Nonresponse bias can lead to the misinterpretation of published data. The National Center for Edu... more Nonresponse bias can lead to the misinterpretation of published data. The National Center for Education Statistics requested that the Census Bureau conduct a Nonresponse Bias Study for the 2003-2004 School and Staffing Survey. Our discussion focuses on portions of the methodology applied and results, as well as future endeavors for nonresponse bias studies.
The purpose of this research was to examine the consistency of responses for similar questions as... more The purpose of this research was to examine the consistency of responses for similar questions asked in the American Community Survey (ACS) and the National Survey of College Graduates (NSCG) to better leverage the use of the ACS in the sample design and estimation of the NSCG. This research evaluated the ACS responses used in the NSCG sample design to determine if the current NSCG design was efficient for key statistics, and to identify new ACS responses that may strengthen the design. Additionally, this evaluation identified variables that could be used in model assisted and model-based estimation for the NSCG. The research consisted of two phases: response comparison and examination of response differences. We looked at individual responses to find patterns that lead to the inconsistency of occupation and education. This included constructing statistical models involving questionnaire responses and paradata to find covariates related to inconsistency. The use of the American Comm...
Nonresponse bias can lead to the misinterpretation of published data. The National Center for Edu... more Nonresponse bias can lead to the misinterpretation of published data. The National Center for Education Statistics requested that the Census Bureau conduct a Nonresponse Bias Study for the 2003-2004 School and Staffing Survey. Our discussion focuses on portions of the methodology applied and results, as well as future endeavors for nonresponse bias studies.
The purpose of this research was to examine the consistency of responses for similar questions as... more The purpose of this research was to examine the consistency of responses for similar questions asked in the American Community Survey (ACS) and the National Survey of College Graduates (NSCG) to better leverage the use of the ACS in the sample design and estimation of the NSCG. This research evaluated the ACS responses used in the NSCG sample design to determine if the current NSCG design was efficient for key statistics, and to identify new ACS responses that may strengthen the design. Additionally, this evaluation identified variables that could be used in model assisted and model-based estimation for the NSCG. The research consisted of two phases: response comparison and examination of response differences. We looked at individual responses to find patterns that lead to the inconsistency of occupation and education. This included constructing statistical models involving questionnaire responses and paradata to find covariates related to inconsistency. The use of the American Comm...
dim [Lα,L−α]≧1 for every root α of a symrnetrizable Kac- Moody Lie algebra L =L(A). Let α be the ... more dim [Lα,L−α]≧1 for every root α of a symrnetrizable Kac- Moody Lie algebra L =L(A). Let α be the sum of the fundamental roots of an arbitrary Kac-Moody algebra L. If there is some nonrepeating sequence il,…,ikwith then dim [Lα,L−α]≧ k. In May 1993 A. John Coleman asked me what would happen if there were two such sequences. This note contains a partial solution. In special cases, if there are k such (disjoint) sequences, of lengths slthrough sk, then dim .
The U.S. Census Bureau relies on reinterview programs as the primary method to evaluate field wor... more The U.S. Census Bureau relies on reinterview programs as the primary method to evaluate field work and monitor the work of the interviewers. One purpose of the reinterviews is to identify falsification. Since falsification is a rare occurrence, reinterview programs generally identify very few falsified cases even when the reinterview sample is reasonably large. This study examines methods for designing a reinterview sample with the goal of identifying more falsified cases. With the Current Population Survey (CPS) as an example, we explore data that could be used for reinterview sampling beyond that currently used in the CPS program. We fit a logistic regression model to predict the likelihood of falsification with the data from original interviews, and use the predicted probabilities to construct alternative reinterview sampling designs. The alternative designs are compared to the current sampling method through cross validation and simulation methods.
... a=klal+k2a2) . Let a~G-{+a~}, arO, Ga#O. If ht a=l, then a=al and [Gal Ca]=khl. Recall that C... more ... a=klal+k2a2) . Let a~G-{+a~}, arO, Ga#O. If ht a=l, then a=al and [Gal Ca]=khl. Recall that Cna #(O} if and only i if n=0,+1. If ht a=2, then a=al+a2 and [Gar Za]= % [ [elfez] [fll f21 I=% ( ~ ~ h ~ + ~ ~ ~ h ~ ) = kh ~ (since A12=Or ~~~=-nfO) . ...
Nonresponse bias can lead to the misinterpretation of published data. The National Center for Edu... more Nonresponse bias can lead to the misinterpretation of published data. The National Center for Education Statistics requested that the Census Bureau conduct a Nonresponse Bias Study for the 2003-2004 School and Staffing Survey. Our discussion focuses on portions of the methodology applied and results, as well as future endeavors for nonresponse bias studies.
The purpose of this research was to examine the consistency of responses for similar questions as... more The purpose of this research was to examine the consistency of responses for similar questions asked in the American Community Survey (ACS) and the National Survey of College Graduates (NSCG) to better leverage the use of the ACS in the sample design and estimation of the NSCG. This research evaluated the ACS responses used in the NSCG sample design to determine if the current NSCG design was efficient for key statistics, and to identify new ACS responses that may strengthen the design. Additionally, this evaluation identified variables that could be used in model assisted and model-based estimation for the NSCG. The research consisted of two phases: response comparison and examination of response differences. We looked at individual responses to find patterns that lead to the inconsistency of occupation and education. This included constructing statistical models involving questionnaire responses and paradata to find covariates related to inconsistency. The use of the American Comm...
Nonresponse bias can lead to the misinterpretation of published data. The National Center for Edu... more Nonresponse bias can lead to the misinterpretation of published data. The National Center for Education Statistics requested that the Census Bureau conduct a Nonresponse Bias Study for the 2003-2004 School and Staffing Survey. Our discussion focuses on portions of the methodology applied and results, as well as future endeavors for nonresponse bias studies.
The purpose of this research was to examine the consistency of responses for similar questions as... more The purpose of this research was to examine the consistency of responses for similar questions asked in the American Community Survey (ACS) and the National Survey of College Graduates (NSCG) to better leverage the use of the ACS in the sample design and estimation of the NSCG. This research evaluated the ACS responses used in the NSCG sample design to determine if the current NSCG design was efficient for key statistics, and to identify new ACS responses that may strengthen the design. Additionally, this evaluation identified variables that could be used in model assisted and model-based estimation for the NSCG. The research consisted of two phases: response comparison and examination of response differences. We looked at individual responses to find patterns that lead to the inconsistency of occupation and education. This included constructing statistical models involving questionnaire responses and paradata to find covariates related to inconsistency. The use of the American Comm...
dim [Lα,L−α]≧1 for every root α of a symrnetrizable Kac- Moody Lie algebra L =L(A). Let α be the ... more dim [Lα,L−α]≧1 for every root α of a symrnetrizable Kac- Moody Lie algebra L =L(A). Let α be the sum of the fundamental roots of an arbitrary Kac-Moody algebra L. If there is some nonrepeating sequence il,…,ikwith then dim [Lα,L−α]≧ k. In May 1993 A. John Coleman asked me what would happen if there were two such sequences. This note contains a partial solution. In special cases, if there are k such (disjoint) sequences, of lengths slthrough sk, then dim .
The U.S. Census Bureau relies on reinterview programs as the primary method to evaluate field wor... more The U.S. Census Bureau relies on reinterview programs as the primary method to evaluate field work and monitor the work of the interviewers. One purpose of the reinterviews is to identify falsification. Since falsification is a rare occurrence, reinterview programs generally identify very few falsified cases even when the reinterview sample is reasonably large. This study examines methods for designing a reinterview sample with the goal of identifying more falsified cases. With the Current Population Survey (CPS) as an example, we explore data that could be used for reinterview sampling beyond that currently used in the CPS program. We fit a logistic regression model to predict the likelihood of falsification with the data from original interviews, and use the predicted probabilities to construct alternative reinterview sampling designs. The alternative designs are compared to the current sampling method through cross validation and simulation methods.
... a=klal+k2a2) . Let a~G-{+a~}, arO, Ga#O. If ht a=l, then a=al and [Gal Ca]=khl. Recall that C... more ... a=klal+k2a2) . Let a~G-{+a~}, arO, Ga#O. If ht a=l, then a=al and [Gal Ca]=khl. Recall that Cna #(O} if and only i if n=0,+1. If ht a=2, then a=al+a2 and [Gar Za]= % [ [elfez] [fll f21 I=% ( ~ ~ h ~ + ~ ~ ~ h ~ ) = kh ~ (since A12=Or ~~~=-nfO) . ...
Uploads
Papers by Phyllis Singer