Correction of Misclassification Error in Disability Rates

Gosling, Amanda; Saloniki, Eirini-Christina

doi:10.1002/hec.3080

Cited by 6 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…That misclassification is a significant concern when assessing both disability status and labor market outcomes is not new. In terms of disability, Kreider and Pepper (2007, p. 432) note that “there is widespread concern about the accuracy of self‐reported disability status in survey datasets.” Gosling and Saloniki (2014, p. 1085) find a “strong estimated bias” in the employment gap between the non‐disabled and the disabled using a self‐reported measure of disability status similar to ours. This misclassification reflects many underlying causes, the most obvious which is a lack of an agreed‐upon definition of disability and a corresponding appropriately worded survey question, leading to subjectiveness in the understanding of concepts such as disability or work limitation (Baker et al., 2004; Hale, 2001).…”

Section: Introductionsupporting

confidence: 80%

“…14 Supplementary Figure A12 reveals that the unemployment gap can no longer be signed if Q exceeds roughly 0.05 in all years even under our most stringent assumptions. 15 It is difficult to report a precise comparison from Gosling and Saloniki (2014) since they report separate estimates of the gap by gender, schooling, and age. That said, the estimated gaps are below roughly 0.60 only for young females with no education qualifications.…”

Section: Conflict Of Interestmentioning

confidence: 99%

“…In terms of disability, Kreider and Pepper (2007, p. 432) note that "there is widespread concern about the accuracy of self-reported disability status in survey datasets." Gosling andSaloniki (2014, p. 1085) find a "strong estimated bias" in the employment gap between the non-disabled and the disabled using a self-reported measure of disability status several years after the passage of the ADAAA. Both the Great Recession and the ADAAA may have altered the relationship between disability status and labor market outcomes.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bounding the joint distribution of disability and employment with misclassification

Liu

Millimet

2021

Health Economics

View full text Add to dashboard Cite

Understanding the relationship between disability and employment is critical and has long been the subject of study. However, estimating this relationship is difficult, particularly with survey data, since both disability and employment status are known to be misreported. Here, we use a partial identification approach to bound the joint distribution of disability and employment status in the presence of misclassification. Allowing for a modest amount of misclassification leads to bounds on the labor market status of the disabled that are not overly informative given the relative size of the disabled population. Thus, absent further assumptions, even a modest amount of misclassification creates much uncertainty about the employment gap between the non‐disabled and disabled. However, additional assumptions considered are shown to have some identifying power. For example, under our most stringent assumptions, we find that the employment gap is at least 15.2% before the Great Recession and 22.0% afterward.

show abstract

Section: Introductionsupporting

confidence: 80%

Section: Conflict Of Interestmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bounding the joint distribution of disability and employment with misclassification

Liu

Millimet

2021

Health Economics

View full text Add to dashboard Cite

show abstract

“…More recently, Benítez-Silva et al (2004, p.649) are "unable to reject the hypothesis that self-reported disability is an unbiased indicator", while in contrast, Baker et al (2004Baker et al ( , p.1090 find "evidence that the error in self-reported chronic conditions is related to labor market status", and the results in Kerkhofs (2009, p.1042) "show that justification bias is substantial and that failing to account for this may change estimation results considerably". Further recent evidence on the importance of justification bias can be found in Gannon (2009), Datta Gupta and Larsen (2010), Datta Gupta and Jürges (2012), and Gosling and Saloniki (2014).…”

Section: Introductionmentioning

confidence: 90%

Justification bias in self-reported disability: New evidence from panel data

Black

Johnston

Suziedelyte

2017

Journal of Health Economics

View full text Add to dashboard Cite

The relationship between health and work is frequently investigated using self-assessments of disability from social surveys. The complication is that respondents may overstate their level of disability to justify non-employment and welfare receipt. This study provides new evidence on the existence and magnitude of justification bias by exploiting a novel feature of a large longitudinal survey: each wave respondents are asked identical disability questions twice; near the beginning and end of the face-to-face interview. Prior to answering the second disability question, respondents are asked a series of questions that increase the salience of their employment and welfare circumstances. Justification bias is identified by comparing the variation between the two measures within-individuals over time, with the variation in employment status over time. Results indicate substantial and statistically significant justification bias; especially for men and women who receive disability pensions.

show abstract

“…By replicating prior studies, the authors showed that key estimated parameters varied by up to 100% due to data errors. Other papers also indicated errors in aggregate statistical data for suicide [14], disability [15], mortality [16], and life satisfaction [17]. Thus, despite the fact that many studies have examined methods of correcting classification bias [9], [11], we can conclude based on previously mentioned cases that it is essential to analyze this bias together with a mathematical formula 1 for calculating the indicator for social indicators research.…”

Section: Introductionmentioning

confidence: 99%

Misclassification Bias in Computational Social Science: A Simulation Approach for Assessing the Impact of Classification Errors on Social Indicators Research

Smetanin

Komarov

2022

IEEE Access

View full text Add to dashboard Cite

A growing body of literature has examined the potential of machine learning algorithms in constructing social indicators based on the automatic classification of digital traces. However, as long as the classification algorithms' predictions are not completely error-free, the estimate of the relative occurrence of a particular class may be affected by misclassification bias, thereby affecting the value of the calculated social indicator. Although a significant amount of studies have investigated misclassification bias correction techniques, they commonly rely on a set of assumptions that are likely to be violated in practice, which calls into question the effectiveness of these methods. Thus, there is a knowledge gap with respect to the assessment of misclassification bias's impact on a specific social indicator formula without strict reference to the number of classes. Moreover, given the erroneous nature of automatic classification algorithms, the quality of a predicted indicator can be assessed not only using regression quality metrics, as was done in existing literature, but also using correlation metrics. In this paper, we propose a simulation approach for assessing the impact of misclassification bias on the calculated social indicators in terms of regression and correlation metrics. The proposed approach focuses on indicators calculated based on the distribution of classes and can process any number of classes. The proposed approach allows selecting the most appropriate classification model for a particular social indicator, and vice versa. Moreover, it allows for assessment of the optimistic level of correlation between the indicator calculated based on the results of the classification algorithm and the true underlying indicator.

show abstract

Correction of Misclassification Error in Disability Rates

Cited by 6 publications

References 17 publications

Bounding the joint distribution of disability and employment with misclassification

Bounding the joint distribution of disability and employment with misclassification

Justification bias in self-reported disability: New evidence from panel data

Misclassification Bias in Computational Social Science: A Simulation Approach for Assessing the Impact of Classification Errors on Social Indicators Research

Contact Info

Product

Resources

About