Search

Big data has been fruitfully leveraged as a supplement for survey data—and sometimes as its replacement—and in the best of worlds, as a "force multiplier" to improve survey analytics and insight. We detail a use case, the big data classifier (BDC), as a replacement to the more traditional methods of targeting households in survey sampling for given specific household and personal attributes. Much like geographic targeting and the use of commercial vendor flags, we detail the ability of BDCs to predict the likelihood that any given household is, for example, one that contains a child or someone who is Hispanic. We specifically build 15 BDCs with the combined data from a large nationally representative probability-based panel and a range of big data from public and private sources, and then assess the effectiveness of these BDCs to successfully predict their range of predicted attributes across three large survey datasets. For each BDC and each data application, we compare the relative effectiveness of the BDCs against historical sample targeting techniques of geographic clustering and vendor flags. Overall, BDCs offer a modest improvement in their ability to target subpopulations. We find classes of predictions that are consistently more effective, and others where the BDCs are on par with vendor flagging, though always superior to geographic clustering. We present some of the relative strengths and weaknesses of BDCs as a new method to identify and subsequently sample low incidence and other populations. [ABSTRACT FROM AUTHOR]

Recent advances in computing technologies have enabled the development of low-cost, compact weather and air quality monitors. The U.S. federally funded Array of Things (AoT) project has deployed more than 140 such sensor nodes throughout the City of Chicago. This article combines a year's worth of AoT sensor data with household data collected from 450 elderly Chicagoans in order to explore the feasibility of using previously unavailable data on local environmental conditions to improve traditional neighborhood research. Specifically, we pilot the use of AoT sensor data to overcome limitations in research linking air pollution to poor physical and mental health and find support for recent findings that exposure to pollutants contributes to both respiratory- and dementia-related diseases. We expect that this support will become even stronger as sensing technologies continue to improve and more AoT nodes come online, enabling additional applications to social science research where environmental context matters. [ABSTRACT FROM AUTHOR]

This is a correction to: David Dutwin and others, Leveraging Predictive Modelling from Multiple Sources of Big Data to Improve Sample Efficiency and Reduce Survey Nonresponse Error, Journal of Survey Statistics and Methodology, 2023, smad016, https://doi.org/10.1093/jssam/smad016In the originally published version of this manuscript, the author Joshua Lerner was inadvertently omitted from the author byline.This error has been corrected.By David Dutwin; Patrick Coyle; Joshua Lerner; Ipek Bilgen and Ned EnglishReported by Author; Author; Author; Author; Author [Extracted from the article]

Objective: Studies assessing sociodemographic disparities in the tobacco retail environment have relied heavily on non-spatial analytical techniques, resulting in potentially misleading conclusions. We utilized a spatial analytical framework to evaluate neighborhood sociodemographic disparities in the tobacco retail environment in Washington, DC (DC) and the DC metropolitan statistical area (DC MSA).Methods: Retail tobacco availability for DC (n=177) and DC MSA (n=1,428) census tract was assessed using adaptive-bandwidth kernel density estimation. Density surfaces were constructed from DC (n=743) and DC MSA (n=4,539) geocoded tobacco retailers. Sociodemographics were obtained from the 2011-2015 American Community Survey. Spearman's correlations between sociodemographics and retail density were computed to account for spatial autocorrelation. Bivariate and multivariate spatial lag models were fit to predict retail density.Results: DC and DC MSA neighborhoods with a higher percentage of Hispanics were positively correlated with retail density (rho = .3392, P = .0001 and rho = .1191, P = .0000, respectively). DC neighborhoods with a higher percentage of African Americans were negatively correlated with retail density (rho = -.3774, P = .0000). This pattern was not significant in DC MSA neighborhoods. Bivariate and multivariate spatial lag models found a significant inverse relationship between the percentage of African Americans and retail density (Beta = -.0133, P = .0181 and Beta = -.0165, P = .0307, respectively).Conclusions: Associations between neighborhood sociodemographics and retail density were significant, although findings regarding African Americans are inconsistent with previous findings. Future studies should analyze other geographic areas, and account for spatial autocorrelation within their analytic framework. [ABSTRACT FROM AUTHOR]

Frames of residential mailing addresses based on U.S. Postal Service (USPS) sources are often used for selecting samples of housing units for surveys to be conducted in-person, by mail, by web, or using mixed modes. Address lists designed for mail delivery require some modifications for sampling purposes, which therefore require familiarity with aspects of the address files themselves. This paper takes a detailed look at the address components of these files along with the vendors from which such frames and samples are available. More specifically, this paper describes the types of address records and the components of an address record that are available. It discusses how vendors differ in both the services they provide with respect to sampling frames as well as in how those sampling frames are updated and maintained. The paper also details ways in which address-based frames can be enhanced with auxiliary data such as geocodes, area-level demographic variables, and commercial indicators at the individual address level. The match rate, append rate, accuracy, and utility of auxiliary variables are described, along with potential uses in survey design and estimation. Information contained within the paper will be useful for survey designers who have an interest in tailoring address lists for a given survey's target population and budget. [ABSTRACT FROM AUTHOR]

Address frames based largely on United States Postal Service lists can have shortcomings for in-person surveys in areas with non-locatable addresses. To give every housing unit a known, nonzero probability of selection, field methods may be implemented to improve frame coverage. Three coverage improvement methods currently in use in the survey research industry are compared—Enhanced Listing, Check for Housing Units Missed (CHUM), and Address Coverage Enhancement (ACE)—so that the reader may understand their similarities and differences. Such a comparison is useful for research practitioners and survey designers who need to consider nuances in frame construction methods that weigh cost, effectiveness, and ease of implementation. [ABSTRACT FROM AUTHOR]

The rising costs and declining response rates of traditional survey modes have spurred many organizations to conduct surveys online. Less expensive broadband connections and the popularity of smartphones have also made it easier for many to access the Internet. The General Social Survey (GSS) shows that the percentage of American adults who use the Internet increased from 69 percent in 2006 to 86 percent in 2014. The increased access has reduced some concerns about the representativeness of Internet surveys. However, there remains little research into coverage bias, which occurs if those not in the sampling frame differ from the target population on variables of interest. This raises a question: With the increase in Internet access, has there been any change in the coverage bias of web surveys? To assess coverage bias, we analyze the GSS to determine whether those with Internet access and those without it became more or less similar between 2006 and 2014. We calculate the potential coverage bias of web-only surveys over these years for sex, age, education, income, race, political ideology, urbanicity, and life satisfaction. We also compare the bias observed in the United States to that in Europe. Our results illustrate that relative coverage bias associated with education, income, race, and age declined between 2006 and 2014, but bias still exists. [ABSTRACT FROM AUTHOR]

Objectives. The sample for the second wave (2010) of National Social Life, Health, and Aging Project (NSHAP) was designed to increase the scientific value of the Wave 1 (2005) data set by revisiting sample members 5 years after their initial interviews and augmenting this sample where possible. Method. There were 2 important innovations. First, the scope of the study was expanded by collecting data from coresident spouses or romantic partners. Second, to maximize the representativeness of the Wave 2 data, nonrespondents from Wave 1 were again approached for interview in the Wave 2 sample. Results. The overall unconditional response rate for the Wave 2 panel was 74%; the conditional response rate of Wave 1 respondents was 89%; the conditional response rate of partners was 84%; and the conversion rate for Wave 1 nonrespondents was 26%. Discussion. The inclusion of coresident partners enhanced the study by allowing the examination of how intimate, household relationships are related to health trajectories and by augmenting the size of the NSHAP sample size for this and future waves. The uncommon strategy of returning to Wave 1 nonrespondents reduced potential bias by ensuring that to the extent possible the whole of the original sample forms the basis for the field effort. NSHAP Wave 2 achieved its field objectives of consolidating the panel, recruiting their resident spouses or romantic partners, and converting a significant proportion of Wave 1 nonrespondents. [ABSTRACT FROM AUTHOR]

Objectives. The sample for the second wave (2010) of National Social Life, Health, and Aging Project (NSHAP) was designed to increase the scientific value of the Wave 1 (2005) data set by revisiting sample members 5 years after their initial interviews and augmenting this sample where possible. Method. There were 2 important innovations. First, the scope of the study was expanded by collecting data from coresident spouses or romantic partners. Second, to maximize the representativeness of the Wave 2 data, nonrespondents from Wave 1 were again approached for interview in the Wave 2 sample. Results. The overall unconditional response rate for the Wave 2 panel was 74%; the conditional response rate of Wave 1 respondents was 89%; the conditional response rate of partners was 84%; and the conversion rate for Wave 1 nonrespondents was 26%. Discussion. The inclusion of coresident partners enhanced the study by allowing the examination of how intimate, household relationships are related to health trajectories and by augmenting the size of the NSHAP sample size for this and future waves. The uncommon strategy of returning to Wave 1 nonrespondents reduced potential bias by ensuring that to the extent possible the whole of the original sample forms the basis for the field effort. NSHAP Wave 2 achieved its field objectives of consolidating the panel, recruiting their resident spouses or romantic partners, and converting a significant proportion of Wave 1 nonrespondents. [ABSTRACT FROM AUTHOR]

An introduction to the journal is presented in which the editor discusses surveying the hard-to-reach (H2R) groups.

Objectives. We examined neighborhood-level foreclosure rates and their association with onset of depressive symptoms in older adults. Methods. We linked data from the National Social Life, Health, and Aging Project (2005-2006 and 2010-2011 waves), a longitudinal, nationally representative survey, to data on zip code-level foreclosure rates, and predicted the onset of depressive symptoms using logit-linked regression. Results. Multiple stages of the foreclosure process predicted the onset of depressive symptoms, with adjustment for demographic characteristics and changes in household assets, neighborhood poverty, and visible neighborhood disorder. A large increase in the number of notices of default (odds ratio [OR] = 1.75; 95% confidence interval [CI] = 1.14, 2.67) and properties returning to ownership by the bank (OR = 1.62; 95% CI = 1.06, 2.47) were associated with depressive symptoms. A large increase in properties going to auction was suggestive of such an association (OR = 1.45; 95% CI = 0.96, 2.19). Age, fewer years of education, and functional limitations also were predictive. Conclusions. Increases in neighborhood-level foreclosure represent an important risk factor for depression in older adults. These results accord with previous studies suggesting that the effects of economic crises are typically first experienced through deficits in emotional well-being. [ABSTRACT FROM AUTHOR]

Survey researchers have recently begun using the U.S. Postal Service Delivery Sequence File (DSF) as a sampling frame. In addition to address, the DSF contains ancillary data regarding the status and structure of the address. These ancillary data may prove useful in further enhancing frame construction and sampling methodologies. We review three variables on the DSF: the vacancy flag, the college flag, and the drop point indicator. We conclude that the ancillary information available on the DSF may be used with caution. Despite the limitations, the flags may be used to enhance the current frame. [ABSTRACT FROM AUTHOR]

Survey researchers increasingly look to commercially available address databases to create survey frames. These databases offer obvious cost and speed advantages over in-field listing of housing units but rely on geocoding to associate the addresses with census geographies such as tracts and blocks. While much recent research has looked at the coverage of frames derived from these databases, none has specifically explored the contributions of geocoding to frame coverage. Nationally, 83.3% of all residential addresses on the January 2010 Valassis database geocode precisely enough that they are likely to be placed in the correct block. Among the city-style addresses, 93.9% geocode at this level. Comparing the counts of addresses on the database to housing unit counts from the 2010 Census, we find that the national net coverage rate for in-person surveys can vary from 86.7% to 92.3%, depending on which addresses a survey uses on its frame. [ABSTRACT FROM AUTHOR]