Who are the people in my neighborhood? The 'contextual fallacy' of measuring individual context with census geographies
February 2018
Working Paper Number:
CES-18-11
Abstract
Scholars deploy census-based measures of neighborhood context throughout the social sciences and epidemiology. Decades of research confirm that variation in how individuals are aggregated into geographic units to create variables that control for social, economic or political contexts can dramatically alter analyses. While most researchers are aware of the problem, they have lacked the tools to determine its magnitude in the literature and in their own projects. By using confidential access to the complete 2010 U.S. Decennial Census, we are able to construct'for all persons in the US'individual-specific contexts, which we group according to the Census-assigned block, block group, and tract. We compare these individual-specific measures to the published statistics at each scale, and we then determine the magnitude of variation in context for an individual with respect to the published measures using a simple statistic, the standard deviation of individual context (SDIC). For three key measures (percent Black, percent Hispanic, and Entropy'a measure of ethno-racial diversity), we find that block-level Census statistics frequently do not capture the actual context of individuals within them. More problematic, we uncover systematic spatial patterns in the contextual variables at all three scales. Finally, we show that within-unit variation is greater in some parts of the country than in others. We publish county-level estimates of the SDIC statistics that enable scholars to assess whether mis-specification in context variables is likely to alter analytic findings when measured at any of the three common Census units.
Document Tags and Keywords
Keywords
Keywords are automatically generated using KeyBERT, a powerful and innovative
keyword extraction tool that utilizes BERT embeddings to ensure high-quality and contextually relevant
keywords.
By analyzing the content of working papers, KeyBERT identifies terms and phrases that capture the essence of the
text, highlighting the most significant topics and trends. This approach not only enhances searchability but
provides connections that go beyond potentially domain-specific author-defined keywords.
:
data census,
census research,
respondent,
black,
hispanic,
immigrant,
white,
metropolitan,
segregation,
geographically,
population,
racial,
race,
geography,
immigration,
ancestry,
neighborhood,
census bureau,
resident,
geographic,
neighbor,
2010 census,
census responses,
race census
Tags
Tags are automatically generated using a pretrained language model from spaCy, which excels at
several tasks, including entity tagging.
The model is able to label words and phrases by part-of-speech,
including "organizations." By filtering for frequent words and phrases labeled as "organizations", papers are
identified to contain references to specific institutions, datasets, and other organizations.
:
National Science Foundation,
Decennial Census,
Chicago Census Research Data Center,
Research Data Center,
Geographic Information Systems,
2010 Census,
Federal Statistical Research Data Center
Similar Working Papers
Similarity between working papers are determined by an unsupervised neural
network model
know as Doc2Vec.
Doc2Vec is a model that represents entire documents as fixed-length vectors, allowing for the
capture of semantic meaning in a way that relates to the context of words within the document. The model learns to
associate a unique vector with each document while simultaneously learning word vectors, enabling tasks such as
document classification, clustering, and similarity detection by preserving the order and structure of words. The
document vectors are compared using cosine similarity/distance to determine the most similar working papers.
Papers identified with 🔥 are in the top 20% of similarity.
The 10 most similar working papers to the working paper 'Who are the people in my neighborhood? The 'contextual fallacy' of measuring individual context with census geographies' are listed below in order of similarity.
-
Working PaperImproving Estimates of Neighborhood Change with Constant Tract Boundaries
May 2022
Working Paper Number:
CES-22-16
Social scientists routinely rely on methods of interpolation to adjust available data to their research needs. This study calls attention to the potential for substantial error in efforts to harmonize data to constant boundaries using standard approaches to areal and population interpolation. We compare estimates from a standard source (the Longitudinal Tract Data Base) to true values calculated by re-aggregating original 2000 census microdata to 2010 tract areas. We then demonstrate an alternative approach that allows the re-aggregated values to be publicly disclosed, using 'differential privacy' (DP) methods to inject random noise to protect confidentiality of the raw data. The DP estimates are considerably more accurate than the interpolated estimates. We also examine conditions under which interpolation is more susceptible to error. This study reveals cause for greater caution in the use of interpolated estimates from any source. Until and unless DP estimates can be publicly disclosed for a wide range of variables and years, research on neighborhood change should routinely examine data for signs of estimation error that may be substantial in a large share of tracts that experienced complex boundary changes.View Full Paper PDF
-
Working PaperStructural versus Ethnic Dimensions of Housing Segregation
March 2016
Working Paper Number:
CES-16-22
Racial residential segregation is still very high in many American cities. Some portion of segregation is attributable to socioeconomic differences across racial lines; some portion is caused by purely racial factors, such as preferences about the racial composition of one's neighborhood or discrimination in the housing market. Social scientists have had great difficulty disaggregating segregation into a portion that can be explained by interracial differences in socioeconomic characteristics (what we call structural factors) versus a portion attributable to racial and ethnic factors. What would such a measure look like? In this paper, we draw on a new source of data to develop an innovative structural segregation measure that shows the amount of segregation that would remain if we could assign households to housing units based only on non-racial socioeconomic characteristics. This inquiry provides vital building blocks for the broader enterprise of understanding and remedying housing segregation.View Full Paper PDF
-
Working PaperMetropolitan Segregation: No Breakthrough in Sight
May 2022
Working Paper Number:
CES-22-14
The 2020 Census offers new information on changes in residential segregation in metropolitan regions across the country as they continue to become more diverse. We take a long view, assessing trends since 1980 and extrapolating to the future. These new data mostly reinforce patterns that were observed a decade ago: high but slowly declining black-white segregation, and less intense but hardly changing segregation of Hispanics and Asians from whites. Enough time has passed since the civil rights era of the 1960s and 1970s to draw this conclusion: segregation will continue to divide Americans well into the 21st Century.View Full Paper PDF
-
Working PaperExamining Multi-Level Correlates of Suicide by Merging NVDRS and ACS Data
January 2017
Working Paper Number:
CES-17-25
This paper describes a novel database and an associated suicide event prediction model that surmount longstanding barriers in suicide risk factor research. The database comingles person-level records from the National Violent Death Reporting System (NVDRS) and the American Community Survey (ACS) to establish a case-control study sample that includes all identified suicide cases, while faithfully reflecting general population sociodemographics, in sixteen USA states during the years 2005 2011. It supports a statistical model of individual suicide risk that accommodates person-level factors and the moderation of these factors by their community rates. Named the United States Multi-Level Suicide Data Set (US-MSDS), the database was developed outside the RDC laboratory using publicly available ACS microdata, and reconstructed inside the laboratory using restricted access ACS microdata. Analyses of the latter version yielded findings that largely amplified but also extended those obtained from analyses of the former. This experience shows that the analytic precision achievable using restricted access ACS data can play an important role in conducting social research, although it also indicates that publicly available ACS data have considerable value in conducting preliminary analyses and preparing to use an RDC laboratory. The database development strategy may interest scientists investigating sociodemographic risk factors for other types of low-frequency mortality.View Full Paper PDF
-
Working PaperLocation, Location, Location: The 3L Approach to House Price Determination
May 2004
Working Paper Number:
CES-04-06
The immobility of houses means that their location affects their values. This explains the common belief that three things determine the price of a house: location, location, and location. We use this notion to develop the 3L Approach to house price determination. That is, prices are determined by the Metropolitan Statistical Area (MSA), town, and street where the house is located. This study creates a unique data set based on data from the American Housing Survey (AHS) consisting of small 'clusters' of housing units with information on their housing characteristics and resident characteristics that is merged with census tract-level attributes. We use this data to verify the 3L Approach: we find that all three levels of location are significant when estimating the house price hedonic equation. This indicates that individuals care about their local neighborhood, i.e. the general upkeep of their street and possibly their neighbors' characteristics (cluster variables), a broader area such as the school district and/or the town (tract variables) that account for school quality and crime rates, and the particular amenities found in their MSA.View Full Paper PDF
-
Working PaperAssociations Between Public Housing and Individual Earnings in New Orleans
October 2015
Working Paper Number:
CES-15-32
This study uses a sample of the civilian labor force aged 16-64 constructed from the Decennial Census and American Community Survey, along with data from the HUD dataset Picture of Subsidized Households, to compare the likelihood for job earnings in relation to public housing developments in the New Orleans MSA before and after Hurricane Katrina. Results from a series of hierarchical linear models (HLM) indicate significant relationships are altered between time periods, including those from public and mixed-income developments, suggesting a fluid relationship between neighborhoods and economic outcomes during physical, demographic and economic restructuring.View Full Paper PDF
-
Working PaperDynamics of Race: Joining, Leaving, and Staying in the American Indian/Alaska Native Race Category between 2000 and 2010
August 2014
Working Paper Number:
carra-2014-10
Each census for decades has seen the American Indian and Alaska Native population increase substantially more than expected. Changes in racial reporting seem to play an important role in the observed net increases, though research has been hampered by data limitations. We address previously unanswerable questions about race response change among American Indian and Alaska Natives (hereafter 'American Indians') using uniquely-suited (but not nationally representative) linked data from the 2000 and 2010 decennial censuses (N = 3.1 million) and the 2006-2010 American Community Survey (N = 188,131). To what extent do people change responses to include or exclude American Indian? How are people who change responses similar to or different from those who do not? How are people who join a group similar to or different from those who leave it? We find considerable race response change by people in our data, especially by multiple-race and/or Hispanic American Indians. This turnover is hidden in cross-sectional comparisons because people joining the group are similar in number and characteristics to those who leave the group. People in our data who changed their race response to add or drop American Indian differ from those who kept the same race response in 2000 and 2010 and from those who moved between a single-race and multiple-race American Indian response. Those who consistently reported American Indian (including those who added or dropped another race response) were relatively likely to report a tribe, live in an American Indian area, report American Indian ancestry, and live in the West. There are significant differences between those who joined and those who left a specific American Indian response group, but poor model fit indicates general similarity between joiners and leavers. Response changes should be considered when conceptualizing and operationalizing 'the American Indian and Alaska Native population.'View Full Paper PDF
-
Working PaperWHITE-LATINO RESIDENTIAL ATTAINMENTS AND SEGREGATION IN SIX CITIES: ASSESSING THE ROLE OF MICRO-LEVEL FACTORS
January 2016
Working Paper Number:
CES-16-51
This study examines the residential outcomes of Latinos in major metropolitan areas using new methods to connect micro-level analyses of residential attainments to overall patterns of segregation in the metropolitan area. Drawing on new formulations of standard measures of evenness, we conduct micro-level multivariate analyses using the restricted-use census microdata files to predict segregation-relevant neighborhood outcomes for individuals by race. We term the dependent variables segregation-relevant neighborhood outcomes because the differences in average outcomes for each group on these variables determine the values of the aggregate measures of evenness. This approach allows me to use standardization and components analysis to quantitatively assess the separate contributions that differences in social characteristics and differences in rates of return make towards determining the overall disparity in residential outcomes ' that is, the level of segregation ' between Whites and Latinos. Based on our micro-level residential attainment analyses we find that for Latinos, acculturation and gains in socioeconomic status are associated with greater residential contact with Whites, in agreement with spatial assimilation theory, which promotes lower segregation. However, our standardization and components analyses reveals that a substantial portion of White-Latino disparities in residential contact with Whites can be attributed to differences in rates of return; that is White-Latino differences in the ability to translate acculturation and gains in socioeconomic status into more residential contact with Whites. This is further elaborated upon by assessing the changes in contact with Whites for Whites and Latinos after manipulating single variables while holding all others constant. This can be interpreted as the role of discrimination which is emphasized by place stratification theory. Therefore we conclude that while members of minority groups make gains in residential outcomes that reduce segregation by attaining parity with Whites on social characteristics as spatial assimilation theory would predict, a substantial disparity will persist as Latinos cannot translate those gains into greater contact with Whites at the rate that Whites can. At the aggregate level of analysis, this means that White-Latino segregation remains substantial even when groups are equalized on social and economic characteristics.View Full Paper PDF
-
Working PaperPlace of Work and Place of Residence: Informal Hiring Networks and Labor Market Outcomes
October 2005
Working Paper Number:
CES-05-23
We use a novel dataset and research design to empirically detect the effect of social interactions among neighbors on labor market outcomes. Specifically, using Census data that characterize residential and employment locations down to the city block, we examine whether individuals residing in the same block are more likely to work together than those in nearby blocks. We find evidence of significant social interactions operating at the block level: residing on the same versus nearby blocks increases the probability of working together by over 33 percent. The results also indicate that this referral effect is stronger when individuals are similar in sociodemographic characteristics (e.g., both have children of similar ages) and when at least one individual is well attached to the labor market. These findings are robust across various specifications intended to address concerns related to sorting and reverse causation. Further, having determined the characteristics of a pair of individuals that lead to an especially strong referral effect, we provide evidence that the increased availability of neighborhood referrals has a significant impact on a wide range of labor market outcomes including employment and wages.View Full Paper PDF
-
Working PaperPeer Income Exposure Across the Income Distribution
February 2025
Working Paper Number:
CES-25-16
Children from families across the income distribution attend public schools, making schools and classrooms potential sites for interaction between more- and less-affluent children. However, limited information exists regarding the extent of economic integration in these contexts. We merge educational administrative data from Oregon with measures of family income derived from IRS records to document student exposure to economically diverse school and classroom peers. Our findings indicate that affluent children in public schools are relatively isolated from their less affluent peers, while low- and middle-income students experience relatively even peer income distributions. Students from families in the top percentile of the income distribution attend schools where 20 percent of their peers, on average, come from the top five income percentiles. A large majority of the differences in peer exposure that we observe arise from the sorting of students across schools; sorting across classrooms within schools plays a substantially smaller role.View Full Paper PDF
