Annals of the Institute of Statistical Mathematics
Ranked set sampling was introduced by McIntyre (1952, Australian Journal of Agricultural Research... more Ranked set sampling was introduced by McIntyre (1952, Australian Journal of Agricultural Research, 3, 385-390) as a cost-effective method of selecting data if observations are much more cheaply ranked than measured. He proposed its use for estimating the population mean when the distribution of the data was unknown. In this paper, we examine the advantage, if any, that this method of sampling has if the distribution is known, for a specific family of distributions. Specifically, we consider estimation of # and G for the family of random variables with cdf's of the form F(~-~a). We find that the ranked set sample does provide more information about both # and G than a random sample of the same number of observations. We examine both maximum likelihood and best linear unbiased estimation of # and G, as well as methods for modifying the ranked set sampling procedure to provide even better estimation.
Due to declining response rates and coverage loss in random digit dial surveys, the NOAA Fisherie... more Due to declining response rates and coverage loss in random digit dial surveys, the NOAA Fisheries is exploring using mail surveys to estimate saltwater fishing activity in the U.S. The results of an initial pilot mail survey were promising in terms of response rates and coverage, but there was evidence of avidity bias. This paper describes a revision of pilot study in which the mail survey was re-designed to reduce avidity bias, and experiments were implemented to explore methods of improving the response rates. Some of the findings with respect to avidity bias and mailing methods to increase response rates are described.
In the United States, information about recreational saltwater fishing effort is generally collec... more In the United States, information about recreational saltwater fishing effort is generally collected through the Coastal Household Telephone Survey (CHTS), a random-digit-dial (RDD) survey of coastal county households. Like many RDD surveys, CHTS response rates have declined in recent years, increasing the potential for nonresponse error in survey estimates. Because little is known about nonrespondents, the CHTS accounts for unit nonresponse through a simple stratum cell-weighting adjustment that assumes random nonresponse. This may be an invalid assumption as previous studies have demonstrated that avid anglers are more likely to respond to fishing surveys than non anglers. Here we describe efforts to enrich the CHTS sample, and subsequently define more meaningful weighting cells, by matching the sample to external data sources, specifically, databases of licensed saltwater anglers. We assess the extent to which CHTS estimates are subjected to nonresponse error resulting from diffe...
Kernel Density Estimator From Ranked Set Samples Johan Lim, Min Chen, Sangun Park, Xinlei Wang & ... more Kernel Density Estimator From Ranked Set Samples Johan Lim, Min Chen, Sangun Park, Xinlei Wang & Lynne Stokes a Department of Statistics, Seoul National University, Seoul, Korea b Department of Mathematical Sciences, University of Texas at Dallas, Dallas, Texas, USA c Department of Applied Statistics, Yonsei University, Seoul, Korea d Department of Statistical Science, Southern Methodist University, Dallas, Texas, USA Published online: 23 Apr 2014.
Journal of the American Statistical Association, 1988
The loss of precision in estimates of means due to variability among interviewers can be substant... more The loss of precision in estimates of means due to variability among interviewers can be substantial for some questionnaire items and survey designs. The most commonly used methods for estimating the magnitude of this loss are inappropriate for binary items in complex surveys. This article shows how parameters from a model for variance components in binary variables (Anderson and Aitkin 1985) are related to the increased variance of population estimates. It is suggested that one of these parameters, a measure of correlation between interviewer observations on a latent variable, may be more appropriate than the intrainterviewer correlation p for measuring the magnitude of interviewer effects. This is because it is unaffected by the level of the attribute in the population, which is not true of p. Interviewer effects for a household respondent's recorded labor-force status in a random digit dial telephone survey are examined using the new model and estimation process. Small but positive effects are present for at least one of the labor-force categories.
Recently, voter ID laws have been instituted, modified, or overturned in many states in the U.S. ... more Recently, voter ID laws have been instituted, modified, or overturned in many states in the U.S. As these laws change, it is important to have accurate measures of their impact. We present the data collection methods and results of class projects that attempted to quantify the impact of the voter ID laws in areas of three states. We also summarize the types of data used to assess the impact of voter ID laws and discuss how our data address some of the shortcomings of the usual techniques for assessing the impact of voter ID laws.
This research was motivated by the desire to more efficiently estimate catch by recreational angl... more This research was motivated by the desire to more efficiently estimate catch by recreational anglers than current methods do. The method illustrated here combines data from angler self-reports made by a smartphone app and dockside validation samples. The two data sources can be thought of as a capture and recapture, where the parameter of interest is the population total (catch) instead of the population size. We developed several estimators of the total and compared them to one that makes use only of catch observed in the validation sample but not self-reports of catch. All the proposed estimators allow measurement error in the self-reports and do not make any assumptions about their representativeness. The validation sample must be a probability sample for valid inference, and our estimators can accommodate a complex sample design. We provide recommendations about conditions under which one of the estimators discussed may be preferred to another. Finally, we illustrate the method with analysis of data from a pilot project to estimate recreational red snapper catch in the Gulf of Mexico off the coast of Texas.
Information organizations, such as statistical agencies, must ensure that data access does not co... more Information organizations, such as statistical agencies, must ensure that data access does not compromise the confidentiality afforded data providers, whether individuals or establishments. Recognizing that deidentification of data is generally inadequate to protect confidentiality against attack by a data snooper, information organizations (IOs)-such as statistical agencies, data archives, and trade associations-can implement a variety of disclosure limitation (DL) techniques-such as topcoding, noise addition and data swapping-in developing data products. Desirably, the resulting restricted data have both high data utility U to data users and low disclosure risk R from data snoopers. IOs lack a framework for examining tradeoffs between R and U under a specific DL procedure. They also lack systematic ways of comparing the performance of distinct DL procedures. To provide this framework and facilitate comparisons, the R-U confidentiality map is introduced to trace the joint impact on R and U to changes in the parameters of a DL procedure. Implementation of an R-U confidentiality map is illustrated in the case of multivariate noise addition. Analysis is provided for two important multivariate estimation problems: a data user seeks to estimate linear combinations of means and to estimate regression coefficients.
Fellegi's (1974) improved method for estimating the interviewer component of correlated resp... more Fellegi's (1974) improved method for estimating the interviewer component of correlated response variance is extended to l groups of k interviewer assignments for general multistage survey designs. Using a linear models approach suggestive of Hartley and Rao's (1978), the independence of the two estimators of interviewer variance is established and the forms of the variances of the estimators are derived.
The accuracy of the Content should not be relied upon and should be independently verified with p... more The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
We generalize the denition of a concomitant of an order statistic in the multivariate case, devel... more We generalize the denition of a concomitant of an order statistic in the multivariate case, develop general expressions for its density, and establish related properties. The concomitant of a normal random vector is studied in detail, and methods for calculating its moments are discussed. Furthermore, we apply the theory to develop new estimators of the mean from a judgement post-stratied sample, where post-strata are formed by rank classes of auxiliary variables. Our estimators are shown to be more ecient than existing ones and robust against violations of the normality assumption. They are also well suited to applications requiring cost eciency.
Ranked-set sampling has been shown to provide improved estimators of the mean and variance when a... more Ranked-set sampling has been shown to provide improved estimators of the mean and variance when actual measurement of the observations is difficult but ranking of the elements in a sample is relatively easy. This result holds even if the ranking is imperfect. In ...
The importance of assessing fidelity has been emphasized recently with increasingly sophisticated... more The importance of assessing fidelity has been emphasized recently with increasingly sophisticated definitions, assessment procedures, and integration of fidelity data into analyses of outcomes. Fidelity is often measured through observation and coding of instructional sessions either live or by video. However, little guidance has been provided about how to determine the number of observations needed to precisely measure fidelity. We propose a practical method for determining a reasonable sample size for fidelity data collection when fidelity assessment requires observation. The proposed methodology is based on consideration of the power of tests of the treatment effect of outcome itself, as well as of the relationship between fidelity and outcome. It makes use of the methodology of probability sampling from a finite population, because the fidelity parameters of interest are estimated over a specific, limited time frame using a sample. For example, consider a fidelity measure define...
Annals of the Institute of Statistical Mathematics
Ranked set sampling was introduced by McIntyre (1952, Australian Journal of Agricultural Research... more Ranked set sampling was introduced by McIntyre (1952, Australian Journal of Agricultural Research, 3, 385-390) as a cost-effective method of selecting data if observations are much more cheaply ranked than measured. He proposed its use for estimating the population mean when the distribution of the data was unknown. In this paper, we examine the advantage, if any, that this method of sampling has if the distribution is known, for a specific family of distributions. Specifically, we consider estimation of # and G for the family of random variables with cdf's of the form F(~-~a). We find that the ranked set sample does provide more information about both # and G than a random sample of the same number of observations. We examine both maximum likelihood and best linear unbiased estimation of # and G, as well as methods for modifying the ranked set sampling procedure to provide even better estimation.
Due to declining response rates and coverage loss in random digit dial surveys, the NOAA Fisherie... more Due to declining response rates and coverage loss in random digit dial surveys, the NOAA Fisheries is exploring using mail surveys to estimate saltwater fishing activity in the U.S. The results of an initial pilot mail survey were promising in terms of response rates and coverage, but there was evidence of avidity bias. This paper describes a revision of pilot study in which the mail survey was re-designed to reduce avidity bias, and experiments were implemented to explore methods of improving the response rates. Some of the findings with respect to avidity bias and mailing methods to increase response rates are described.
In the United States, information about recreational saltwater fishing effort is generally collec... more In the United States, information about recreational saltwater fishing effort is generally collected through the Coastal Household Telephone Survey (CHTS), a random-digit-dial (RDD) survey of coastal county households. Like many RDD surveys, CHTS response rates have declined in recent years, increasing the potential for nonresponse error in survey estimates. Because little is known about nonrespondents, the CHTS accounts for unit nonresponse through a simple stratum cell-weighting adjustment that assumes random nonresponse. This may be an invalid assumption as previous studies have demonstrated that avid anglers are more likely to respond to fishing surveys than non anglers. Here we describe efforts to enrich the CHTS sample, and subsequently define more meaningful weighting cells, by matching the sample to external data sources, specifically, databases of licensed saltwater anglers. We assess the extent to which CHTS estimates are subjected to nonresponse error resulting from diffe...
Kernel Density Estimator From Ranked Set Samples Johan Lim, Min Chen, Sangun Park, Xinlei Wang & ... more Kernel Density Estimator From Ranked Set Samples Johan Lim, Min Chen, Sangun Park, Xinlei Wang & Lynne Stokes a Department of Statistics, Seoul National University, Seoul, Korea b Department of Mathematical Sciences, University of Texas at Dallas, Dallas, Texas, USA c Department of Applied Statistics, Yonsei University, Seoul, Korea d Department of Statistical Science, Southern Methodist University, Dallas, Texas, USA Published online: 23 Apr 2014.
Journal of the American Statistical Association, 1988
The loss of precision in estimates of means due to variability among interviewers can be substant... more The loss of precision in estimates of means due to variability among interviewers can be substantial for some questionnaire items and survey designs. The most commonly used methods for estimating the magnitude of this loss are inappropriate for binary items in complex surveys. This article shows how parameters from a model for variance components in binary variables (Anderson and Aitkin 1985) are related to the increased variance of population estimates. It is suggested that one of these parameters, a measure of correlation between interviewer observations on a latent variable, may be more appropriate than the intrainterviewer correlation p for measuring the magnitude of interviewer effects. This is because it is unaffected by the level of the attribute in the population, which is not true of p. Interviewer effects for a household respondent's recorded labor-force status in a random digit dial telephone survey are examined using the new model and estimation process. Small but positive effects are present for at least one of the labor-force categories.
Recently, voter ID laws have been instituted, modified, or overturned in many states in the U.S. ... more Recently, voter ID laws have been instituted, modified, or overturned in many states in the U.S. As these laws change, it is important to have accurate measures of their impact. We present the data collection methods and results of class projects that attempted to quantify the impact of the voter ID laws in areas of three states. We also summarize the types of data used to assess the impact of voter ID laws and discuss how our data address some of the shortcomings of the usual techniques for assessing the impact of voter ID laws.
This research was motivated by the desire to more efficiently estimate catch by recreational angl... more This research was motivated by the desire to more efficiently estimate catch by recreational anglers than current methods do. The method illustrated here combines data from angler self-reports made by a smartphone app and dockside validation samples. The two data sources can be thought of as a capture and recapture, where the parameter of interest is the population total (catch) instead of the population size. We developed several estimators of the total and compared them to one that makes use only of catch observed in the validation sample but not self-reports of catch. All the proposed estimators allow measurement error in the self-reports and do not make any assumptions about their representativeness. The validation sample must be a probability sample for valid inference, and our estimators can accommodate a complex sample design. We provide recommendations about conditions under which one of the estimators discussed may be preferred to another. Finally, we illustrate the method with analysis of data from a pilot project to estimate recreational red snapper catch in the Gulf of Mexico off the coast of Texas.
Information organizations, such as statistical agencies, must ensure that data access does not co... more Information organizations, such as statistical agencies, must ensure that data access does not compromise the confidentiality afforded data providers, whether individuals or establishments. Recognizing that deidentification of data is generally inadequate to protect confidentiality against attack by a data snooper, information organizations (IOs)-such as statistical agencies, data archives, and trade associations-can implement a variety of disclosure limitation (DL) techniques-such as topcoding, noise addition and data swapping-in developing data products. Desirably, the resulting restricted data have both high data utility U to data users and low disclosure risk R from data snoopers. IOs lack a framework for examining tradeoffs between R and U under a specific DL procedure. They also lack systematic ways of comparing the performance of distinct DL procedures. To provide this framework and facilitate comparisons, the R-U confidentiality map is introduced to trace the joint impact on R and U to changes in the parameters of a DL procedure. Implementation of an R-U confidentiality map is illustrated in the case of multivariate noise addition. Analysis is provided for two important multivariate estimation problems: a data user seeks to estimate linear combinations of means and to estimate regression coefficients.
Fellegi's (1974) improved method for estimating the interviewer component of correlated resp... more Fellegi's (1974) improved method for estimating the interviewer component of correlated response variance is extended to l groups of k interviewer assignments for general multistage survey designs. Using a linear models approach suggestive of Hartley and Rao's (1978), the independence of the two estimators of interviewer variance is established and the forms of the variances of the estimators are derived.
The accuracy of the Content should not be relied upon and should be independently verified with p... more The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
We generalize the denition of a concomitant of an order statistic in the multivariate case, devel... more We generalize the denition of a concomitant of an order statistic in the multivariate case, develop general expressions for its density, and establish related properties. The concomitant of a normal random vector is studied in detail, and methods for calculating its moments are discussed. Furthermore, we apply the theory to develop new estimators of the mean from a judgement post-stratied sample, where post-strata are formed by rank classes of auxiliary variables. Our estimators are shown to be more ecient than existing ones and robust against violations of the normality assumption. They are also well suited to applications requiring cost eciency.
Ranked-set sampling has been shown to provide improved estimators of the mean and variance when a... more Ranked-set sampling has been shown to provide improved estimators of the mean and variance when actual measurement of the observations is difficult but ranking of the elements in a sample is relatively easy. This result holds even if the ranking is imperfect. In ...
The importance of assessing fidelity has been emphasized recently with increasingly sophisticated... more The importance of assessing fidelity has been emphasized recently with increasingly sophisticated definitions, assessment procedures, and integration of fidelity data into analyses of outcomes. Fidelity is often measured through observation and coding of instructional sessions either live or by video. However, little guidance has been provided about how to determine the number of observations needed to precisely measure fidelity. We propose a practical method for determining a reasonable sample size for fidelity data collection when fidelity assessment requires observation. The proposed methodology is based on consideration of the power of tests of the treatment effect of outcome itself, as well as of the relationship between fidelity and outcome. It makes use of the methodology of probability sampling from a finite population, because the fidelity parameters of interest are estimated over a specific, limited time frame using a sample. For example, consider a fidelity measure define...
Uploads
Papers by Lynne Stokes