of these differences in accessibility we predicted (1) greater asymmetry in the impact of positive and negative information on the general satisfaction response in the GS order, (2) higher correspondence between domain-specific satisfaction and general satisfaction in the SG order, and (3) higher levels of general satisfaction in the SG order. These predictions were supported in analyses of customers' responses in a national survey of satisfaction with the Israel Telecommunication Corporation (Bezeq).

The impact of messages on survey participation in answering machine households

Abstract: This study investigated the impact of telephone an- swering machines on telephone survey participation. The study found that households with answering machines were more likely to be contacted, more likely to complete the interview, and less likely to refuse to participate in the study compared to house- holds where there was no answer on the initial call attempt. The study also investigated the utility of leaving messages on the an- swering machine as a means of encouraging participation. While leaving messages did result in higher participation rates, there were no significant differences among the types of messages tested.

Meeting the challenge of answering machines

Abstract: Answering machines have become one of the major challenges to obtaining high response rates in telephone surveys. This article presents some data that may be helpful in answering two questions that frequently arise: (1) What are the chances that additional calls to a household known to use an answering machine will result in a completed interview? (2) When is the best time to call, in order to minimize the chances of encountering an answering machine? The data used to analyze these questions are based on the calling records of the 1990 California Disability Survey, a large random-digit dialing survey that generated about 330,000 calls and completed interviews at over 24,000 households. This large number of calls is interesting not only in absolute terms but because of the high number of callbacks that they represent, designed to bring the response rate up over 80 percent. The records of calls analyzed here, consequently, can show convincingly what really happens when 10, 20, 30, or even more callbacks are attempted. Such opportunities are rare. Researchers can take advantage of these results to generate more informed calling strategies and consequently improve response rates in their surveys. Answering machines are becoming a major challenge for survey researchers to overcome. Although the exact percentage of households that have such machines is unknown, current estimates put the figure nationwide at about 25 percent-and rising (Tuckel and Feinberg 1991). For the California survey reported on here, an answering machine was encountered once or more at 31 percent of confirmed households, but many additional households may have had answering machines that did not happen to be in use when the interviewer called. THOMAS PIAZZA iS the manager of technical services at the Survey Research Center, University of California, Berkeley, and the coordinator for survey design and analysis in the Computer-assisted Survey Methods Program, also at the University of California, Berkeley. An earlier version of this article was presented at the 1992 annual meeting of the American Association for Public Opinion Research, St. Petersburg Beach, FL. Public Opinion Quarterly Volume 57:219-231 ? by the American Association for Public Opinion Research All rights reserved. 0033-362X/93/5702-0006$02.50 This content downloaded from 157.55.39.27 on Tue, 06 Sep 2016 06:00:37 UTC All use subject to http://about.jstor.org/terms

Aids—an update

Abstract: This summary of public opinion about acquired immune deficiency syndrome (AIDS) updates our earlier review (Singer, Rogers, and Corcoran 1987). Since then, there has been a dramatic increase in public information about AIDS. The Surgeon General's Report on AIDS was released in 1986, and figures compiled by the Centers for Disease Control indicate that media coverage of the disease surged in 1987, with a total of 11,852 stories, up sharply from approximately 5,000 stories each in 1986 and 1985. In early 1988, the federal government mounted a nationwide campaign to inform the public about AIDS, in particular about ways to prevent its spread. One consequence of this information explosion seems to have been a large increase in the number of people who feel they know a lot about the disease; the percentage giving this response more than doubled between 1987 and 1991. The number of media stories on AIDS in 1988, 1989, and 19907,584, 7,091, and 8,364, respectively-did not match the nearly 12,000 in 1987. But 1991 figures indicate the highest media interest yet13,209 stories. Nearly half (6,038) appeared in the last quarter of the year, which is when Magic Johnson, the superstar Los Angeles Lakers basketball player, announced that he was HIV positive. The total for the first two quarters of 1992 remained high: 7,981 stories, evidence that the media continue to find AIDS highly newsworthy. For this review, we assembled national surveys from the Roper Center for Public Opinion Research that had asked one or more questions about AIDS. Between January 1987 and July 1992, our cutoff date, we identified 116 such surveys: 27 in 1987, 13 in 1988, 9 in 1989, 24 in

Intensifiers in behavioral frequency questions

Abstract: Intensifiers, words such as "very" and "extremely," are used to magnify the meaning of the phrases to which they are applied. In a series of studies we investigate how intensifiers in question stems affect response patterns in social surveys. Our research indicates that even apparently important differences in question wording may in some situations have little or no impact; adding an intensifier to a root did not create a response shift for several intensifier/root combinations in large-scale surveys. Using both field and laboratory techniques we explore the situations in which shifts do occur and go some lengths toward describing why this happens. Response shifts were observed for two situations. First, when "extreme" was applied to "physical pain" there was a substantial and significant response shift. Second, when respon- dents were asked a question without an intensifier and then had the question repeated immediately afterward with an intensifier, a response shift was produced. In addition to the practical sig- nificance for survey methodologists, these results are important to cognitive psychologists interested in word meaning.