The Opportunities and Challenges of Using Administrative Data Linkages to Evaluate Mobility

• Published date: 01 January 2015
• Author: Amy O’Hara,David S. Johnson,Catherine Massey
• Date: 01 January 2015
• DOI: 10.1177/0002716214552780
• Subject Matter: Section III: Issues of Implementation

ANNALS, AAPSS, 657, January 2015 247

DOI: 10.1177/0002716214552780

The

Opportunities

and Challenges

of Using

Administrative

Data Linkages

to Evaluate

Mobility

By

DAVID S. JOHNSON,

CATHERINE MASSEY,

and

AMY O’HARA

552780ANN The Annals of the American AcademyUsing Administrative Data Linkages to Evaluate Mobility

research-article2014

Since Alan Krueger’s christening of the Great Gatsby

curve, there has been increased attention given to the

relationship between inequality and intergenerational

social mobility in the United States. Studying intergen-

erational mobility (IGM) requires longitudinal data

across large spans of time as well as the ability to follow

parents and children over multiple generations. Few

longitudinal datasets meet this need. This article sur-

veys available data and the current and potential issues

surrounding the use of administrative records to vastly

extend the study of IGM. First, we describe the U.S.

Census Bureau’s current uses of administrative records

in the linkage of households across household surveys

such as the Current Population Survey (CPS), American

Community Survey (ACS), Survey of Income and

Program Participation (SIPP), and the decennial cen-

suses. Then, we describe the possibilities of creating

additional parent-child linkages using the SIPP linked

to decennial censuses and the ACS. Last, we outline

our model to create linkages across earlier census data

(e.g., 1980 and 1990) and contemporary surveys.

Keywords: intergenerational mobility; child develop-

ment; record linkage; administrative

records

With the advent of the Great Gatsby curve,

there has been increased attention on

inequality and intergenerational mobility

(IGM) in the United States.1 Much of the

recent interest is in determining absolute and

relative differences in economic and social

well-being between generations. Although

David S. Johnson is the chief economist at the Bureau

of Economic Analysis (BEA), where he coordinates

cross-program research projects and develops new sup-

plemental measures and distributional measures for the

national accounts. Before joining BEA, he was chief of

the Social, Economic, and Housing Statistics Division

at the U.S. Census Bureau.

Catherine Massey is an economist in the Center for

Administrative Records Research and Applications at

the U.S. Census Bureau. Her current areas of research

include intergenerational mobility, immigration, and

record linkage.

248 THE ANNALS OF THE AMERICAN ACADEMY

multiple datasets observe both parents and children at one point in time, the

study of mobility requires linked parent-child information across multiple gen-

erations. One way to address this issue is to create linked parent-child data from

existing surveys that include both parents and children. In this article, we

describe the current uses of administrative records data at the U.S. Census

Bureau and their use in the linkage of individuals across household surveys (e.g.,

Current Population Survey [CPS], American Community Survey [ACS], Survey

of Income Program Participation [SIPP]) and the decennial censuses, as well as

their potential to aid in the collection of intergenerational information.

The recent Committee on National Statistics (CNSTAT) report, “Reengineering

the Survey of Income and Program Participation,” suggests two types of uses for

administrative data—indirect and direct (National Research Council 2009). The

indirect use includes assessing data quality, imputing for missing data, and aiding

in linking survey data. The direct use includes research using only administrative

data (Chetty etal. 2014; Mitnik etal. 2014), replacing survey data, and research

to augment survey data.

This article outlines how the U.S. Census Bureau uses administrative data both

directly and indirectly, and how these uses are applicable to the study of mobility.

First, we highlight the difficulty in acquiring administrative data, current uses of

administrative data, and appropriate disclosure avoidance requirements necessary

to utilize administrative records for research. Next, we describe the role of admin-

istrative records in the record linkage activities of the Census Bureau. We also

discuss the creation of a core set of linked decennial census and ACS data, as well

as the possibilities of creating additional parent-child linkages using the SIPP. We

build upon the concepts presented in Warren (2013) and the statement of purpose

for the National Academy of Science (NAS) background paper “Issues in Designing

a New National Survey on Social Mobility.” Last, we present the issues and possi-

bilities of using data from the Social Security Administration (SSA) and Internal

Revenue Service (IRS) to create further links between parents and children.

Precedents for Linking Census Data and Measuring

Mobility

The practice of linking decennial census data dates back to the late 1930s

(Ferrie 2004) but became widely used following Steckel (1988, 1989) and

Ferrie (1996). Their work created the first national linked samples of the

Amy O’Hara is chief of the Center for Administrative Records Research and Applications at

the U.S. Census Bureau. She leads efforts to integrate federal, state, and third party data with

censuses and surveys to create new social and economic measurements, improve operational

efficiency, and reduce respondent burden.

NOTE: This article is intended to inform interested parties of ongoing research at the U.S.

Census Bureau and the Bureau of Economic Analysis and to encourage discussion. The views

expressed are those of the authors and not necessarily those of the U.S. Census Bureau or the

Bureau of Economic Analysis.