Sex Differences in Measurement Error in Status Attainment Models

Corcoran, Mary

doi:10.1177/004912418000900205

Cited by 19 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reliability, defined in this way, has been used in a number of applications aimed at understanding the extent of randomness in survey response (e.g. see Achen, 1975;Alwin, 1989aAlwin, , 1989bKrosnick, 1989a, 1989b;Alwin and Thornton, 1984;Andrews, 1984;Asher, 1974;Bieiby and Hauser, 1977;Featherman, 1977a, 1977b;Borus and Nestle, 1973;Converse and Markus, 1979;Corcoran, 1980;Erikson, 1979;Hauser, Tsai and Sewell, 1983;Jagodzinski and Kuhnei, 1987;Marquis and Marquis, 1977;Siegel and Hodge, 1968;Smith and Stephenson, 1979). The general conclusion of this body of work is that unreliability of measurement occurs with respect to a wide variety of content areas, and while the measurement of attitudes may be particularly difficult (Converse, 1964), substantial amounts of error occur in the measurement of other phenomena as well.…”

Section: The Concept Of Reliabilitymentioning

confidence: 99%

Problems in the estimation and interpretation of the reliability of survey data

Alwin

1989

Qual Quant

View full text Add to dashboard Cite

Abstraet. In this paper I discuss several of the difficulties involved in estimating the reliability of survey measurement. Reliability is defined on the basis of classical true-score theory, as the correlational consistency of multiple measures of the same construct, net of true change. This concept is presented within the framework of a theoretical discussion of the sources of error in survey data and the design requirements for separating response variation into components representing such response consistency and measurement errors. Discussion focuses on the potential sources of random and nonrandom errors, including "invalidity" of measurement, the term frequently used to refer to components of method variance. Problems with the estimation of these components are enumerated and discussed with respect to both cross-sectional and panel designs. Empirical examples are given of the estimation of the quantities of interest, which are the basis of a discussion of the interpretational difficulties encountered in reliability estimation. Data are drawn from the ISR's Quality of Life surveys, the National Election Studies and the NORC's General Social Surveys. The general conclusion is that both crosssectional and panel estimates of measurement reliability are desirable, but for the purposes of isolating the random component of error, panel designs are probably the most advantageous.

show abstract

Section: The Concept Of Reliabilitymentioning

confidence: 99%

Problems in the estimation and interpretation of the reliability of survey data

Alwin

1989

Qual Quant

View full text Add to dashboard Cite

show abstract

“…Further, note should be taken of the work which has shown that sub-groups in the population may exhibit varying patterns of response error (Mason et al, 1976;Bielby et al, 1977: Corcoran, 1980Wyner, 1981;Mare and Mason, 1980;Hamilton, 1981) and, of the research on the stability of underlying true scores across time (Wheaton et al, 1977). We would like some information on the Highers you have taken, N.B.…”

Section: Discussionmentioning

confidence: 99%

Towards a typology of information to aid reliability of response in social surveys

Cuttance

1986

Qual Quant

View full text Add to dashboard Cite

IntroductionIn this paper I use a covariance structure analysis to model response reliability and measurement error in survey data and to test the validity of a preliminary typology of variable types. The impetus to the construction of such a typology derives from the need to be able to identify the information in a data set which may have a less than acceptable degree of reliability. As a basis for distinguishing between which types of information may yield more or less reliable responses a preliminary typology based on the formal characteristics of the information is outlined. From this classification, models for the different classes of information are used to estimate the reliability of the observed variables. Differential reliability across variables within each information class is then investigated in terms of a subsidiary hypothesis that the degree of reliability varies systematically according to other dimensions of the formal characteristics not encompassed by this typology.

show abstract

“…The effects of social background on son's educational attainment and the effect of father's occupational status on son's occupational status are stronger when fathers' information was used. Corcoran (1980Corcoran ( , 1981 used data from interviews with both young adults and their parents. The offspring turned out to make answers on different parental characteristics more consistent than they really were.…”

Section: Previous Researchmentioning

confidence: 99%

“…The studies of Broom et al (1978), Corcoran (1980Corcoran ( , 1981, Hauser et al (1983), Massagli and Hauser (1983), and Van Eijck (1996) discussed above use this design. This is the design we prefer, since the multiple source type design is not feasible for most family background variables, and because the multiple moment design respondents can give the same wrong answers in each interview.…”

Section: Multiple Informant Designmentioning

confidence: 99%