Determinants of social desirability bias in sensitive surveys: a literature review

Survey questions asking about taboo topics such as sexual activities, illegal behaviour such as social fraud, or unsocial attitudes such as racism, often generate inaccurate survey estimates which are distorted by social desirability bias. Due to self-presentation concerns, survey respondents underreport socially undesirable activities and overreport socially desirable ones. This article reviews theoretical explanations of socially motivated misreporting in sensitive surveys and provides an overview of the empirical evidence on the effectiveness of specific survey methods designed to encourage the respondents to answer more honestly. Besides psychological aspects, like a stable need for social approval and the preference for not getting involved into embarrassing social interactions, aspects of the survey design, the interviewer’s characteristics and the survey situation determine the occurrence and the degree of social desirability bias. The review shows that survey designers could generate more valid data by selecting appropriate data collection strategies that reduce respondents’ discomfort when answering to a sensitive question.     

Has the Internet changed the wage structure too?

By using recent Current Population Survey (CPS) data sets, this paper investigates whether workers who use the Internet at work earn a higher wage than otherwise similar workers who do not use the Internet at work. Estimates suggest that there was a substantial premium to Internet use in 1997 of about 8%. However, this premium disappeared very quickly thereafter, and there are grounds to believe that it is now negative. This result is in stark contrast with that for computer use.     

苏州市中青年购买健康险产品消费行为研究

本文主要针对苏州市的年龄在20岁～60岁间的居民为研究对象,研究苏州市中青年消费者对于购买商业健康保险的购买行为。研究后发现,各生活形态集群对于未购买原因有比较显著的差异,而对于购买动机,给付偏好、购买评估准则以及产品利益等则基本相同。因而,对各大保险公司来说,针对不同的子市场执行相应的保险营销策略,则有利于提高消费者的购买意愿,提高投保率。     

Stereotypes, Asian Americans, and Wages: An Empirical Strategy Applied to Computer Use at Work

This article examines the effect on wages of the Asian-American stereotype as mathematically and technically adept, and the role this stereotype may play in explaining racial wage differences. We propose an empirical strategy to examine the influence of stereotypes on labor market outcomes, with a specific application to the wage premium associated with computer use at work. Using Current Population Survey data, ordinary least squares estimates do not provide compelling evidence that a positive stereotype affects wages for Asian Americans.     

Measuring the Measurable: Why can’t we Agree on the Number of Telecommuters in the U.S.?

Using telecommuting as a case study, we demonstrate that definitions, measurement instruments, sampling and sometimes vested interests affect the quality and utility even of seemingly objective and “measurable” data. Little consensus exists with respect to the definition of telecommuting, or to possible distinctions from related terms such as teleworking. Such a consensus is unlikely, since the “best” definition of telecommuting depends on one’s point of reference and purpose. However, differing definitions confound efforts to measure the amount of telecommuting and how it is changing over time. This paper evaluates estimates of the amounts of telecommuting occurring in the U.S. obtained from several different sources: the U.S. Census, the American Housing Survey, several Work at Home supplements to the Current Population Survey, a series of market research surveys, and the trade association-sponsored Telework America surveys. Many of the issues raised here are transferable to other contexts, and indirectly serve as suggestions for improving data collection in the future.     

Wage determination in the union and nonunion sectors.

This study contains estimates of wage equations for white male union and nonunion employees. The authors find that nonunion wages are generally more responsive than union wages to individuals' education and experience and to regional price-level variation. Despite those differences, however, estimates of union-nonunion wage differentials based on these separate equations do not differ greatly from a differential obtained from a union dummy variable in an equation based on combined union and nonunion observations. Union-nonunion differentials vary widely across occupational groups and are generally larger in the lower skilled and more highly unionized occupations. The results for manufacturing, for which additional industry data are available, indicate a negative impact of high concentration ratios on the wages of all workers and a greater impact of establishment size on nonunion than on union wages. Data were drawn from the May 1973 Current Population Survey.     

The Meanings of ‘Yes’ and ‘No’. An Explanation for the Forbid/Allow Asymmetry

Survey questions worded with the verb ‘forbid’ prove not to elicit opposite answers to equivalent questions worded with the verb ‘allow’ (Rugg 1941). Although ‘forbid’ and ‘allow’ are generally considered each other’s counterparts, respondents rather answer ‘no, not forbid’ than ‘yes, allow’. In order to find out which question is a more valid measure of the underlying attitude, this asymmetry in the answers has to be explained. Experiments show that the asymmetry arises because respondents translate similar attitudes differently into the answering options to forbid/allow questions are equally valid, but the way the attitudes are expressed on the answering scale differs due to the use of ‘forbid’ or ‘allow’. How does this translation process work? The leading hypothesis in forbid/allow research predicts that respondents holding moderate opinions feel that ‘yes forbid’ and ‘yes allow’ are very extreme, causing moderate respondents to prefer answering ‘not forbid’, or ‘not allow’. This article presents the results of 10 experiments investigating the meanings of the answering options to forbid/allow questions. Extreme connotations are shown to only provide part of the explanation for the occurrence of the forbid/allow asymmetry. In order to describe the answering process for forbid/allow questions, well-definedness of meanings proves to be an important additional factor. The meanings of answering options to allow questions are ill-defined compared of those to forbid questions, which causes allow questions to be less homogeneous measures of the underlying attitude than forbid questions.     

The impact of unemployment insurance taxes on wages

This paper examines the incidence of state unemployment insurance taxes on wages paid to workers in various demographic groups. The empirical work matches state level measures of unemployment insurance tax and benefit variables to data aggregated from the Current Population Survey on worker earnings for the period 1992 to 2002. Econometric evidence presented in the paper supports the paper's main hypothesis that UI tax rate structure has its greatest adverse impact on less mobile workers (prime age married women and young workers) and little impact on more mobile workers (prime age men).     

Retrospective questions: data quality,task difficulty,and the use of a checklist

Data from a national panel study with waves in 1987 and 1991 (N=1257), present evidence of the low accuracy of responses to retrospective questions, concerning both attitudes and behaviour. Applying a split ballot design, it is investigated whether using a checklist improves the response accuracy for a retrospective question about one single event: how respondents (N=363) did obtain the job they had four years ago. Furthermore interaction effects of ‘task difficulty’ are examined. The response accuracy indeed increases by using a checklist; however, this increase is not statistically significant. The expected increase of the checklist effect with higher ‘task difficulty’ appears for longer recall intervals, but not for more frequent changes of jobs. It turned out that for male respondents all the predicted effects are indeed clearly present. But for female respondents the checklist appeared to have no effect, irrespective of the task difficulty. A tentative explanation for this unexpected gender effect is suggested by pointing at indications that the male and female respondents differed in the way they obtained their job.     

Disability and the labor force participation of older men in Canada

I use a latent variable model to estimate the effect of disability status on the labor force participation of older men in Canada, with data from the National Population Health Survey. The estimates from this paper indicate that disability status has a large negative effect on the labor force participation of older Canadian males. It is difficult to compare these estimates with previous Canadian estimates because of differences in methodology and data. However, my estimates appear to be larger than those from US data. Some of the implications of these estimates for disability policy are also discussed.     

Accuracy of self-reports to the burdensome question: survey response and nonresponse error trade-offs

Survey self-reports to questions involving respondent burden are examined for: (1) overall accuracy, (2) direction of bias, and (3) influence on relationships. Self-reports by respondents to questions on an area of household finances are compared to error-free supplier records. The evidence shows self-reports in this case to be accurate at the overall, or aggregate, level, but inaccurate at the individual, or disaggregate, level. The trade-offs inherent in the nonresponse occurring as a consequence of attempting to acquire verified data versus response bias resulting from a reliance on self-reports from surveys are discussed. It is suggested that attempts by researchers to acquire error-free, verified information in surveys will lead in most cases to unacceptably high levels of item nonresponse. An alternative, algebraic procedure for adjusting for disaggregate level response bias in self-reports on the basis of a subsample of verified responses is provided.     

Health insurance and retirement of married couples

Most health insurance in the USA is provided by employers until eligibility for public health insurance (Medicare) begins at age 65. Retiring before 65 exposes workers who lack retiree health insurance coverage to the risk of catastrophic medical expenditure. We solve and estimate a dynamic model of the employment behavior of older married couples that includes risky medical expenditure and health insurance. Parameter estimates imply that the risk‐reducing feature of health insurance can account for about half of the observed association between retiree health insurance and employment for married men, but can account for only one tenth of the much larger observed association for married women. Policy simulations imply very small effects on employment of changing the age of eligibility for Medicare from 65 to 67. Copyright © 2006 John Wiley & Sons, Ltd.     

Flexible Working Hours, Family Responsibilities, and Female Self-Employment

ABSTRACT Using cross-sectional data from the Contingent Work Survey of the February 1995 Current Population Survey, we present direct evidence that there are substantial gender differences in the reasons why individuals become self-employed. In particular, women–especially women with young children–are more likely than men to cite flexibility of schedule and family-related reasons for becoming self-employed. Men's reasons for becoming self-employed show little association with their parental status. Our findings suggest that employers should be encouraged to offer working conditions that are more friendly to workers whose family obligations conflict with traditional, forty-hour-per-week jobs.     

Localised Estimates of Dynamics of Multi‐dimensional Disadvantage: An Application of the Small Area Estimation Technique Using Australian Survey and Census Data

Deep and persistent disadvantage is an important, but statistically rare, phenomenon in the population, and sample sizes are usually not large enough to provide reliable estimates for disaggregated analysis. Survey samples are typically designed to produce estimates of population characteristics of planned areas. The sample sizes are calculated so that the survey estimator for each of the planned areas is of a desired level of precision. However, in many instances, estimators are required for areas of the population for which the survey providing the data was unplanned. Then, for areas with small sample sizes, direct estimation of population characteristics based only on the data available from the particular area tends to be unreliable. This has led to the development of a class of indirect estimators that make use of information from related areas through modelling. A model is used to link similar areas to enhance the estimation of unplanned areas; in other words, they borrow strength from the other areas. Doing so improves the precision of estimated characteristics in the small area, especially in areas with smaller sample sizes. Social science researchers have increasingly employed small area estimation to provide localised estimates of population characteristics from surveys. We explore how to extend this approach within the context of deep and persistent disadvantage in Australia. We find that because of the unique circumstances of the Australian population distribution, direct estimates of disadvantage have substantial variation, but by applying small area estimation, there are significant improvements in precision of estimates.     

Health insurance loss: the case of the displaced worker.

Upon termination, most displaced workers lose employer-financed health insurance along with their jobs; data for 1983-84 suggest that such workers ran a high risk of remaining uninsured for extended periods, even after new employment was secured.     

The role of health insurance in labor supply decisions of divorced females

Labor economics literature provides evidence that marital dissolution induces an increase in labor supply of females. This paper explores an explanation for this finding: Marital separation might place wives at risk of losing health insurance or increase the need for expanded health coverage. Thus, wives must increase their labor supply in order to qualify for health benefits. Using data from the National Longitudinal Survey of Youth, results confirm that marital dissolution is associated with increased female labor supply. However, this effect is mostly concentrated among women who were not previously enrolled in their husbands’ health insurance plans. For wives who were dependent on their husbands for coverage, continuing coverage laws appear to mitigate the effect of marital dissolution on female labor supply.     

Respondent Related Correlates of Response Behaviour in Audience Research

Audience research by means of surveys has a long tradition, certainly withinarts and humanities oriented research. Yet, due to selective sampling and unitnonresponse it frequently lacks the methodological rigour to make scientificallyvalid statements based on sample estimates. This is one of the first attempts toexplore unit nonresponse in audience research. More specifically, it focuses onthe explanation of nonresponse by the socio-demographic and more topic relatedcharacteristics of a theatre audience. Using a two-step procedure for the on-sitecollection of data, the characteristics of respondents are compared with those ofnonrespondents. In step 1 the composition of the theatre audience is comparedto a proxy of a theatre population benchmark based on a weighted sample of theFlemish population (APS-2000). The validity of this best available method isdiscussed. Step 2 compares respondents with nonrespondents on a micro-level:ignoring unit nonresponse in step 1, we use logistic regression to map selectionin step 2. The chance of cooperating with the survey has been found to increasewith educational attainment and vary according to occupational category. Moreover,involvement with survey topic is confirmed as a strong predictor of survey participation.Gender, age and experience with theatre remain insignificant in predicting responsebehaviour. These findings are compared with the socio-demographic correlates ofresponse behaviour in general populations. Implications for statistically controllingfor nonresponse bias in audience research are discussed. Suggestions for furtherresearch are presented.     

Eastern European Public Opinion on Economic Issues:

Abstract Survey data on Eastern Europe is investigated with respect to two major questions: How do people view the creation of a market economy and how do they assess the speed of economic transformation and privatization. These questions are analyzed by looking at the development over 1990 to 1992 and by performing regressions with a number of demographic variables as regressors. Some findings are that the opposition to the creation of a market economy is increasing over time, that in Hungary and Poland people are quite satisfied with the speed of transformation as compared to Czechs and Slovaks, that the Polish are divided over the speed of privatization, and that the relationship between the speed of transformation and privatization is not considered to be very tight by the respondents.     

Small‐sample bias in synthetic cohort models of labor supply

This paper investigates small‐sample biases in synthetic cohort models (repeated cross‐sectional data grouped at the cohort and year level) in the context of a female labor supply model. I use the Current Population Survey to compare estimates when group sizes are extremely large to those that arise from randomly drawing subsamples of observations from the large groups. I augment this approach with Monte Carlo analysis so as to precisely quantify biases and coverage rates. In this particular application, thousands of observations per group are required before small‐sample issues can be ignored in estimation and sampling error leads to large downward biases in the estimated income elasticity. Copyright © 2007 John Wiley & Sons, Ltd.