@article {5679, title = {Enhancing the Quality of Data on the Measurement of Income and Wealth}, number = {WP 2007-151}, year = {2007}, institution = {Michigan Retirement and Disability Research Center, University of Michigan}, address = {Ann Arbor, MI}, abstract = {Over the last decade or so, a substantial effort has gone into the design of a series of methodological investigations aimed at enhancing the quality of survey data on income and wealth. These investigations have largely been conducted at the Survey Research Center at the University of Michigan, and have mainly involved two longitudinal surveys: the Health and Retirement Study (HRS), with a first wave beginning in 1992 and continued thereafter every other year through 2004; and the Assets and Health Dynamics Among the Oldest Old (AHEAD) Study, begun in 1993 and continued in 1995 and 1998, then in every other year through 2004. Surveys for the year 2006 are currently in the field. This paper provides an overview of the main studies and summarizes what has been learned so far. The studies include; a paper by Juster and Smith (Improving the Quality of Economic Data: Lessons from the HRS and AHEAD, JASA, 1997); a paper by Juster, Cao, Perry and Couper (The Effect of Unfolding Brackets on the Quality of Wealth Data in HRS, MRRC Working Paper, WP 2006-113, January 2006); a paper by Hurd, Juster and Smith (Enhancing the Quality of Data on Income: Recent Innovations from the HRS, Journal of Human Resources, Summer 2003); a paper by Juster, Lupton and Cao (Ensuring Time-Series Consistency in Estimates of Income and Wealth, MRRC Working Paper, WP 2002-030, July 2002); a paper by Cao and Juster (Correcting Second-Home Equity in HRS/AHEAD: MRRC Working Paper WP 2004-081, June 2004); and a paper by Rohwedder, Haider and Hurd (RAND Working Paper, 2004).}, keywords = {Income, Methodology, Net Worth and Assets}, url = {https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1095815}, author = {Juster, F. Thomas and Cao, Honggao and Mick P. Couper and Daniel H. Hill and Michael D Hurd and Joseph P. Lupton and Michael M. Perry and James P Smith} } @article {5635, title = {Enhancing the Quality of Data on Income and Wealth}, year = {2005}, institution = {The University of Michigan, Michigan Retirement Research Center}, abstract = {Over the last decade or so, a substantial effort has gone into the design of a series of methodological investigations aimed at enhancing the quality of survey data on income and wealth. These investigations have largely been conducted at the Survey Research Center at the University of Michigan, and have mainly involved two longitudinal surveys: the Health and Retirement Study (HRS), with a first wave beginning in 1992 and continued thereafter every other year through 2004; and the Assets and Health Dynamics Among the Oldest Old (AHEAD) Study, begun in 1993 and continued in 1995 and 1998, then in every other year through 2004. This provides and overview of the main studies and summarizes what has been learned about correcting longitudinal inconsistencies that arise.}, keywords = {Income, Methodology, Net Worth and Assets}, url = {http://www.mrrc.isr.umich.edu/publications/papers/}, author = {Cao, Honggao and Daniel H. Hill and Juster, F. Thomas and Michael M. Perry} } @article {5636, title = {Knowledge and Preference in Reporting Financial Information}, year = {2005}, institution = {Michigan Retirement Research Center, University of Michigan}, address = {Ann Arbor, MI}, abstract = {This article models respondent behavior in a financial survey with a framework explicitly integrating a respondent s knowledge of and willingness to reveal his or her financial status. Whether a respondent provides a valid answer, a don t know , or a refusal to a financial question depends on the interaction of his or her financial knowledge and preferences regarding revealing the knowledge. Using asset response and nonresponse data from the Health and Retirement Study (2000), we found that knowledge and preferences play interrelated roles in reporting financial information, that a respondent s age, gender, education, and race and ethnicity are important predictors of respondent behavior, and that race and ethnicity affect a respondent behavior only via their influence on preferences, while gender only via its influence on knowledge. We also found strong heterogeneity in respondents financial knowledge and their willingness to reveal the knowledge.}, keywords = {Methodology}, url = {https://mrdrc.isr.umich.edu/pubs/knowledge-and-preference-in-reporting-financial-information/}, author = {Cao, Honggao and Daniel H. Hill} } @article {6730, title = {Reducing Panel Attrition: A Search for Effective Policy Instruments}, journal = {Journal of Human Resources}, volume = {36}, year = {2001}, note = {ProCite field 3 : Survey Research Center, Institute for Social Research, U MI; Survey Research Center, Institute for Social Research, U MI}, pages = {416-38}, publisher = {36}, abstract = {In this paper we develop a theory of the survey response decision process and apply it to the analysis of field office policy measures in an attempt to see which of these are effective in reducing panel attrition. We use data from the Health and Retirement Study (HRS) to assess the effectiveness of 1) reducing the length of the interview and 2) assigning the same initial interviewer wave after wave. There is virtually no evidence in the data that interview length affects subsequent wave response. Assigning the same interviewer wave after wave, however, has a strong positive effect on response rates.}, keywords = {Methodology}, doi = {10.2307/3069625}, author = {Daniel H. Hill and Robert J. Willis} } @article {5395, title = {Panel Attrition Bias in Health-Event Models}, year = {1999}, institution = {University of Michigan Institute for Social Resear}, keywords = {Methodology}, author = {Daniel H. Hill} } @inbook {5119, title = {Unfolding Bracket Method in the Measurement of Expenditures and Wealth}, booktitle = {Wealth, Work, and Health: Innovations in Measurement in the Social Sciences: Essays in honor of F. Thomas Juster.}, year = {1999}, note = {ProCite field[8]: eds.}, pages = {64-86}, publisher = {University of Michigan Press}, organization = {University of Michigan Press}, address = {Ann Arbor, MI}, keywords = {Consumption and Savings, Methodology, Net Worth and Assets}, url = {https://books.google.com/books?id=lKvp4D1HuH8C\&pg=PA64\&lpg=PA64\&dq=Unfolding+Bracket+Method+in+the+Measurement+of+Expenditures+and+Wealth\&source=bl\&ots=hFIAdSfSt6\&sig=ACfU3U2f2HvZoq6nQLadPt_pPaidMEBcyQ\&hl=en\&sa=X\&ved=2ahUKEwjAruG52p3qAhWIZM0KHTMiBpAQ6AEwA}, author = {Daniel H. Hill}, editor = {James P Smith and Robert J. Willis} } @article {5370, title = {The Empirical Validity of the HRS Medical Expenditure Data: A Model to Account for Different Reference Periods}, year = {1998}, institution = {University of Michigan Institute for Social Research}, keywords = {Medicare/Medicaid/Health Insurance, Methodology}, author = {Daniel H. Hill and Mathiowetz, Nancy A.} } @article {6591, title = {Question Difficulty and Respondents{\textquoteright} Cognitive Ability: The Impact on Data Quality}, journal = {Journal of Official Statistics}, volume = {13}, year = {1997}, pages = {181-199}, publisher = {13}, abstract = {Older respondents lower in cognitive ability are more affected by question difficulty than older respondents higher in cognitive ability. Respondents lower in cognitive ability gave more DK responses to difficult than to easy questions, while the differences between the difficult and easy questions were not as apparent for respondents with higher cognitive abilities. This effect may result in biased survey findings because responses to difficult questions from respondents with lower cognitive ability would be underrepresented.}, keywords = {Health Conditions and Status, Medicare/Medicaid/Health Insurance, Methodology}, url = {https://www.psc.isr.umich.edu/pubs/abs/1046}, author = {Barbel Knauper and Robert F. Belli and Daniel H. Hill and A. Regula Herzog} } @article {10985, title = {Unfolding brackets for reducing item nonresponse in economic surveys}, year = {1995}, institution = {Survey Research Center, Institute for Social Research, University of Michigan}, address = {Ann Arbor, MI}, abstract = {This paper describes and analyzes a new survey methodology for reducing item non-response on financial measures. This "unfolding bracket" method is systematic and applicable in both face-to-face and telephone surveys. The proportion of missing observations for financial variables in national surveys is often in the 20-25\% range and in some cases is as high as a third, With the unfolding bracket method the proportion of completely missing data can be cut by two-thirds. Furthermore, with appropriately chosen bracket breakpoints, the amount of the variance in the underlying measure recovered is quite high. We propose and demonstrate on method for choosing the breakpoints which employs the Downhill Simplex algorithm to maximize their explanatory value. Additionally, use of a Box-Cox transform of the actual data in conjunction with this algorithm, can result in breakpoints which are effective in explaining most of the underlying variance in both actual values and their log transforms. Since each of these metrics is appropriate for some uses this compromise is quite useful in meeting the needs of a wide variety of potential users. Finally, we investigate the effect of bracketing on the empirical validity of survey data. While we do find lower empirical validity for data from individuals exposed to brackets early in the survey instrument, this appears to be the result of self-selection rather than a direct effect of exposure to the methodology. }, keywords = {economic survey, item nonreponse, Missing data}, author = {Steven G Heeringa and Daniel H. Hill and Howell, David A.} } @article {5313, title = {Unfolding Brackets for Reducing Item Non-Response in Economic Surveys}, year = {1995}, institution = {University of Michigan}, abstract = {This paper describes and analyzes a new survey methodology for reducing item non-response on financial measures. This unfolding bracket method is systematic and applicable in both face-to-face and telephone surveys. The proportion of missing observations for financial variables in national surveys is often in the 20-25 range and in some cases is as high as a third. With the unfolding bracket method the proportion of completely missing data can be cut by two-thirds. Furthermore, with appropriately chosen bracket breakpoints, the amount of the variance in the underlying measure recovered is quite high. We propose and demonstrate one method for choosing the breakpoints which employs the Downhill Simplex algorithm to maximize their exploratory value. Additionally, use of a Box-Cox transform of the actual data in conjunction with this algorithm, can result in breakpoints which are effective in explaining most of the underlying variance in both actual values and their log transforms. Since each of these metrics is appropriate for some uses this compromise is quite useful in meeting the needs of a wide variety of potential users. Finally, we investigate the effects of bracketing on the empirical validity of survey data. While we do find lower empirical validity for data from individuals exposed to brackets early in the survey instrument, this appears to be the result of self-selection rather than a direct effect of exposure to the methodology.}, keywords = {Methodology}, url = {http://www.psc.isr.umich.edu/pubs/series.html}, author = {Steven G Heeringa and Daniel H. Hill and Howell, David A.} } @article {10497, title = {Unfolding Brackets for Reducing Item Nonresponse in Economic Surveys}, year = {1995}, abstract = {This paper describes and analyzes a new survey methodology for reducing item non-response on financial measures. This unfolding bracket method is systematic and applicable in both face-to-face and telephone surveys. The proportion of missing observations for financial variables in national surveys is often in the 20-25 range and in some cases is as high as a third. With the unfolding bracket method the proportion of completely missing data can be cut by two-thirds. Furthermore, with appropriately chosen bracket breakpoints, the amount of the variance in the underlying measure recovered is quite high. We propose and demonstrate one method for choosing the breakpoints which employs the Downhill Simplex algorithm to maximize their exploratory value. Additionally, use of a Box-Cox transform of the actual data in conjunction with this algorithm, can result in breakpoints which are effective in explaining most of the underlying variance in both actual values and their log transforms. Since each of these metrics is appropriate for some uses this compromise is quite useful in meeting the needs of a wide variety of potential users. Finally, we investigate the effects of bracketing on the empirical validity of survey data. While we do find lower empirical validity for data from individuals exposed to brackets early in the survey instrument, this appears to be the result of self-selection rather than a direct effect of exposure to the methodology.}, keywords = {Methodology}, author = {Steven G Heeringa}, editor = {Daniel H. Hill} } @article {5297, title = {Respondent Rules and the Quality of Net Worth Data in the HRS}, year = {1993}, institution = {University of Michigan}, abstract = {The HRS uses an informed selection procedure in choosing with spouse to ask about the household{\textquoteright}s assets and debts. This is more expensive than presumptive selection which has been commonly employed in past surveys. This paper evaluates the quality of the resulting net worth data in terms of the exactness of the reports. A trivariate ordered probit model is developed and estimated. This model controls for the self-selectivity of both the gender and proxy interview status of the net-worth respondent. The results suggest that the informed selection procedure does indeed result in lower item non-response and less reliance on approximation strategies than would a presumptive husband R1 respondent rule.}, keywords = {Methodology, Net Worth and Assets}, url = {http://www.psc.isr.umich.edu/pubs/series.html}, author = {Daniel H. Hill} }