Using information from telephone panel surveys to predict reasons for refusal

One of the key problems in conducting surveys is convincing people to participate.¦However, it is often difficult or impossible to determine why people refuse. Panel surveys¦provide information from previous waves that can offer valuable clues as to why people¦refuse to participate. If we are able to anticipate the reasons for refusal, then we¦may be able to take appropriate measures to encourage potential respondents to participate¦in the survey. For example, special training could be provided for interviewers¦on how to convince potential participants to participate.¦This study examines different influences, as determined from the previous wave,¦on refusal reasons that were given by the respondents in the subsequent wave of the¦telephone Swiss Household Panel. These influences include socio-demography, social¦inclusion, answer quality, and interviewer assessment of question understanding and¦of future participation. Generally, coefficients are similar across reasons, and¦between-respondents effects rather than within-respondents effects are significant.¦While 'No interest' reasons are easier to predict, the other reasons are more situational. Survey-specific issues are able to distinguish¦different reasons to some extent.

Criminal victimization surveys in 13 American cities : national crime panel surveys in Boston, Buffalo, Cincinnati, Houston, Miami, Milwaukee, Minneapolis, New Orleans, Oakland, Pittsburgh, San Diego, San Francisco, and Washington, D.C..

Mode of access: Internet.

Missing value imputation in longitudinal measures of alcohol consumption

Attrition in longitudinal studies can lead to biased results. The study is motivated by the unexpected observation that alcohol consumption decreased despite increased availability, which may be due to sample attrition of heavy drinkers. Several imputation methods have been proposed, but rarely compared in longitudinal studies of alcohol consumption. The imputation of consumption level measurements is computationally particularly challenging due to alcohol consumption being a semi-continuous variable (dichotomous drinking status and continuous volume among drinkers), and the non-normality of data in the continuous part. Data come from a longitudinal study in Denmark with four waves (2003-2006) and 1771 individuals at baseline. Five techniques for missing data are compared: Last value carried forward (LVCF) was used as a single, and Hotdeck, Heckman modelling, multivariate imputation by chained equations (MICE), and a Bayesian approach as multiple imputation methods. Predictive mean matching was used to account for non-normality, where instead of imputing regression estimates, "real" observed values from similar cases are imputed. Methods were also compared by means of a simulated dataset. The simulation showed that the Bayesian approach yielded the most unbiased estimates for imputation. The finding of no increase in consumption levels despite a higher availability remained unaltered. Copyright (C) 2011 John Wiley & Sons, Ltd.

Voices of the Poor: Poverty and Growth in Albania

This paper uses three waves of panel surveys at the household level to study growth and poverty in Albania over the period 2002-2004. It attempts to answer two main questions. The first question is directed at finding the micro determinants of growth and aims to expose the obstacles households face to improve their economic situation. The main focus of the analysis is to investigate the importance of health, education, and infrastructure indicators for income growth. The second question asks whether growth in Albania during the period 2002-2004 has been pro-poor. I find that there is some evidence for a convergence of incomes and a pro-poor growth, which has led to a substantial decrease in the number of people living under the poverty line. I also find that infrastructure has not been an important determinant for income mobility, and neither has health. Only the higher education of poor urban households seems to have affected prospects for growing out of poverty, and unexpectedly, the relationship is negative.

Combining Information from Multiple Sources in Bayesian Modeling

Surveys can collect important data that inform policy decisions and drive social science research. Large government surveys collect information from the U.S. population on a wide range of topics, including demographics, education, employment, and lifestyle. Analysis of survey data presents unique challenges. In particular, one needs to account for missing data, for complex sampling designs, and for measurement error. Conceptually, a survey organization could spend lots of resources getting high-quality responses from a simple random sample, resulting in survey data that are easy to analyze. However, this scenario often is not realistic. To address these practical issues, survey organizations can leverage the information available from other sources of data. For example, in longitudinal studies that suffer from attrition, they can use the information from refreshment samples to correct for potential attrition bias. They can use information from known marginal distributions or survey design to improve inferences. They can use information from gold standard sources to correct for measurement error.

This thesis presents novel approaches to combining information from multiple sources that address the three problems described above.

The first method addresses nonignorable unit nonresponse and attrition in a panel survey with a refreshment sample. Panel surveys typically suffer from attrition, which can lead to biased inference when basing analysis only on cases that complete all waves of the panel. Unfortunately, the panel data alone cannot inform the extent of the bias due to attrition, so analysts must make strong and untestable assumptions about the missing data mechanism. Many panel studies also include refreshment samples, which are data collected from a random sample of new

individuals during some later wave of the panel. Refreshment samples offer information that can be utilized to correct for biases induced by nonignorable attrition while reducing reliance on strong assumptions about the attrition process. To date, these bias correction methods have not dealt with two key practical issues in panel studies: unit nonresponse in the initial wave of the panel and in the

refreshment sample itself. As we illustrate, nonignorable unit nonresponse

can significantly compromise the analyst's ability to use the refreshment samples for attrition bias correction. Thus, it is crucial for analysts to assess how sensitive their inferences---corrected for panel attrition---are to different assumptions about the nature of the unit nonresponse. We present an approach that facilitates such sensitivity analyses, both for suspected nonignorable unit nonresponse

in the initial wave and in the refreshment sample. We illustrate the approach using simulation studies and an analysis of data from the 2007-2008 Associated Press/Yahoo News election panel study.

The second method incorporates informative prior beliefs about

marginal probabilities into Bayesian latent class models for categorical data.

The basic idea is to append synthetic observations to the original data such that

(i) the empirical distributions of the desired margins match those of the prior beliefs, and (ii) the values of the remaining variables are left missing. The degree of prior uncertainty is controlled by the number of augmented records. Posterior inferences can be obtained via typical MCMC algorithms for latent class models, tailored to deal efficiently with the missing values in the concatenated data.

We illustrate the approach using a variety of simulations based on data from the American Community Survey, including an example of how augmented records can be used to fit latent class models to data from stratified samples.

The third method leverages the information from a gold standard survey to model reporting error. Survey data are subject to reporting error when respondents misunderstand the question or accidentally select the wrong response. Sometimes survey respondents knowingly select the wrong response, for example, by reporting a higher level of education than they actually have attained. We present an approach that allows an analyst to model reporting error by incorporating information from a gold standard survey. The analyst can specify various reporting error models and assess how sensitive their conclusions are to different assumptions about the reporting error process. We illustrate the approach using simulations based on data from the 1993 National Survey of College Graduates. We use the method to impute error-corrected educational attainments in the 2010 American Community Survey using the 2010 National Survey of College Graduates as the gold standard survey.

Large panel-survey data demonstrated country-level and ethnic minority variation in consent for health record linkage

Objectives
To investigate individual, household and country variation in consent to health record linkage.

Study Design and Setting
Data from 50,994 individuals aged 16-74 years recruited to wave 1 of a large UK general purpose household survey (January 2009 – December 2010) were analysed using multi-level logistic regression models.

Results
Overall, 70.7% of respondents consented to record linkage. Younger age, marriage, tenure, car ownership and education were all significantly associated with consent, though there was little deviation from 70% in subgroups defined by these variables. There were small increases in consent rates in individuals with poor health when defined by self-reported long term limiting illness (adjusted OR 1.11; 95%CIs 1.06, 1.16), less so when defined by General Health Questionnaire score (adjusted OR=1.05; 95%CIs 1.00, 1.10), but the range in absolute consent rates between categories was generally less than 10%. Larger differences were observed for those of non-white ethnicity who were 38% less likely to consent (adjusted OR 0.62; 95%CIs 0.59, 0.66). Consent was higher in Scotland than England (adjusted OR 1.17; 95%CIs 1.06, 1.29) but lower in Northern Ireland (adjusted OR 0.56; 95%CIs 0.50, 0.63).

Conclusion
The modest overall level of systematic bias in consent to record linkage provides reassurance for record linkage potential in general purpose household surveys. However, the low consent rates amongst non-white ethnic minority survey respondents will further compound their low survey participation rates. The reason for the country-level variation requires further study.

A panel data approach to economic forecasting: the bias-corrected average forecast

In this paper, we propose a novel approach to econometric forecasting of stationary and ergodic time series within a panel-data framework. Our key element is to employ the bias-corrected average forecast. Using panel-data sequential asymptotics we show that it is potentially superior to other techniques in several contexts. In particular it delivers a zero-limiting mean-squared error if the number of forecasts and the number of post-sample time periods is sufficiently large. We also develop a zero-mean test for the average bias. Monte-Carlo simulations are conducted to evaluate the performance of this new technique in finite samples. An empirical exercise, based upon data from well known surveys is also presented. Overall, these results show promise for the bias-corrected average forecast.

Mexico: Combining monthly inflation predictions from surveys

We examine the problem of combining Mexican inflation predictions or projections provided by a biweekly survey of professional forecasters. Consumer price inflation in Mexico is measured twice a month. We consider several combining methods and advocate the use of dimension reduction techniques whose performance is compared with different benchmark methods, including the simplest average prediction. Missing values in the database are imputed by two different databased methods. The results obtained are basically robust to the choice of the imputation method. A preliminary analysis of the data was based on its panel data structure and showed the potential usefulness of using dimension reduction techniques to combine the experts' predictions. The main findings are: the first monthly predictions are best combined by way of the first principal component of the predictions available; the best second monthly prediction is obtained by calculating the median prediction and is more accurate than the first one.

Panel Discussion - Local Finns in Illinois

Finnish North American labor contributions and involvement in strikes such as the 1913-14 Michigan Copper Strike are being restored to the historical record and even commemorated; yet some Finnish American communities’ labor history still goes untold. We contend that in the case of DeKalb, Illinois, the Finnish American labor and strike history has been, in part, overshadowed in contemporary remembrance by the city’s promotion of traditional history and commemoration focused on the barbed wire barons. Local Finnish American labor involvement and participation in strikes appears to have been marginalized in favor of a subsequent historical narrative surrounding the capitalist entrepreneurship of elites. However, counter memories of labor struggles may be lost for a variety of reasons. External and internal forces make it difficult for marginalized groups to offer alternatives to the construction of collective memories that exclude them. These forces include, but are not limited to gradual assimilation into dominant culture, internal conflict within social movements, and fear of, or experience with, governmental repression. In our archival research, surveys and interviews with 2nd and 3rd generation Finnish American residents reveal the many forces of “forgetting” that can influence the counter memory of Finnish American labor history in certain communities.

Core outcomes and definitions for pediatric fever and neutropenia research: a consensus statement from an international panel.

BACKGROUND There are no specific recommendations for the design and reporting of studies of children with fever and neutropenia (FN). As a result, there is marked heterogeneity in the variables and outcomes that are reported and new definitions continue to emerge. These inconsistencies hinder the ability of researchers and clinicians to compare, contrast and combine results. The objective was to achieve expert consensus on a core set of variables and outcomes that should be measured and reported, as a minimum, in pediatric FN studies. PROCEDURE The Delphi method was used to achieve consensus among an international group of clinicians, pharmacists, researchers, and patient representatives. Four surveys focusing on (i) the identification of a core set of variables and outcomes; and (ii) definitions of these variables and outcomes, were administered electronically. Consensus was predefined as more than 80% agreement on any statement. RESULTS There were forty-five survey participants and the response rate ranged between 84 and 96%. There was consensus on eight core variables and 10 core outcomes that should be collected and reported in all studies of children with FN. Consensus definitions were identified for all of the core outcomes. CONCLUSION Using the Delphi method, expert consensus on a set of core variables and outcomes, and their corresponding definitions, was achieved. These core sets represent the minimum that should be collected and reported in all studies of children with FN. This will promote collaboration and ensure consistency and comparability between studies.

Outpatient prescription drugs : data collection and editing in the 1996 Medical Expenditure Panel Survey (HC-010A).

"June 2001."

Illinois Rural Life Panel survey : spring 2000 report : addressing issues about the quality of life, availability of skilled labor, housing, access to computer technology, impact of the Internet, and womens' health /

"First printing: March 2001"--T.p. verso.

Precast Prestressed Concrete Panel Subdecks in Skewed Bridges, Interim Report, 1990

Precast prestressed concrete panels have been used in bridge deck construction in Iowa and many other states. To investigate the performance of these panels at abutment or pier diaphragm locations for bridges with various skew angles, a research program involving both analytical and experimental aspects, is being conducted. This interim report presents the status of the research with respect to four tasks. Task 1 which involves a literature review and two surveys is essentially complete. Task 2 which involved field investigations of three Iowa bridges containing precast panel subdecks has been completed. Based on the findings of these investigations, future inspections are recommended to evaluate potential panel deterioration due to possible corrosion of the prestressed strands. Task 3 is the experimental program which has been established to monitor the behavior of five configurations of full scale composite deck slabs. Three dimensional test and instrumentation frameworks have been constructed to load and monitor the slab specimens. The first slab configuration representing an interior panel condition is being tested and preliminary results are presented for one of these tests in this interim report. Task 4 involves the analytical investigation of the experimental specimens. Finite element methods are being applied to analytically predict the behavior of the test specimens. The first test configuration of the interior panel condition has been analyzed for the same loads used in the laboratory, and the results are presented herein. Very good correlation between the analytical and experimental results has occurred.