Determinants of social desirability bias in sensitive surveys: a literature review
TL;DR: This article reviewed theoretical explanations of socially motivated misreporting in sensitive surveys and provided an overview of the empirical evidence on the effectiveness of specific survey methods designed to encourage the respondents to answer more honestly.
Abstract: 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.
Citations
More filters
TL;DR: Adolescents who used social media more - both overall and at night - and those who were more emotionally invested in social media experienced poorer sleep quality, lower self-esteem and higher levels of anxiety and depression.
715 citations
Cites background from "Determinants of social desirability..."
...This may have resulted in less accurate data from certain participants due to poor understanding of questions or increased social desirability bias, especially on sensitive measures concerning mood or selfesteem, when language support was given by a researcher or teacher (Krumpal, 2013)....
[...]
TL;DR: An empirical account of how one research team developed and employed strategies to detect and limit social desirability bias in rural Ethiopia is provided.
Abstract: Many qualitative research studies acknowledge the possibility of social desirability bias (a tendency to present reality to align with what is perceived to be socially acceptable) as a limitation t...
356 citations
Cites background from "Determinants of social desirability..."
...…answering a question in a particular way has links to rational choice theory—that is, to maximize utility and goal achievement—and the subjective expected utility theory—considering the risks, losses, and outcomes associated with the decision to admit to a sensitive behavior or not (Krumpal, 2013)....
[...]
...Based on the use of standardized surveys, social desirability bias is considered to stem from the sensitive nature of an answer, and is distinct from sensitivities associated with the intrusiveness of a question or the respondent’s perceived risk of disclosure (Krumpal, 2013)....
[...]
...The rationale for answering a question in a particular way has links to rational choice theory—that is, to maximize utility and goal achievement—and the subjective expected utility theory—considering the risks, losses, and outcomes associated with the decision to admit to a sensitive behavior or not (Krumpal, 2013)....
[...]
TL;DR: Personal attitudes towards mental illness or help-seeking are associated with active help-seek for mental problems, and campaigns promoting help- seeking and fighting mental illness-related stigma should target these personal attitudes rather than broad public opinion.
Abstract: BackgroundMental disorders create high individual and societal costs and burden, partly because help-seeking is often delayed or completely avoided. Stigma related to mental disorders or mental health services is regarded as a main reason for insufficient help-seeking.AimsTo estimate the impact of four stigma types (help-seeking attitudes and personal, self and perceived public stigma) on active help-seeking in the general population.MethodA systematic review of three electronic databases was followed by random effect meta-analyses according to the stigma types.ResultsTwenty-seven studies fulfilled eligibility criteria. Participants' own negative attitudes towards mental health help-seeking (OR = 0.80, 95% CI 0.73-0.88) and their stigmatising attitudes towards people with a mental illness (OR = 0.82, 95% CI 0.69-0.98) were associated with less active help-seeking. Self-stigma showed insignificant association (OR = 0.88, 95% CI 0.76-1.03), whereas perceived public stigma was not associated.ConclusionsPersonal attitudes towards mental illness or help-seeking are associated with active help-seeking for mental problems. Campaigns promoting help-seeking by means of fighting mental illness-related stigma should target these personal attitudes rather than broad public opinions.
332 citations
TL;DR: The estimated proportion of PSP reporting current symptom clusters consistent with 1 or more mental disorders appears higher than previously published estimates for the general population; however, direct comparisons are impossible because of methodological differences.
Abstract: Background:Canadian public safety personnel (PSP; e.g., correctional workers, dispatchers, firefighters, paramedics, police officers) are exposed to potentially traumatic events as a function of their work. Such exposures contribute to the risk of developing clinically significant symptoms related to mental disorders. The current study was designed to provide estimates of mental disorder symptom frequencies and severities for Canadian PSP.Methods:An online survey was made available in English or French from September 2016 to January 2017. The survey assessed current symptoms, and participation was solicited from national PSP agencies and advocacy groups. Estimates were derived using well-validated screening measures.Results:There were 5813 participants (32.5% women) who were grouped into 6 categories (i.e., call center operators/dispatchers, correctional workers, firefighters, municipal/provincial police, paramedics, Royal Canadian Mounted Police). Substantial proportions of participants reported current ...
287 citations
TL;DR: The authors conclude that the UCT is a promising alternative to RRT in self-administered surveys and that future research should be directed toward evaluating and improving the technique.
Abstract: Gaining valid answers to so-called sensitive questions is an age-old problem in survey research. Various techniques have been developed to guarantee anonymity and minimize the respondent's feelings of jeopardy. Two such techniques are the randomized response technique (RRT) and the unmatched count technique (UCT). In this study we evaluate the effectiveness of different implementations of the RRT (using a forced-response design) in a computer-assisted setting and also compare the use of the RRT to that of the UCT. The techniques are evaluated according to various quality criteria, such as the prevalence estimates they provide, the ease of their use, and respondent trust in the techniques. Our results indicate that the RRTs are problematic with respect to several domains, such as the limited trust they inspire and non-response, and that the RRT estimates are unreliable due to a strong false "no" bias, especially for the more sensitive questions. The UCT, however, performed well compared to the RRTs on all the evaluated measures. The UCT estimates also had more face validity than the RRT estimates. We conclude that the UCT is a promising alternative to RRT in self-administered surveys and that future research should be directed towards evaluating and improving the technique.
259 citations
References
More filters
Book•
01 Jan 1983
TL;DR: In this article, the authors present a survey of the use of truncated distributions in the context of unions and wages, and some results on truncated distribution Bibliography Index and references therein.
Abstract: Preface 1. Introduction 2. Discrete regression models 3. Probabilistic-choice models 4. Discriminant analysis 5. Multivariate qualitative variables 6. Censored and truncated regression models 7. Simultaneous-equations models with truncated and censored variables 8. Two-stage estimation methods 9. Models with self-selectivity 10. Disequilibrium models 11. Some applications: unions and wages Appendix: Some results on truncated distributions Bibliography Index.
13,828 citations
TL;DR: It seems clear that the items in the Edwards Social Desirability Scale would, of necessity, have extreme social desirability scale positions or, in other words, be statistically deviant.
Abstract: It has long been recognized that personality test scores are influenced by non-test-relevant response determinants. Wiggins and Rumrill (1959) distinguish three approaches to this problem. Briefly, interest in the problem of response distortion has been concerned with attempts at statistical correction for "faking good" or "faking bad" (Meehl & Hathaway, 1946), the analysis of response sets (Cronbach, 1946,1950), and ratings of the social desirability of personality test items (Edwards, 19 5 7). A further distinction can be made, however, which results in a somewhat different division of approaches to the question of response distortion. Common to both the Meehl and Hathaway corrections for faking good and faking bad and Cronbach's notion of response sets is an interest in the test behavior of the subject(S). By social desirability, on the other hand, Edwards primarily means the "scale value for any personality statement such that the scale value indicates the position of the statement on the social desirability continuum . . ." (1957, p. 3). Social desirability, thus, has been used to refer to a characteristic of test items, i.e., their scale position on a social desirability scale. Whether the test behavior of 5s or the social desirability properties of items are the focus of interest, however, it now seems clear that underlying both these approaches is the concept of statistical deviance. In the construction of the MMPI K scale, for example, items were selected which differentiated between clinically normal persons producing abnormal te¥Tpfpfiles~snd^cTinically abnormal individuals with abnormal test profiles, and between clinically abnormal persons with normal test profiles and abnormal 5s whose test records were abnormal. Keyed responses to the K scale items tend to be statistically deviant in the parent populations. Similarly, the development of the Edwards Social Desirability Scale (SDS) illustrates this procedure. Items were drawn from various MMPI scales (F, L, K, and the Manifest Anxiety Scale [Taylor, 1953]) and submitted to judges who categorized them as either socially desirable or socially undesirable. Only items on which there was unanimous agreement among the 10 judges were included in the SDS. It seems clear that the items in Edwards SDS would, of necessity, have extreme social desirability scale positions or, in other words, be statistically deviant. Some unfortunate consequences follow from the strict use of the statistical deviance model in the development of-sOcialTtesirSbTBty scales. With items drawn from the MMPI, it is apparent that in addition to their scalability for social desirability the items may also be characterized by their content which,^n a general sense, has pathological implications. When a social desrrabtltty^scale constructed according to this procedure is then applied to a college student population, the meaning of high social desirability scores is not at all clear. When 5s given the Edwards SDS deny, for example, that their sleep is fitful and disturbed (Item 6) or that they worry quite a bit over possible misfortunes (Item 35), it cannot be determined whether these responses are attributable to social desirability or to a genuine absence of such symptoms. The probability of occurrence of the symptoms represented in MMPI items (and incorportated in the SDS)
8,478 citations
01 Jan 1991
3,388 citations
Book•
13 Mar 2000TL;DR: In this article, the role of memory in response to survey questions is discussed. And the impact of the application of cognitive models to survey measurement is discussed, as well as the effect of these models on survey reporting of sensitive topics.
Abstract: 1. Introduction 2. Respondents' understanding of survey questions 3. The role of memory in survey responding 4. Answering questions about date and durations 5. Attitude questions 6. Factual judgments and numerical estimates 7. Attitude judgments and context effects 8. Mapping and formatting 9. Survey reporting of sensitive topics 10. Mode of data collection 11. Impact of the application of cognitive models to survey measurement.
3,114 citations
TL;DR: A survey technique for improving the reliability of responses to sensitive interview questions is described, which permits the respondent to answer "yes" or "no" to a question without the interviewer knowing what information is being conveyed by the respondent.
Abstract: For various reasons individuals in a sample survey may prefer not to confide to the interviewer the correct answers to certain questions. In such cases the individuals may elect not to reply at all or to reply with incorrect answers. The resulting evasive answer bias is ordinarily difficult to assess. In this paper it is argued that such bias is potentially removable through allowing the interviewee to maintain privacy through the device of randomizing his response. A randomized response method for estimating a population proportion is presented as an example. Unbiased maximum likelihood estimates are obtained and their mean square errors are compared with the mean square errors of conventional estimates under various assumptions about the underlying population.
2,929 citations