1986
DOI: 10.1080/01621459.1986.10478282
|View full text |Cite
|
Sign up to set email alerts
|

Response Bias and Reliability in Sensitive Topic Surveys

Abstract: Estimates of survey response bias and reliability are presented for six topics: receipt of welfare, income, alcohol use, drug use, criminal history, and embarrassing medical conditions. The estimates are derived from published full-design criterion validity studies. The common assumption that these characteristics are underreported is, in part, based on partial validity studies; the bias in estimating response parameters using partial designs is demonstrated. Evidence from the full-design studies suggests that… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

1
28
0

Year Published

1988
1988
2023
2023

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 71 publications
(29 citation statements)
references
References 28 publications
1
28
0
Order By: Relevance
“….strategy group. This result confrrms results of other investigators (Burton & Blair, 1991;Marquis, Marquis, & Polich, 1986) that the normal fast pace of interviews often leads to less accu. .…”
Section: Discussionsupporting
confidence: 76%
See 2 more Smart Citations
“….strategy group. This result confrrms results of other investigators (Burton & Blair, 1991;Marquis, Marquis, & Polich, 1986) that the normal fast pace of interviews often leads to less accu. .…”
Section: Discussionsupporting
confidence: 76%
“…Although individual sub.. ject's reports were not accurate under the free-strategy condition at the longer retention interval, when we examine the net error for the group as a whole (rather than as the mean absolute value for the errors as reported above), subject's errors tend to cancel each other out; that is, some subjects gave reports that were too high, whereas others gave reports that were too low. Epidemiologic research .. ers who are interested in the relationship between the amount smoked and health status should have the same concerns with accuracy of individual reports as do psychologists and survey researchers who are interested in studying the relationships between self-reports and other variables (e.g., Marquis et al, 1986).…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…In addition to the Ashenfelter and Kreuger (1994) twins data, replicates have been given by Barron, Berger, and Black (1997a), Card (1996), Freeman (1984, Kaestner, Joyce, and Wehbeh (1996), Loewenstein and Spletzer (1998), Mellow and Sider (1983), Romeo (1997), and Romeo and Sun (1996). Alternatively, other studies have examined the accuracy of data by comparing questionnaire responses to administrative data (Bollinger 1998;Bollinger and David 1997;Bound, Brown, Duncan, andRodgers 1990,1994;Bound and Krueger 1991;Duncan and Hill 1985;Marquis, Marquis, and Polich 1986;Rodgers, Brown, and Duncan 1993). Taken together, these studies offer little support for the classic EVM; measurement error often exhibits a covariance with the true value.…”
Section: Health Care Policy and Research Grant 1-r01-hs08188-01 And Nmentioning
confidence: 99%
“…The attenuation will be proportional to the reliability of the income measure. Estimates of the reliability of selfreported annual income as a measure of true annual income range from .7 to .9; estimates of its reliability as a measure of permanent income are roughly .5 (Marquis, Marquis, & Polich, 1986;Mazumder, 2001). If we take a value of .8 as the reliability of income, then we can get a corrected estimate of income segregation in 2010 as H ' .089/.8 ' .11 (a reliability of .5 would imply H ' .089/.5 ' .18).…”
mentioning
confidence: 99%