Some Survey Sampling Biases

Lowe, Francis E.; McCormick, Thomas C.

doi:10.1086/266574

Cited by 14 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Studies are also divided as to whether nonrespondents vary by sex. Crossley and Fink (1951), Hawkins (1975), andDeMaio (1980) found no difference, but Bartholomew (1961), Lowe and McCormick (1955), and Smith (1979) found an underrepresentation of men. .7209 a A parallel analysis using nonparametric methods (Davis, 1975) showed similar results except for marital status.…”

Section: Difficultymentioning

confidence: 98%

See 1 more Smart Citation

The Hidden 25 Percent: An Analysis of Nonresponse on the 1980 General Social Survey

Smith

1983

Public Opinion Quarterly

109

View full text Add to dashboard Cite

NONRESPONSE can seriously bias survey estimates and distort inferences. The relationship between nonresponse and survey estimates is simple and well defined, but the actual impact is ill known. For the sample mean of a particular variable {Y) the association iswhere W t and W 2 are the proportion respondents and nonrespondents. The relative bias (RB) of using the response mean to equal the sample means is = W 2We can see that the relative bias is serious only when the nonresponse rate (W^ is large and the difference in the means is great. Given this simple formula we can easily measure the magnitude of the nonresponse bias. The problem is that while we know the nonresAbstract Methods for estimating nonresponse bias are reviewed and several methods are tried on the 1980 GSS. The results indicate that various estimating procedures are inappropriate and that even the more promising techniques can provide faulty estimates of nonresponse bias. By its nature, nonresponse bias is very difficult to assess accurately and no simple, certain method exists.

show abstract

Section: Difficultymentioning

confidence: 98%

“…Most previous research finds final nonresponse to be highest among the older ages(Lowe and McCormick, 1955; Lansing et al, 1971;Weaver et al, 1975;Hawkins, 1975; at New York University on May 28, 2015 http://poq.oxfordjournals.org/ Downloaded from…”

mentioning

confidence: 99%

The Hidden 25 Percent: An Analysis of Nonresponse on the 1980 General Social Survey

Smith

1983

Public Opinion Quarterly

109

View full text Add to dashboard Cite

show abstract

“…Older individuals are less likely to participate in personal interviews (Lowe and McCormick, 1955) and surveys (Goyder, 1986;Kandel and Raveis, 1983;Moore and Tarnai, 2002). Using a sample of the elderly, Kaldenberg and his colleagues (Kaldenberg, Koenig, and Becker, 1994) found that non-response increased over one half of a percentage point for each year increase of respondents' age.…”

Section: The General Populationmentioning

confidence: 99%

Non-response in student surveys: The Role of Demographics, Engagement and Personality

2005

View full text Add to dashboard Cite

“…In general, certain groups, such as males (Curtin, Presser & Singer, 2000;Groves & Couper, 1996;Moore & Tarnai, 2002;Singer, van Hoewyk & Mather, 2000), and individuals with low incomes (Curtin et al, 2000;Goyder, 1986;Goyder, Warriner & Miller, 2002;Groves, Singer & Corning, 2000;Kandel & Raveis, 1983;Singer, Groves & Corning, 1999), are less likely to respond to surveys than females and individuals with high incomes. Research has also shown that the response rate for the old is lower than for the young (Goyder, 1986;Kaldenberg, Koenig & Becker, 1994;Kandel & Raveis, 1983;Lowe & McCormack, 1955;Moore & Tarnai, 2002) and that in the United States Blacks are less likely to complete surveys than Whites (Curtin et al, 2000;Groves et al, 2000;Singer et al, 1999;Singer et al, 2000;Voigt, Koepsell, & Daling, 2003). In general, independent of the demographic group under consideration, potential survey participants are more likely to respond to surveys if they have an interest in the topic under study (Filion, 1975;Kojetin, Borgida & Snyder, 1993;Pearl & Fairley, 1985;Stinchcombe, Jones & Sheatsley, 1981;Van Kenhove, Wijnen & De Wulf, 2002).…”

Section: Response Ratesmentioning

confidence: 99%

How research ethics boards are undermining survey research on Canadian university students

Grayson

Myles

2005

J Acad Ethics

View full text Add to dashboard Cite

In Canada, all research conducted by individuals associated with universities must be subjected to review by research ethics boards (REB). Unfortunately, decisions reached by REBs may seriously compromise the integrity of university-based research. In this paper attention will focus on how requirements of REBs and a legal department in four Canadian universities affected response rates to a survey of domestic and international students. It will be shown that in universities in which students were sent a legalistic cover letter to a mail survey, or were required to sign a consent form, lower response rates were achieved than in universities in which students were sent a relatively friendly letter. In turn, lower response rates resulted in: sample characteristics that deviated from population characteristics; a reduction in the possibility of testing research hypotheses; and increased survey costs. As a consequence, it is argued that the unreasonable demands of REBs are seriously compromising the quality of research that can be carried out on Canadian university students.

show abstract

Some Survey Sampling Biases

Cited by 14 publications

References 0 publications

The Hidden 25 Percent: An Analysis of Nonresponse on the 1980 General Social Survey

The Hidden 25 Percent: An Analysis of Nonresponse on the 1980 General Social Survey

Non-response in student surveys: The Role of Demographics, Engagement and Personality

How research ethics boards are undermining survey research on Canadian university students

Contact Info

Product

Resources

About