Showing papers in "Public Opinion Quarterly in 1982"

Measuring Depression in the Community: A Comparison of Telephone and Personal Interviews

Abstract: Interview method effects in response to the Center for Epidemiologic Studies-Depression Scale (CES-D) are compared among a community sample (N = 546) of adults randomly assigned to either telephone or in-person interviews. The interviews were conducted in Los Angeles during 1980. No significant differences were observed between the two interview methods in nonresponse to symptom items, preference for specific response categories, reliability, mean level of depression, or proportion classified as depressed. Furthermore, no significant interactions were found between sociodemographic characteristics and the method of interview.

Pluralistic Ignorance and the Spiral of Silence: A Formal Analysis

Abstract: Do PEOPLE accurately perceive public opinion? The "spiral of silence" and "pluralistic ignorance" are theories that draw on psychological principles of perception, communication, and certainty to answer this question. The theory of the spiral of silence states that one's perception of the distribution of public opinion motivates one's willingness to express political opinions. The act of self-expression, however, changes the global environment of opinion, altering the perceptions of other persons and, ultimately, affecting their willingness to express their own opinions. Because individuals monitor their social environment as one cue to opinion and action, opinions with visible adherents appear to be more widely held than they are in fact. The appearance of strength becomes a self-fulfilling prophecy; those who think they are in the majority are more willing to speak out, those who think they are in the minority have an extra incentive to remain silent. The description of the plight of those who believe they are in the minority gives the name to this theory: the spiral of silence (Noelle-Neumann 1974; 1977; 1979). In recent writings on this theory, Noelle-Neumann emphasizes the importance of this phenomenon for

Indicators of Consumer Behavior: The University of Michigan Surveys of Consumers

Abstract: THE Survey Research Center at The University of Michigan began a series of periodic consumer surveys more than 35 years ago, under the direction of George Katona. The Surveys of Consumers were developed and based on a theoretical view designated as behavioral or psychological economics (Katona, 1951; 1956; 1964; 1975; 1978; 1980). The underlying theory focuses on the human factor in economic affairs. During the past several decades, changes in consumer expenditures for housing and vehicles have played a major role in determining whether the entire economy slipped into recession or moved toward recovery and growth. The consumer, rather than business or the government, has become the dominant actor in shaping the course of the aggregate economy. For the analysis of short-run fluctuations in economic activity, consumer outlays can be roughly categorized as either necessary or discretionary expenditures. The aggregate level of expenditures for necessary budget items is more stable over time and shows only small cyclical variation. Consumer decisions regarding necessary expenditures are frequently governed by need and guided by the force of habit, and in some cases, by contractual requirements-as for example, expenditures on shelter or loan repayments. Discretionary expenditures, in contrast, more frequently involve active decision mak

Jumping on the Bandwagon with the Underdog: The Impact of Attitude Polls on Polling Behavior

Abstract: An experimental design was used to examine the effect which previous polling information about candidate preference had upon subsequent polling responses. Subjects were given one of three types of information (Carter commanding a lead, Reagan commanding a lead, no information about current poll results) at two different points in time. This yielded nine information conditions. The data clearly demonstrate that both strength of attitude and candidate preference are influenced by knowledge of previous polling results. The patterns of both shift in attitude and switching candidate preference are explained in terms of an oppositional reactivity hypothesis. Stephen J. Ceci is a psychologist and Edward L. Kain is a sociologist. Both are Assistant Professors in the Department of Human Development and Family Studies, Cornell University, Ithaca, N.Y. 14853. Portions of this report were underwritten by a grant from the College of Human Ecology to S. J. Ceci. Public Opinion Quarterly Vol. 46:228-242 ?) 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-228/$2.50 This content downloaded from 157.55.39.93 on Wed, 13 Jul 2016 05:53:02 UTC All use subject to http://about.jstor.org/terms IMPACT OF ATTITUDE POLLS ON POLLING BEHAVIOR 229 possible that the absence of any bandwagon or underdog effects is a result of the fact that candidate preference grows more stable as election day approaches. Thus, these studies of bandwagon effects sampled voter opinions at a point in time when the impact of information such as televised election returns would be minimal (Mendelsohn and Crespi, 1970; Navazio, 1977). A further concern with these studies involves the response validity of self-reports of voting behavior. Although two of the studies examined precinct rosters to verify the self-reports of going to the voting booth (Fuchs, 1966, Mendelsohn, 1966) there was no way to validate the self-reports of candidate preference when those votes were cast. A long research tradition has examined response error in surveys of voting behavior (Clausen, 1968; Dinerman, 1948; Miller, 1952; Parry and Crossley, 1950; Traugott and Katosh, 1979). The recent research by Traugott and Katosh (1979) indicates that a fairly large number of respondents misreported their behavior in the 1976 CPS National Election Study. This error was large both in terms of reports of voter registration (15 percent) and reports of voting in the election (14 percent). The substantial majority of misreporting was in the direction of claiming to have voted or registered to vote when this was not, in fact, the case. It appears that some of this misreporting, therefore, was a result of feelings of pressure to give a socially appropriate response. Two further findings of the Traugott and Katosh research are important to note. The authors suggest that for those whose vote could not be verified, there was a distinct preference for Carter, indicating a small postelection bandwagon effect. Also of interest is the finding that being interviewed prior to the election is clearly linked to levels of voter participation. Further, this effect is cumulative, so that respondents who were interviewed in more than one wave of the election study had successively higher rates of validated voter participation (Traugott and Kotosh, 1979: 370-71). Navazio (1977) has suggested that research using an experimental design can help assess the impact of previous information upon polling responses. He used questionnaires including four opinion questions which were sent to both an experimental and control group. The experimental questionnaire differed only in the fact that the questions were preceded by the most recent national poll results on those questions. The response differences between the experimental and control groups did not indicate either an underdog or a bandwagon effect when examined as a whole. When occupational groups were compared, however, consistent differences appeared. The national poll results had been strongly negative in evaluating President Nixon. This content downloaded from 157.55.39.93 on Wed, 13 Jul 2016 05:53:02 UTC All use subject to http://about.jstor.org/terms

Race-of-Interviewer Effects in Telephone Interviews

Abstract: Previous studies have found a race-of-interviewer effect on survey questions dealing with racial issues. This effect has been found in both personal interviews and on questionnaires filled out in the presence of an interviewer. This study examines whether a race-of-interviewer effect is also present in telephone interviews. The results show that a race-of-interviewer effect does occur in telephone interviews on racial questions. Patrick R. Cotter is an Assistant Professor in the Department of Political Science and the Director of Survey Research and Service in the Center for Administrative and Policy Studies at the University of Alabama. Jeffrey Cohen is an Assistant Professor in the Department of Political Science at the University of Alabama. Philip B. Coulter is Professor and Chairperson of the Department of Political Science at the University of Alabama. Public Opinion Quarterly Vol. 46:278-284 ? 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-278/$2.50 1 Weeks and Moore (1981) found that an ethnicity-of-interviewer effect can occur among other groups. This content downloaded from 207.46.13.124 on Wed, 22 Jun 2016 05:39:44 UTC All use subject to http://about.jstor.org/terms RACE-OF-INTERVIEWER EFFECTS 279 1981). The presence of a race-of-interviewer effect has been found in both personal interviews and on questionnaires filled out in the presence of an interviewer (Sudman and Bradbum, 1974; Campbell, 1981). In recent years, survey researchers have increasingly relied on telephone interviews to conduct their surveys, primarily because of the lower per unit costs of telephone interviews. The increasing use of this method of interviewing makes it important to determine whether a factor such as race-of-interviewer influences responses in the same manner in telephone interviews as it does in other forms of surveying. There are at least two reasons why race of interviewer may have little or no effect on responses in telephone interviews. First, in telephone interviews respondents cannot see the interviewer and thus may be unable to determine the interviewer's race. As a result, respondents may feel no need to avoid offending the interviewer. Second, even if the respondent can identify the interviewer's race, the greater physical and psychological distance between respondent and interviewer in telephone surveys may reduce any need the respondent might feel to avoid offending the interviewer. A race-of-interviewer effect may, however, be present in telephone surveys. Specifically, during the interview respondents may use verbal cues to identify the race of their interviewers. This ability, along with a sense of politeness, may result in respondents seeking to avoid offending their interviewers, despite the distance between them and the interviewers.

Changes in Americans' Policy Preferences, 1935–1979

Abstract: Using responses to 3,315 survey questions asked of national samples, we examine how policy preferences of Americans have changed over the last 45 years. The data indicate that there has been considerable stability in public opinion: responses to half the 613 repeated policy items show no significant change at all; approximately half the detectable changes were less than 10 percentage points; and rarely did preferences fluctuate significantly back and forth within a short time period. Foreign policy changes were no larger or more frequent than domestic, but they did tend to occur more abruptly. When opinion shifts occurred, they were not random or capricious; they were usually related to important changes in citizens' social and economic environments. Rapid shifts generally coincided with major events in international affairs or the economy. Benjamin I. Page is an Associate Professor of Political Science at the University of Chicago and a Senior Study Director at the National Opinion Research Center. Robert Y. Shapiro is a Ph.D. candidate in political science at the University of Chicago and an Associate Study Director at NORC. The authors are grateful for research support from the National Science Foundation, Grant #SES-7912969 AOI, and wish especially to thank Tom Smith, Patrick Bova, Michael Galati, and John Gillroy for their advice and help. Eric Schmaler, Kathleen Bawn, and Leah Knowlton provided capable assistance. The data were originally collected by the National Opinion Research Center and the American Institute of Public Opinion, or were made available through the Inter-University Consortium for Political and Social Research; the responsibility for analysis and interpretation is our own. An earlier version of this paper was presented at the 1980 Annual Meeting of the Midwest Association for Public Opinion Research, December 5, 1980, Chicago, Illinois. Public Opinion Quarterly Vol. 46 24-42 ? 1982 by The Trustees of Columbia University Published by ElIeviei North-Holland, Inc. 0033-362X/82/0046-24/$2.50 This content downloaded from 157.55.39.112 on Wed, 07 Sep 2016 04:26:41 UTC All use subject to http://about.jstor.org/terms AMERICANS' POLICY PREFERENCES, 1935-1979 25 In the literature of public opinion, some scholars have maintained that opinions are highly labile, especially concerning foreign policy matters (Almond, 1950; but cf. Caspary, 1970). Such a view seems consistent with the notion that the average person is uninterested in politics, unaware of what is going on, and subject to influence by a host of arbitrary forces. Others have suggested that although the opinions of individuals may be volatile (Converse, 1964), the aggregate distributions of preferences are generally quite stable and change slowly (Key, 1961; Erikson and Luttbeg, 1973; Monroe, 1975; Erikson et al., 1980). Generalization has been hampered, however, by reliance on fragmentary or unsatisfactory data. Journalistic accounts have often stressed the instability of public opinion while reporting sudden jumps in the president's popularity rating or wild fluctuations in preelection polls.2 But these data do not gauge changes in polity preferences; the stimuli are not fixed policy alternatives but rather ever-changing politicians about whose latest words and actions new information is constantly available. When policy preferences are actually discussed, alleged fluctuations often represent only sampling error, or are artifacts of variations in question wording, sample design, or the research procedures followed by different polling organizations (see Lipset, 1976). The comparison of responses to even slightly different questions is hazardous; yet scholars have not generally had access to any comprehensive collection of responses to identical policy preference questions, asked repeatedly over time by the same survey organizations (see Cantril, 1951; Hastings and Southwick, 1975; Smith, 1980; Miller et al., 1980, for helpful but limited collections).

Blacks, Whites, and Attitudes Toward Abortion

Abstract: PIP: Patterns of black and white support for abortion from 1972-80 in the US were examined in order to investigate the significance of race in attitudes toward abortion, analyze the extent to which other factors such as religious practice (religiosity) and demographic characteristics affect these racial differences; and determine what changes, if any, occurred in the salience of race on abortion attitudes during this 9-year period The General Social Survey (GSS) conducted from 1972-80 were used These surveys, done annually with the exception of 1979, use national samples of approximately 1500 respondents each year A modified probability sampling design was used in 1972 through 1974, full probability sampling in 1977-80, and a combination of the 2 in 1975 and 1976 6 standard questions tapping abortion attitudes were asked in each of these surveys From these 6 items an additive scale was created, with values ranging from 0, when abortion was opposed in every case, to 6, where approval was given for abortion in each case This scale was used as a dependent variable in the analyses There was a great deal of public stability in public attitudes toward abortion This was particularly the case for the health, rape, and birth defect items, where little change was evident in either race, except for increased black support for abortion when the mother's health is threatened For the other items, support by blacks appears to have increased during this time, while support by whites dropped slightly The amount of convergence was not statistically significant Blacks remained significantly less likely to favor abortion in all 6 instances, they were, by 1978-80, almost equally unlikely to oppose abortion in all cases (9% for whites, 12% for blacks) The mean number of abortion items supported by blacks increased from 31 in 1972-74 to 33 in 1978-80; white support dropped from 40 to 39 in 1978-80 (though returning to 40 in 1980) Much of the difference in support of abortion, though not all, was due to the different demographic characteristics of blacks and whites, and the greater degree of religiosity of blacks Most of the factors hypothesized to affect black attitudes appeared to do so: education; income; urban and northern residence; and lesser religiosity where each related to greater support for abortion There was no direct evidence to support or refute the legitimacy of the "abortion as black genocide" argument, but the fact that opposition to abortion comes from the seemingly more traditional segment of the black community seems to provide some indirect evidence that the argument does not account for the greater black opposition to abortion

Sex Differences in Attitude Expression: A Generational Explanation

Abstract: Females have consistently shown higher DK response rates than males in surveys. Using the 1972 Center for Political Studies National Election Study, this sex difference is shown to be a largely generational phenomenon which is much greater among older than younger respondents. It also declines at higher levels of political knowledge and interest. Finally, using the CPS 1972-76 national panel, DK response rate shows high test-retest reliability. Ronald B. Rapoport is an Associate Professor of Government, College of William and Mary. The data for this study were made available by the Inter-University Consortium for Social and Political Research. Neither the Consortium nor the original collectors of the data are responsible for the analyses or interpretations presented here. The author would like to acknowledge the insightful suggestions of Jere Bruner on this project, and the helpful comments of Alan Abramowitz, Fritz Gaenslen, John McGlennon, Patricia Rapoport, and the anonymous reviewers on an earlier draft of this article. Public Opinion Quar-terly Vol. 46.86-96 ? 1982 by The Trustees of Columbia University Published by Elsevier North-Holland, Inc. 0033-362X/82/0046-86/$2.50 This content downloaded from 157.55.39.127 on Mon, 27 Jun 2016 06:10:40 UTC All use subject to http://about.jstor.org/terms SEX DIFFERENCES IN ATTITUDE EXPRESSION 87 data in their own right, and reveal something important about respon-

Presidential Popularity and Presidential Vote

Abstract: The president's popularity rating is highly predictive of his vote share in a reelection bid, especially when popularity is assessed in June of that year. This June popularity-vote model predicts about as well as the Gallup final preelection poll, and a 50 percent approval rating will ensure reelection. Michael S. Lewis-Beck is a Professor in the Department of Political Science, University of Iowa. Tom W. Rice is a graduate student in the Department of Political Science, University of Iowa. Public Opinion Quarterly Vol. 46:534-537 ? 1982 by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-534/$2.50 This content downloaded from 207.46.13.48 on Wed, 12 Oct 2016 05:48:46 UTC All use subject to http://about.jstor.org/terms RESIDENTIAL POPULARITY AND PRESIDENTIAL VOTE 535 Vt= 30.53 + .41 Pt (1) (3.72) R2=.70 n=8 D-W=2.18 where Vt = president's percentage of the popular vote, Pt = percentage approving of the president's handling of his job in the last Gallup popularity poll before the election, the value in parentheses = the t-ratio, R2 = the coefficient of determination, n = the number of observations (Gallup began asking the popularity item in 1938, and incumbents sought reelection in 1940, 1944, 1948, 1956, 1964, 1972, 1976, and 1980; these data are taken from The Gallup Poll Index, Reports No. 182 and 183, October-November 1980 and December 1980, respectively), and D-W = the Durbin-Watson statistic. Surprisingly, inclusion of the unusual Carter data actually enhances the model. The goodness of fit of the regression is improved and the t-ratio is greater. However, the vote percentages for Nixon and Carter are rather poorly predicted, with errors of 7.01 percentage points and + 4.7 percentage points, respectively. Perhaps the precision of the model could be improved by slight revision, which we now propose. Sigelman measures popularity according to the last such poll before the election. This strategy has intuitive appeal. The difficulty is that these final preelection popularity polls are not always held at the same time, and thus, responses may be influenced by differing short-term forces. We are especially concerned about the impact of the primaries and the conventions, which may cause a momentary, rather large, shift in the president's job rating. To overcome such distortion, it would seem preferable to employ a survey that was close to the election, yet held in a period of relative political calm, after the primaries and before the conventions. The June poll best meets these criteria. Hence, we sought a June measure of presidential popularity, which was possible for all years except 1940 and 1944. (The exclusion of those years may be preferable, since they were war years in which an incumbent was reelected to unprecedented third and fourth terms). Here are the OLS results for this "June model": Vt = 30.80 + .42 Pt (2) (4.80) R2=.85 n=6 D-W= 2.36 where the definitions are the same as with Equation 1, except Pt = the percentage approving of the president in the June Gallup poll prior to the election. Equation 2, with its June indicator of popularity, improves upon Equation 1, with its last preelection poll indicator of popularity, in This content downloaded from 207.46.13.48 on Wed, 12 Oct 2016 05:48:46 UTC All use subject to http://about.jstor.org/terms 536 LEWIS-BECK AND RICE several measurable ways. First, the larger t-ratio supports a higher level of statistical significance for the slope estimate. Further, the explanatory power of the model is clearly heightened, as the increase in R2 from .70 to .85 shows. By observing popularity shifts six months in advance of the election, we can account for 85 percent of the variation in the incumbent president's vote share. The strong linearity of the relationship can be seen in Figure 1, where the regression line is fitted to the scatter of points.

On Finding and Interviewing the Needles in the Haystack: The Use of Multiplicity Sampling

Abstract: THE Vietnam Era Project of the Center for Policy Research has for the last several years been studying the male Vietnam veteran and his nonveteran peers. We collected comprehensive life history data for the purpose of assessing current functioning and the impact of the Vietnam War upon occupational and educational careers, upon interpersonal relations, and upon psychological mood and symptomatology. The veterans in this group constitute a rare and relatively difficult to access population when contrasted with a typical survey population. To draw a representative sample of sufficient size and

The Consumer as Forecaster

Abstract: Consumer confidence, particularly consumer expectations for the short-term future, has proven to be a leading economic indicator. Over the past 15 years it has foretold all major business cycles signaling both the contraction and recovery phases. The buying plans component of the survey program, however, has not demonstrated a consistently convincing forecasting capacity. Fabian Linden is Executive Director, Consumer Research Center, The Conference Board. Public Opinion Quarterly Vol. 46:353-360 ? 1982 by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-353/$2.50 This content downloaded from 40.77.167.82 on Wed, 13 Jul 2016 05:31:07 UTC All use subject to http://about.jstor.org/terms

Opinion Leaders: Is Anyone Following?

Abstract: ARE there opinion leaders in the sense that some individuals within the general public change their opinions earlier than others? Are there followers in the sense of a lagging public, or are "opinion leaders" the only people discussing and making up their minds on the major issues of the day? The General Electric Company has been conducting interviews with national samples of adults by telephone every quarter since 1964. One purpose of these studies is to alert management to changes in public opinion, providing an early indication of new influences that may affect the company. Beginning in 1972 questions were included to identify respondents whose opinions might be expected to change earlier than the total sample. The Katz/Lazarsfeld opinion leadership questions reported in Personal Influence (1955) were adapted to this purpose. Two basic assumptions about how opinions are formed were implicit in this attempt. First, it was assumed that the process of opinion formation is social. When faced with an unclear situation which requires some

Mass and Elite Foreign Policy Opinions

Abstract: THE CONFLICT between strong executive direction and building the necessary consensus for action makes the conduct of foreign policy in a democracy extremely difficult. The expression of public support for foreign policy initiatives is a source of strength in diplomacy, but such support is difficult to gain because the mass public is generally regarded as less informed and unconcerned about foreign policy problems than about domestic issues (Almond, 1950; Caspary, 1970). In fact, the distinction between the foreign and domestic spheres is seen as so great by some scholars that the office of the American chief executive has been described as two presidencies-one oriented to domestic policy and one for foreign policy (Wildavsky, 1966; Cronin, 1980). According to this perspective, the president chooses to exercise initiatives on foreign policy questions precisely because the public is relatively unconcerned and the extreme conflict which erupts between various domestic factions can be avoided. Furthermore, the mass public is viewed as teachable in foreign policy matters; the president can set the agenda, persuade opinion leaders, and lead the

Conformity to AAPOR Standards in Newspaper Reporting of Public Opinion Polls

Abstract: This study examines three newspapers' level of conformity to AAPOR standards of minimal disclosure in their reporting of public opinion polls. Analysis of 116 polls published in the period of 1972-79 indicates a dramatic increase in the number of polls reported, but not in the level of conformity to AAPOR standards. Conformity is higher when the polls concern elections rather than nonelection topics, and when newspapers themselves, rather than external sources, are the sources of the polls. M. Mark Miller is an Associate Professor of Journalism and a member of the Mass Communications Research Center at the University of Tennessee-Knoxville. Robert Hurd is a reporter for the Baraboo (Wis.) News Republic. Public Opinion Quarterly Vol. 46:243-249 ? 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-243/$2.50 This content downloaded from 157.55.39.104 on Mon, 20 Jun 2016 06:19:15 UTC All use subject to http://about.jstor.org/terms

Television and American Culture: The Mass Medium and the Pluralist Audience

Abstract: The study explores the potential influence of television on trends toward cultural homogenization in American society. Depth interviews focused on two variables: the analytic response-viewers' thoughts about the program itself, including, for example, comments on how plot elements relate to the program as a whole or the character of the script or acting, and the interpretive response-viewers' thoughts about the program's relevance to their own lives or broader issues of society and culture. The data support the hypothesis of cultural homogenization revealing similar indices of analytic and interpretive response across educational levels. W. Russell Neuman is Assistant Professor of Political Science and Co-Director of the Research Program on Communications Policy at the Massachusetts Institute of Technology. This article is a revised version of a paper delivered at the Annual Meeting of the American Sociological Association, New York, 1980. Public Opinion Quarterly Vol. 46:471-487 ? 1982 by the Trustees of Columbia University Published by ElsevierScience Publishing Co., Inc. 0033-362X/82/0046-471/$2.50 This content downloaded from 157.55.39.138 on Sun, 26 Jun 2016 05:56:10 UTC All use subject to http://about.jstor.org/terms 472 W. RUSSELL NEUMAN ratings point for one program over the course of a year amounts to a potential 15-million-dollar loss in revenues for the network. Perhaps the most striking aspect of the television phenomenon from a sociological point of view is its universality. Television is socially defined as the culture of the masses. And although professors, professionals, and the educated elite in general may claim to have time only for occasional news, sports, and a play or concert on the public channel, the recurrent implication of systematic research is that they watch the same situation comedies and action adventure programs just about as often as their neighbors in blue collars and hard hats (Wilensky, 1965). As Hirsch has put it, television has come to provide a "centrally produced, standardized and homogenous culture" (1978, 400). Social scientists, however, have devoted surprisingly little attention to the long-term impact of this core of shared experience in our society. To what extent has television become a major force of social integration and cultural uniformity? Are individuals from diverse social and cultural backgrounds really responding in the same ways to these common cultural forms? We find that better educated viewers watch slightly less television than the less well educated-about three minutes less per day for each additional year of formal education-but the types of programs viewed are virtually identical.' Are better educated viewers more attentive and do they draw on extensive exposure to literature, theater, and cinema in responding to the television they see? In an attempt to explore whether people from diverse backgrounds who watch the same program actually see the same program, we designed a depth interview technique based on a series of increasingly structured questions about which elements in a program stand out in viewers' minds and how they evaluate what they have seen. Transcripts of these interviews were systematically coded to derive quantitative measures of the attentiveness as well as the level of analytic and interpretive thinking among various social groups in response to a random sample of prime-time television programming. The analysis focuses on four questions: (1) Is television thought-provoking? (2) Is the response to television differentiated by educational level? (3) Does the level of response vary across program types, or was McLuhan right after all, that the medium pretty much determines the I The three-minutes-per-day figure is derived from Bower (1973), Table 6-2, p. 132. Some studies have suggested that longer working hours and more out-of-house social activities rather than cultural tastes explain the small negative relationship between education and television exposure (Samuelson, et al., 1963). This content downloaded from 157.55.39.138 on Sun, 26 Jun 2016 05:56:10 UTC All use subject to http://about.jstor.org/terms TELEVISION AND AMERICAN CULTURE 473 message? (4) To what extent do the expectations and attitudes of the viewer (the "social definition of television") influence the medium's impact?

A Study of Experimental Techniques for Telephone Interviewing

Abstract: TELEPHONE surveys are not a new phenomenon, but academic and governmental organizations have only recently made serious investments in the method. As of this writing, for example, the Bureau of the Census has conducted only one study involving "cold" telephone interviews-a random-digit dialing sample of Michigan residents. Reducing the cost of surveys is the primary motivation for developments in telephone interviewing. Many researchers still see the telephone interview as the somewhat disreputable "poor relation" of the personal contact. If it were not for financial exigencies, these people would probably not view the telephone as a preferable (or even a viable) means of collecting survey data. This attitude may be one reason why we know little about what constitutes effective telephone interviewing techniques. The telephone survey has been treated as a necessity, rather than as an

House Effects and the Reproducibility of Survey Measurements: A Comparison of the 1980 GSS and the 1980 American National Election Study

Abstract: This paper examines a collaborative experiment between the General Social Survey, National Opinion Research Center, and the American National Election Study, Center for Political Studies. Identical items were asked by both surveys, thereby permitting a test of reproducibility and house effects. Significant differences were found between the houses, with the largest and most systematic differences involving item nonresponse. Various explanations for the differences are examined and a general assessment of the importance of the differences is presented. Tom W. Smith is a Senior Study Director, National Opinion Research Center. This research was done for the General Social Survey project directed by James A. Davis. The project is supported by the National Science Foundation, Grant No. SOC77-03279. Public Opinion Quartedly Vol. 46 54-68 ? 1982 by The Tiustees of Columbia University Published by Elseviei Noith-Holland, Inc. 0033-362X/82/0046-54/$2.50 This content downloaded from 157.55.39.255 on Mon, 01 Aug 2016 06:02:33 UTC All use subject to http://about.jstor.org/terms

The Effect of Cost and Importance Factors on the Return Rate for Single and Multiple Mailings

Abstract: Employing an experimental design, the effectiveness of cost and importance factors on the return rate for mail questionnaires is analyzed. Importance factors, notably certified mail, are shown to be of greatest consequence. Kent L. Tedin is an Associate Professor and C. Richard Hofstetter is a Professor in the Department of Political Science, University of Houston. Public Opinion Quarteily Vol. 46 122-128 ? 1982 by The Tiustees of Columbia University Published by Elsevier North-Holland. Inc. 0033-362X/82/0046-122/$2.50 1 Importance factors are distinguished from "salience factors" in that the latter indicates the salience of the substance of the survey to the person taking it. Herberlein and Baumgartner (1978) show that return rates increase if the content of the questionnaire is likely to be of interest to the respondent. Of course, investigators are not usually in a position to alter the content of a study for the purpose of increasing the rate of return. This content downloaded from 207.46.13.109 on Sun, 24 Apr 2016 06:35:09 UTC All use subject to http://about.jstor.org/terms SINGLE AND MULTIPLE MAILINGS 123 percent, respectively. Other importance factors found to be effective include the use of first-class postage, certified mail, and various methods of personalizing the communication, such as addressing the respondent by name and hand signing the letter (Blumberg et al., 1974; Linsky, 1975; Carpenter, 1974-75). Costs involve the time and effort a respondent must devote to completing the questionnaire. As costs go down, return rates should go up. In practice, cost and importance factors sometimes interact yielding results counter to expectations, For example, return rates should be higher for short than for long questionnaires-it costs less to complete the former. But long questionnaires may be seen as being more important, which counteracts some cost factors. Thus, Herberlein and Baumgartner (1978) find no zero-order relationship between questionnaire length and return rates, but find under controlled circumstances that longer questionnaires have lower return rates. The most effective cost reduction factor seems to involve offering monetary incentives. The respondent is compensated for the costs involved in filling out the questionnaire. Armstrong (1975), in a literature review, suggests a straightforward monotonic relationship between monetary reward and return rates-the greater the reward the higher the return rate. Others, however (Linsky, 1965; Schewe and Cournoyer, 1976; Cox, 1976), have suggested that certain discrete amounts are optimal. While considerable research has been done on mail questionnaire return rates, much of it involves a single contact, two-cell experiment which varies just a single factor, for example, the type of postage or the color of the paper. It may well be that a stimulus which is effective on an initial contact is not effective when used as part of a second or third contact. One exception to the two-cell experiments is an ex post facto design by Heberlein and Baumgartner (1978), in which the unit of analysis is the reported percentage return in other studies. The effect of more than one stimulus is then analyzed using regression. However, a more desirable approach to determining the effect of multiple contacts and multiple stimuli is a true experimental design. In this paper we report on an experiment involving two importance factors (multiple contacts, certified mail vs. first-class postage) and a cost factor (a monetary incentive) to determine the least expensive combination which yields the highest rate of return.

RRT Meets RDD: Statistical Strategies for Assuring Response Privacy in Telephone Surveys

Abstract: Despite its growing role in survey research, the telephone survey has been largely neglected with respect to the design and testing of statistical procedures for assuring response privacy (e.g., randomized response techniques). This paper demonstrates that such procedures are no less feasible to administer in telephone surveys than in face-to-face surveys. Both randomized response techniques and alternatives to randomized response are considered.

Public Attitudes Toward Life and Death

Abstract: Opinion trends in this country indicate sharp divisions in public sentiment over a number of life-taking actions. While legal abortion and capital punishment clearly head a list, a number of other issues have gained national attention in recent years. The present paper explores the structure of belief systems giving rise to normative conflicts of this kind. Of particular interest is the notion of a "pro-life" or other generic life orientation (e.g., the alleged "right-to-die" orientation of those who favor "mercy killings" in the case of terminally ill patients) as a possible explanation for public attitudes toward specific issues such as suicide and euthanasia. The present analysis assesses the empirical claims associated with such a model. The results offer qualified support for the existence of generic value orientations as revealed by public attitudes toward legal abortion, suicide, euthanasia, and capital punishment. Language: en

The Polls: Attitudes Toward American Jews

Abstract: CONGRESSIONAL debate in the fall of 1981 over the sale of AWACs to Saudi Arabia evoked heated public discussion, including some highly critical remarks in the media about the "Jewish lobby" and Israel's "veto" over American foreign policy. Coupled with news coverage of increasing vandalism of Jewish institutions, such stories have led some observers to conclude that there is a resurgence of antisemitism in the United States. Is there a resurgence? What do studies about the attitudes of Americans toward Jewish fellow Americans tell us? Surveying attitudes toward Jews in the United States started in the 1930s, with the triumph of the Nazis in Germany and their vigorous propaganda abroad. American Jewish organizations and individuals wanted to learn the extent of antisemitism in this country and to develop ways to counteract it. By the early 1960s over a hundred surveys of attitudes toward American Jews had been conducted. These were compiled and analyzed in Jews in the Mind of America.I Charles Herbert Stember, the editor, concluded that antisemitism was "a disappearing problem in the United States." But to this day is has not disappeared entirely. In October 1964 the Survey Research Center of the University of California at Berkeley conducted the first national survey to assess the extent and nature of contemporary antisemitism. The results were published as The Tenacity of Prejudice2 and dealt with attitudes toward discrimination in its various forms, social distance, and traditional antisemitic beliefs. Gertrude Selznick and Stephen Steinberg, who edited the volume, developed an index of 11 antisemitic beliefs which has served as a point of re-

Locating a Special Population Using Random Digit Dialing

Abstract: Frequently, researchers are only interested in specific subgroups of the total population. Waksberg (1978) has designed a two-stage method of selecting equalprobability samples of the general population using random digit dialing. The present research proposes and illustrates a modification of Waksberg's procedure to locate special populations which cluster geographically. Situations arising from the proposed modifications are discussed and alternative solutions suggested. Johnny Blair is the Head of Sampling at the Survey Research Laboratory, University of Illinois, Urbana-Champaign. Ronald Czaja is the Assistant Director of the Survey Research Laboratory and Assistant Professor at the School of Public Health, University of Illinois, Chicago. The authors wish to thank the many colleagues at the Survey Research Laboratory who provided helpful suggestions. Public Opinion Quarterly Vol. 46:585-590 ? 1982 by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-585/$2.50 This content downloaded from 157.55.39.144 on Wed, 07 Sep 2016 04:25:27 UTC All use subject to http://about.jstor.org/terms

Effects of Presenting One Versus Two Sides of an Issue in Survey Questions

Abstract: Data from a split-ballot experiment show that offering respondents an alter- native position on an issue not only affects the marginals, it also influences whether respondents will give an opinion at all. Furthermore, the study demonstrates that these form effects can occur despite the use of filter questions which theoretically screen out those who tend to be most susceptible to such effects: the less educated or uninformed. The analysis does provide evidence, however, that less educated respondents are indeed more affected by differences in question format and that they are much more likely to "acquiesce" to one-sided agree/disagree forms. In discussing the results the authors develop an information-processing model of question form effects and a meth- odological strategy to generate further research on a much-needed theory of the survey instrument. George F. Bishop is a Senior Research Associate, Robert W. Oldendick is Assistant Director, and Alfred J. Tuchfarber is Director of the Behavioral Sciences Laboratory at the University of Cincinnati's Institute for Policy Research. The research reported here was supported by a grant from the National Science Foundation (SOC78-07407). The authors want to thank Howard Schuman and Stanley Presser for their comments on an earlier version of this paper.

The Effects of Branching on Item Nonresponse

Abstract: THE effect of branching on item nonresponse is an important research issue, in that it deals with an aspect in the design of mail questionnaires which is, in some situations, discretionary on the part of the researcher. There are many situations in which a researcher may choose between asking a two-step question with a branch and asking a single question with an expanded list of responses. For example, an ''own" versus "rent" question might be followed by a question for "owners" only on type of housing. Renters would be instructed to branch around the second question. Alternatively, these two questions could be combined into a single more complex question with responses of: own-single-family house; own-other type of dwelling; and rent. While this eliminates the branch, it correspondingy increases the complexity of the question, thereby creating the potential for ambiguity. This apparently leaves the researcher with a choice between the two approaches. However, the literature on item nonresponse does not indicate whether branching actually reduces item response rates, and, if it does, under what conditions. The purpose of this study is to investigate the effect of branching on item nonresponse.

Continuities in Support for Racial Equality, 1972 to 1976

Abstract: Capitalizing upon the panel component of the 1972-1976 National Election Study, we explore change in white Americans' support for racial equality. Two general propositions regarding individual change are examined: frustration-aggression theory, which attributes change to the stresses and strains of private life; and realistic group conflict theory, which attributes change to the tangible threats blacks pose to whites' private interests. We find intermittent support for the first and virtually no support for the second. Much more impressive is the durability of opinion on racial equality, the insulation of racial opinion from the ostensibly powerful predicaments of private life. Donald R. Kinder is Associate Professor of Political Science at the Institute for Social Research, The University of Michigan. Laurie A. Rhodebeck is Lecturer in the Department of Government and International Studies, University of Notre Dame. Data for this paper were provided by the Inter-University Consortium for Political Research, University of Michigan. The Consortium bears no responsibility for our interpretations or conclusions, nor do Shanto Iyengar, Steven Rosenstone, Blair Wheaton, or two anonymous reviewers, who offered valuable advice. Correspondence regarding this paper should be sent to Donald R. Kinder, Department of Political Science, The University of Michigan, Ann Arbor, Michigan, 48106. Public Opinion Quarterly Vol. 46:195-215 ? 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-195/$2.50 This content downloaded from 157.55.39.123 on Mon, 18 Jul 2016 05:47:42 UTC All use subject to http://about.jstor.org/terms 196 KINDER AND RHODEBECK For all its political significance, public opinion on racial matters remains mysterious. Why do some whites grow more sympathetic toward racial equality while others become more hostile? What sorts of experiences enhance or relieve interracial tensions? By rummaging through the splawling literature, it is easy to find answers-but answers backed by persuasive evidence are hard to come by. These days especially, benign neglect applies rather well to theories of racial opinion. In a modest way, our paper begins to redress this neglect. Here we test a pair of eminently plausible and quite general theories: frustration-aggression theory, which ascribes change in whites' support for racial equality to the stresses and strains of their private lives; and realistic group conflict theory, which ascribes change to the threats blacks seem to pose to whites' tangible self-interests. Data well suited to test these two explanations we take from the panel component of the 1972-1976 National Election Study. Estimation proceeds via a maximum likelihood procedure, as described in detail below. The implications of our results-for an adequate social psychology of racial opinion and for an understanding of the political entanglements of race-are the business of the paper's final section. Two Theories of Opinion Change Frustration-aggression theory has its roots in Freud's familiar assertion that the frustrations of modern life lead inexorably to hostility. In Frustration and Aggression (1939), Dollard and his colleagues develop this point. According to their treatment of displacement, the frustrations whites experience in their private lives, though they have no manifest racial content, nevertheless take a racial form. In particular, we assume here that private frustrations feed political resistance. The trials and tribulations of whites' modern lives undermine the support they would otherwise lend to policies designed to bring about racial equality. Such indirect expressions of aggression serve, as Allport (1954:325) put it, "to soften the disappointments . . . of life." Research on displacement has concentrated on the frustrations associated with downward social mobility, with uneven results. Some researchers find the downwardly mobile to be disproportionately prejudiced (e.g., Bettelheim and Janowitz, 1964); others do not (e.g., Hodge and Treiman, 1966). The evidence is no better for the more general proposition that discontent in personal life contributes to racial hostility. Alienation and social isolation, economic grievances, worries over safety, concern for the quality of children's schoolingThis content downloaded from 157.55.39.123 on Mon, 18 Jul 2016 05:47:42 UTC All use subject to http://about.jstor.org/terms SUPPORT FOR RACIAL EQUALITY, 1972-1976 197 none of these personal frustrations appears to drive racial prejudice (Campbell, 1971; Kinder and Sears, 1979; Sears and Kinder, 1971). So the empirical record of the displacement hypothesis is checkered. However, all of the evidence cited above is drawn from crosssectional comparisons.1 A better test of displacement requires panel evidence of the kind we marshal below. If the displacement hypothesis is correct, then whites who have run into serious difficulties in their private lives should grow increasingly hostile toward racial equality. In the panel tests that follow, we concentrate on two ostensibly powerful sources of frustration: economic adversity and vulnerability to crime and violence. We will see whether whites who have suffered economic dislocation or who feel increasingly unsafe on the streets of their own neighborhoods become less willing to support policies promoting racial equality. Realistic group conflict theory (Cummings, 1980; LeVine and Campbell, 1972; Rothbart, 1976), the second perspective on racial opinion examined here, locates the causes of racial prejudice in the competition between blacks and whites for scarce resources. Competitive interdependence produces the realistic perception of threat, which in turn leads to hostility directed at members of the threatening group. Thus whites develop antagonistic attitudes toward blacks to the extent that blacks are seen to threaten their private interests. Plausible as realistic group theory may seem, its empirical support turns out to be anemic. There is not much evidence one way or the other, and what there is often runs counter to expectation. Moreover, tests of realistic group conflict theory are almost always indirect, usually invite alternative interpretations, and in any case are of dubious generality (this literature is reviewed in Ashmore and DelBoca, 1976). Here we try to avoid these shortcomings. Again using panel evidence, we examine the responsiveness of opinion toward racial equality to the tangible threats that blacks seem to pose to whites' private interests. In particular, we will see whether whites who think themselves besieged by black street crime, or menanced by neighborhood desegregation, or personally implicated in racial busing schemes, become less enthusiastic about racial equality. Model Specification and Estimation Procedure Tests of social psychological propositions regarding change in an individual's support for racial equality require individual-level eviI Some support for the displacement hypothesis has been furnished by experimental research. However, such experiments have proven difficult to replicate. For a capable review of this research, see Ashmore and DelBoca, 1976. This content downloaded from 157.55.39.123 on Mon, 18 Jul 2016 05:47:42 UTC All use subject to http://about.jstor.org/terms 198 KINDER AND RHODEBECK dence, preferably longitudinal. With panel evidence provided by the 1972-1976 National Election Study we will assess the stability of whites' racial opinions and, of greater interest, appraise how adequately the frustration-aggression and realistic group conflict theories account for deviations from stability. In schematic form, the model that expresses both these interests appears in Figure 1. Stability in racial opinions at the individual level is represented by the arrow from Racial Equality72 to Racial Equality76. Tests of the displacement and the realistic group conflict theories are represented in the diagram by the arrow running from Personal Experience to Racial Equality76. Together, the two theories suggest three types of personal experience that should prompt whites to reduce their support for racial equality: economic dislocation, vulnerability to crime and violence, and exposure to racial threats. In the analysis that follows, we will examine each separately.2 We will also treat separately the various components of support for racial equality. Table 1 presents the questions used to measure such e72 e72 * e e76 e76 ... e 721 722 ... 72 761 762 ... 76, Fg R aci al Equaliey72 e Rt Racal EqualitytY| |Personal Experience Figure 1. The Impact of Personal Experience on Support for Racial Equality 2 The reader may legitimately object to this procedure. Taking up these three types of experience one at a time raises the specter of specification error: that because we ignore the simultaneous effects of the three, our estimates of each taken separately are necessarily biased. In principle, we agree completely. But in our case, estimates based on separate analyses do not differ from those based on a more elaborate analysis that takes into account simultaneous effects. Consequently, for the sake of presentational clarity, we have chosen to present the results in three parts. This content downloaded from 157.55.39.123 on Mon, 18 Jul 2016 05:47:42 UTC All use subject to http://about.jstor.org/terms SUPPORT FOR RACIAL EQUALITY, 1972-1976 199 Table 1. White Support for Racial Equality, 1972 and 1976

From Parties to Candidates: Examining the Role of the Media

Abstract: This article examines the relationship between campaign media expenditures and the saliency of attitudes about parties and candidates in the electorate using data from the 1978 National Election Study and the Traugott/Goldenberg study of campaign managers. A negative relationship is found between media spending and party saliency. In contrast, a strong positive correlation is shown to exist between media expenditures and candidate saliency. These relationships are found to be accentuated when the party organizations in the congressional district are weak and where political action committees contribute a large proportion of the total campaign expenses. Martin P. Wattenberg is a Visiting Assistant Professor at the University of California, Los Angeles. The author wishes to thank Michael Traugott and Edie Goldenberg for making the data from their 1978 survey of congressional campaign managers available for this analysis. They, of course, bear no responsibility for the analyses or interpretations presented here. Data from the 1958 and 1978 National Election Studies were provided by the Inter-University Consortium for Political and Social Research. Public Opinion Quarterly Vol. 46:216-227 ?) 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-216/$2.50 This content downloaded from 157.55.39.181 on Thu, 29 Sep 2016 05:51:07 UTC All use subject to http://about.jstor.org/terms PARTIES TO CANDIDATES: ROLE OF THE MEDIA 217 frequently stated in the literature (e.g., Sorauf, 1980: 255). Yet the evidence for such an important assertion has remained almost entirely impressionistic. The goal of this paper is to remedy this deficiency by describing the extent of the relationship and the conditions which

Public Attitudes Toward Freedom of the Press

Abstract: This article presents findings on a study of public attitudes toward the rights and responsibilities of the media. Public thinking about freedom of the press appears to be dominated by a principle of fairness. Specifically, people want and expect the communications media to provide equal coverage to opposing political candidates and to present both sides of controversial issues. Despite other studies showing a lack of political tolerance, the public is willing to permit considerable diversity in the points of view that are presented in the media. Although the public rejects many proposals for censorship laws, it does support "fairness-enhancing" laws. The authors hypothesize that the public's perspective on freedom of the press grows out of its role as listeners rather than as speakers. John Immerwahr is an Associate Professor of philosophy at Villanova University, and a research fellow of the Public Agenda foundation. John Doble is a Senior Research Associate at the Public Agenda Foundation. This article was adapted from The Speaker and the Listener: A Public Perspective on Freedom of Expression (New York: Public Agenda Foundation, 1980), by John Immerwahr, Jean Johnson, and John Doble, The research was supported by a grant from the John and Mary R. Markle Foundation. Daniel Yankelovich, President of the Public Agenda, supervised the original project, and the authors are indebted to him for his help and suggestions. Jean Johnson read and commented on earlier drafts of this article. For more detailed information about the findings, contact the Roper Data Center, Storrs, Connecticut 06268. Public Opinion Quarterly Vol. 46:177-194 ? 1982 by The Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-177/$2.50 This content downloaded from 207.46.13.100 on Thu, 02 Mar 2017 04:15:24 UTC All use subject to http://about.jstor.org/terms 178 IMMERWAHR AND DOBLE

Enjoyment of Work by Full-Time Workers in the U.S., 1955 and 1980

Abstract: Two questions concerning enjoyment of work asked on a 1955 American Gallup Poll were asked on a 1980 United States national survey to gauge the net change during a quarter of a century. Only the responses of persons employed full-time were used for this study. Indicated enjoyment of work was substantially lower in 1980 than in 1955. The difference was especially great for manual workers, Protestants, and older persons. A tentative conclusion drawn from examining trend data for one birth cohort is that the change resulted from cohort succession and from orientations and attitudes members of younger cohorts brought to their work rather than from changes in work conditions which affected workers of all ages and in all birth cohorts. Some implications of the findings are disscussed. Norval D. Glenn is Professor of Sociology, University of Texas, Austin. Charles N. Weaver is Professor of Management, St. Mary's University, San Antonio. The research reported here was supported in part by National Science Foundation Grant SES7907917. The 1955 data on enjoyment of work were obtained from the Roper Center at the University of Connecticut. The authors are indebted to the Gallup Organization, Inc., and especially to James Shriver, III, for cooperation in gathering the 1980 data on enjoyment of work. Data from the 1978 and 1980 General Social Surveys, conducted by the National Opinion Research Center with funds from the National Science Foundation, were also used in the research. Public Opinion Quarterly Vol. 46:459-470 ?3 1982 by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-459/$2.50 This content downloaded from 207.46.13.142 on Sat, 28 May 2016 04:45:21 UTC All use subject to http://about.jstor.org/terms 460 GLENN AND WEAVER American national surveys since 1958.1 Quinn et al. (1974) reanalyzed job satisfaction data gathered by 15 national surveys from 1958 to 1973 and concluded that there was no convincing evidence for a decline in job satisfaction during that period. Katzell (1979) extended the series by adding data from surveys conducted in 1974, 1975, 1976, and 1977, and he still found no evidence for any substantial change. Data from the General Social Surveys conducted in 1977, 1978, and 1980 (Davis, 1980) show an aggregate level of reported job satisfaction at about the same high level at which it has existed for over 20 years.2 On the other hand, survey data other than responses to the global job satisfaction question do indicate a recent decline in satisfaction with work. Quinn and Staines (1979), using a composite index based on responses to several questions about specific aspects of job satisfaction asked on the Quality of Employment Surveys conducted by the Survey Research Center at the University of Michigan, found a distinct decline in the mean index scores from 1973 to 1977, and commercial opinion polls also found some changes in attitudes toward work during the 1970s.3 As a whole, the relevant survey data are rather inconsistent and may seem confusing (even though the different data can be reconciled with one another). Data on behavioral indicators of work orientation, such as rates of absenteeism and job turnover, do not make the picture much clearer, because data can be found either to support or to refute the alleged decline in commitment to work (Strauss, 1974). I The most common wording of the question has been: "On the whole, how satisfied are you with the work you do-would you say you are very satisfied, moderately satisfied, a little dissatisfied, or very dissatisfied?" However, the Gallup version of the question only asked the respondents whether they were satisfied or dissatisfied, and thus the responses to the different variants of the question can be made comparable only by collapsing all responses into "satisfied" and "dissatisfied." 2 Data from the 1974, 1975, and 1976 General Social Surveys are included in Katzell's series (see his Table 1). The percentage of GSS respondents who said they were satisfied was 87 in 1977, 87 in 1978, and 83 in 1980. The apparent decline from 1978 to 1980 was probably real (p = .060-two-tailed), but the 1980 level was no lower than that in some other years since 1958, including the 81 percent in 1958, and it was only slightly lower than the 85 percent in 1974. Data on the four cohorts studied by the National Longitudinal Surveys show a decline in job satisfaction in each cohort from 1966 to 1972 (Andrisani, 1978), but since data from repeated cross-sectional surveys show little change during that period, the meaning of the changes shown by the panel data is unclear. The changes could be age effects, but the usual inference made from cross-sectional data is that aging, or some close correlate such as increasing extrinsic job rewards, typically leads to an increase in job satisfaction (Glenn et al., 1977; Wright and Hamilton, 1978). One might even speculate that the repeated interviewing of the panel members on topics relating to jobs and work sensitized them to deficiencies in their work situations and led to a decline in their mean job satisfaction. 3 For a summary of some of the poll data, see Mindell and Gorden (1981). This content downloaded from 207.46.13.142 on Sat, 28 May 2016 04:45:21 UTC All use subject to http://about.jstor.org/terms ENJOYMENT OF WORK, 1955 AND 1980 461 If there have been important recent changes in American workers' orientations to their jobs and work, responses to the global job satisfaction question obviously do not reflect them. Perhaps, as several authors have suggested, the stability in the global job satisfaction data reflects offsetting trends in satisfaction with different aspects of work. For instance, to use Herzberg's well-known distinction (Herzberg, 1966), satisfaction with extrinsic rewards of work (pay, prestige, etc.) may have increased while satisfaction with intrinsic rewards (direct enjoyment of work) has declined. Although data from the Quality of Employment Surveys-conducted in 1969-1970, 1973, and 1977-enable one to distinguish between satisfaction with extrinsic and intrinsic rewards of work (and to distinguish among other kinds of satisfaction as well), comparable data from national samples are not available for dates earlier than 1969-1970. Therefore, there have been no available American national data with which to trace the level of such variables as enjoyment of work during the past quarter of a century or for any period

Classifying Undecided Voters in Pre-Election Polls

Abstract: The treatment of "don't know," "no opinion," or other nonsubstantive responses is a problem in many consumer research surveys. This paper looks at the problem in the context of 1980 presidential election opinion polls. During the campaign, a relatively large proportion of those interviewed said that they were undecided as to whom they planned to vote for. Discriminant analysis is used to allocate undecided voters to candidates. The method is validated by a postelection follow-up survey. Ian Fenwick is Associate Professor of Marketing at York University. Frederick Wiseman is the Patrick F. and Helen C. Walsh Research Professor in the College of Business Administration at Northeastern University. John F. Becker is President and James R. Heiman Vice-President of Becker Research Corporation, Boston. This research was supported in part by a grant from the Dean's Research Fund, College of Business Administration, Northeastern University. The authors express their appreciation to the referees for their helpful suggestions. An earlier version of this paper was presented at the annual conference of the Association of Consumer Research, October 1981. Public Opinion Quarterly Vol. 46:383-391 ? 1982 by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/82/0046-383/$2.50 This content downloaded from 157.55.39.255 on Mon, 01 Aug 2016 05:45:06 UTC All use subject to http://about.jstor.org/terms 384 FENWICK, WISEMAN, BECKER, AND HEIMAN illustrated by applying the new technique to data collected by the Becker Institute from an October 1980 poll of registered voters in Massachusetts. Further evidence is provided by an empirical validation study using data obtained in a November 1980 postelection survey of a sample of previously undecided voters.