836 lines
No EOL
91 KiB
HTML
836 lines
No EOL
91 KiB
HTML
<!doctype html>
|
|
<html lang="en" class="no-js">
|
|
<head>
|
|
<meta charset="UTF-8" />
|
|
<meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1" />
|
|
<meta name="viewport" content="width=device-width, initial-scale=1" />
|
|
<title>Measurement Issues Associated with Using Survey Data Matched with Administrative Data from the Social Security Administration</title>
|
|
<meta name="DCTERMS:dateCreated" content="2009-07" />
|
|
<meta name="DCTERMS:contentOffice" content="ORDP:ORES" />
|
|
<meta name="DCTERMS:contentOwner" content="publications@ssa.gov" />
|
|
<meta name="DCTERMS:coderOffice" content="ORDP:ORES:OD" />
|
|
<meta name="DCTERMS:coder" content="op.webmaster@ssa.gov" />
|
|
<meta name="DCTERMS:dateCertified" content="2025-01-01" />
|
|
<meta name="description" content="Social Security Administration Research, Statistics, and Policy Analysis" />
|
|
<meta property="og:site_name" content="Social Security Administration Research, Statistics, and Policy Analysis"/>
|
|
<link rel="stylesheet" href="/policy/styles/doc.css" />
|
|
<link rel="stylesheet" href="/policy/styles/global.css" />
|
|
<!-- SSA INTERNET HEAD SCRIPTS -->
|
|
<script src="https://code.jquery.com/jquery-3.7.1.min.js" integrity="sha256-/JqT3SQfawRcv/BIHPThkBvs0OEvtFFmqPF/lYI/Cxo=" crossorigin="anonymous"></script>
|
|
<script src="/framework/js/ssa.internet.head.js"></script>
|
|
|
|
<script>(window.BOOMR_mq=window.BOOMR_mq||[]).push(["addVar",{"rua.upush":"false","rua.cpush":"false","rua.upre":"false","rua.cpre":"false","rua.uprl":"false","rua.cprl":"false","rua.cprf":"false","rua.trans":"SJ-3a3bb884-f513-47e3-a86c-84bab05e21dc","rua.cook":"true","rua.ims":"false","rua.ufprl":"false","rua.cfprl":"false","rua.isuxp":"false","rua.texp":"norulematch","rua.ceh":"false","rua.ueh":"false","rua.ieh.st":"0"}]);</script>
|
|
<script>!function(e){var n="https://s.go-mpulse.net/boomerang/";if("False"=="True")e.BOOMR_config=e.BOOMR_config||{},e.BOOMR_config.PageParams=e.BOOMR_config.PageParams||{},e.BOOMR_config.PageParams.pci=!0,n="https://s2.go-mpulse.net/boomerang/";if(window.BOOMR_API_key="LERZW-HECFS-R8H4E-23UQ7-ERMQB",function(){function e(){if(!o){var e=document.createElement("script");e.id="boomr-scr-as",e.src=window.BOOMR.url,e.async=!0,i.parentNode.appendChild(e),o=!0}}function t(e){o=!0;var n,t,a,r,d=document,O=window;if(window.BOOMR.snippetMethod=e?"if":"i",t=function(e,n){var t=d.createElement("script");t.id=n||"boomr-if-as",t.src=window.BOOMR.url,BOOMR_lstart=(new Date).getTime(),e=e||d.body,e.appendChild(t)},!window.addEventListener&&window.attachEvent&&navigator.userAgent.match(/MSIE [67]\./))return window.BOOMR.snippetMethod="s",void t(i.parentNode,"boomr-async");a=document.createElement("IFRAME"),a.src="about:blank",a.title="",a.role="presentation",a.loading="eager",r=(a.frameElement||a).style,r.width=0,r.height=0,r.border=0,r.display="none",i.parentNode.appendChild(a);try{O=a.contentWindow,d=O.document.open()}catch(_){n=document.domain,a.src="javascript:var d=document.open();d.domain='"+n+"';void(0);",O=a.contentWindow,d=O.document.open()}if(n)d._boomrl=function(){this.domain=n,t()},d.write("<bo"+"dy onload='document._boomrl();'>");else if(O._boomrl=function(){t()},O.addEventListener)O.addEventListener("load",O._boomrl,!1);else if(O.attachEvent)O.attachEvent("onload",O._boomrl);d.close()}function a(e){window.BOOMR_onload=e&&e.timeStamp||(new Date).getTime()}if(!window.BOOMR||!window.BOOMR.version&&!window.BOOMR.snippetExecuted){window.BOOMR=window.BOOMR||{},window.BOOMR.snippetStart=(new Date).getTime(),window.BOOMR.snippetExecuted=!0,window.BOOMR.snippetVersion=12,window.BOOMR.url=n+"LERZW-HECFS-R8H4E-23UQ7-ERMQB";var i=document.currentScript||document.getElementsByTagName("script")[0],o=!1,r=document.createElement("link");if(r.relList&&"function"==typeof r.relList.supports&&r.relList.supports("preload")&&"as"in r)window.BOOMR.snippetMethod="p",r.href=window.BOOMR.url,r.rel="preload",r.as="script",r.addEventListener("load",e),r.addEventListener("error",function(){t(!0)}),setTimeout(function(){if(!o)t(!0)},3e3),BOOMR_lstart=(new Date).getTime(),i.parentNode.appendChild(r);else t(!1);if(window.addEventListener)window.addEventListener("load",a,!1);else if(window.attachEvent)window.attachEvent("onload",a)}}(),"".length>0)if(e&&"performance"in e&&e.performance&&"function"==typeof e.performance.setResourceTimingBufferSize)e.performance.setResourceTimingBufferSize();!function(){if(BOOMR=e.BOOMR||{},BOOMR.plugins=BOOMR.plugins||{},!BOOMR.plugins.AK){var n="false"=="true"?1:0,t="cookiepresent",a="eyd7g6aaiaaamjqacqdfqaaaabt3morr-f-1229ed2dd-clienttons-s.akamaihd.net",i="false"=="true"?2:1,o={"ak.v":"39","ak.cp":"1204614","ak.ai":parseInt("728289",10),"ak.ol":"0","ak.cr":3,"ak.ipv":6,"ak.proto":"h2","ak.rid":"14c88de6","ak.r":19138,"ak.a2":n,"ak.m":"dsca","ak.n":"essl","ak.bpcip":"2607:f378:40:6::","ak.cport":40630,"ak.gh":"23.60.168.62","ak.quicv":"","ak.tlsv":"tls1.3","ak.0rtt":"","ak.0rtt.ed":"","ak.csrc":"-","ak.acc":"","ak.t":"1739995697","ak.ak":"hOBiQwZUYzCg5VSAfCLimQ==A2RnzxK97WdYKq4JeKEG9WF2qjp+6K70EUC4y1bBwDS03JLOxaTVihEeJU1QlvH81tIjgeDgh3OXd08+mJo9Ouerq5pdvyvYfNx4bGZp9xZEa8lN5cax72DA/8c9tywqK2v7iyvVGWAyV2QGdG70QorLoKzxy3y6SJKgA0PnLk+FwH98AA8IAeketA33RJE5eZ2apJl3XsjB9lyN6q5xaWQIb1WPkdGPT05LO1ZPXAe07Fqnng646p/dcJqaE9bTyfCyR8znqvqWSB8w+rLkUjtDaAcm8V/Dt5JW+J7YBX2QWENUY8/sZ1r5Z6UinN710Jq1qINYN1j9ZHjdeM3wGG7YKIjVe71pfehz1srWTg2p5iJUyrfCS2pPJ1seF0F/VFZiC9DJuOts91zawYW6zxA6Z2JdSgqmmnGY8+XlksM=","ak.pv":"98","ak.dpoabenc":"","ak.tf":i};if(""!==t)o["ak.ruds"]=t;var r={i:!1,av:function(n){var t="http.initiator";if(n&&(!n[t]||"spa_hard"===n[t]))o["ak.feo"]=void 0!==e.aFeoApplied?1:0,BOOMR.addVar(o)},rv:function(){var e=["ak.bpcip","ak.cport","ak.cr","ak.csrc","ak.gh","ak.ipv","ak.m","ak.n","ak.ol","ak.proto","ak.quicv","ak.tlsv","ak.0rtt","ak.0rtt.ed","ak.r","ak.acc","ak.t","ak.tf"];BOOMR.removeVar(e)}};BOOMR.plugins.AK={akVars:o,akDNSPreFetchDomain:a,init:function(){if(!r.i){var e=BOOMR.subscribe;e("before_beacon",r.av,null,null),e("onbeacon",r.rv,null,null),r.i=!0}return this},is_complete:function(){return!0}}}}()}(window);</script></head>
|
|
<body class="research">
|
|
<article itemscope itemtype="http://schema.org/ScholarlyArticle">
|
|
<meta itemprop="datePublished" content="2009-07" />
|
|
<meta itemprop="image" content="cover.jpg" />
|
|
<header>
|
|
<div id="hLogo"><a class="navLogo" href="/policy/index.html">Social Security</a><a class="navSearch" href="https://search.ssa.gov/search?affiliate=ssa">SEARCH</a></div>
|
|
<div id="hRedBar">
|
|
<div id="hDocInfo">
|
|
<h1 itemprop="headline">Measurement Issues Associated with Using Survey Data Matched with Administrative Data from the Social Security Administration</h1>
|
|
<div id="hByline">by <span itemprop="author">Paul S. Davies and T. Lynn Fisher</span><br>Social Security Bulletin, <abbr title="Volume">Vol.</abbr> 69, <abbr title="Number">No.</abbr> 2, 2009 (released July 2009)</div>
|
|
</div>
|
|
</div>
|
|
</header>
|
|
<nav>
|
|
<div id="breadcrumbs" itemscope itemtype="http://schema.org/BreadcrumbList">You are here: <span itemprop="itemListElement" itemscope itemtype="http://schema.org/ListItem"><a href="/" itemprop="item"><span itemprop="name">Social Security Administration</span></a><meta itemprop="position" content="1" /></span> > <span itemprop="itemListElement" itemscope itemtype="http://schema.org/ListItem"><a href="/policy/index.html" itemprop="item"><span itemprop="name">Research, Statistics & Policy Analysis</span></a><meta itemprop="position" content="2" /></span> > <span itemprop="itemListElement" itemscope itemtype="http://schema.org/ListItem"><a href="/policy/docs/ssb/index.html" itemprop="item"><span itemprop="name">Social Security Bulletin</span></a><meta itemprop="position" content="3" /></span> > <span itemprop="itemListElement" itemscope itemtype="http://schema.org/ListItem"><a href="index.html" itemprop="item"><span itemprop="name"><abbr title="Volume">Vol.</abbr> 69, <abbr title="Number">No.</abbr> 2</span></a><meta itemprop="position" content="4" /></span></div>
|
|
<div id="rspaUtil"><ul><li id="mail"><a class="js-ga-event" href="#" rel="nofollow" data-event="outbound-link" data-event-action="click" data-event-label="email-this">Email</a></li><li id="print"><a href="#" rel="nofollow">Save/Print</a></li></ul></div>
|
|
</nav>
|
|
<div class="innards">
|
|
<div class="introBox">
|
|
<p id="synopsis" itemprop="description">Researchers using survey data matched with administrative data benefit from the rich demographic and economic detail available from survey data combined with detailed programmatic data from administrative records. The research benefits of using these matched data are too numerous to mention. But there are drawbacks as well, and those drawbacks have received less systematic attention from researchers. We focus on survey data matched with administrative data from the Social Security Administration and address the strengths and weaknesses of each in four specific areas: (1) program participation and benefits, (2) disability and health information, (3) earnings, and (4) deferred compensation. We discuss the implications of these strengths and weaknesses for decisions that researchers must make regarding the appropriate data source and definition for the concepts in question. From this discussion, some general conclusions are drawn about measurement issues associated with using matched survey and administrative data for research, policy evaluation, and statistics.</p>
|
|
<hr />
|
|
<div class="eightypercent">
|
|
<p>The authors are with the Division of Policy Evaluation, Office of Research, Evaluation, and Statistics, Office of Retirement and Disability Policy, Social Security Administration.</p>
|
|
<p><i>Acknowledgments</i>: The authors are grateful to Susan Grad, Carolyn Puckett, and Kalman Rupp for helpful comments and suggestions. A previous version of this article was presented at the 2008 Joint Statistical Meetings of the American Statistical Association, Government Statistics Section, Denver, <abbr title="Colorado">CO</abbr>.</p>
|
|
<p>Contents of this publication are <a href="/policy/accessibility.html">not copyrighted</a>; any items may be reprinted, but citation of the <i>Social Security Bulletin</i> as the source is requested. The findings and conclusions presented in the <i>Bulletin</i> are those of the authors and do not necessarily represent the views of the Social Security Administration.</p>
|
|
</div>
|
|
</div>
|
|
<h2>Introduction</h2>
|
|
<div class="abbrtable">
|
|
<table role="presentation">
|
|
<caption>Selected Abbreviations</caption>
|
|
<tr>
|
|
<td><abbr class="spell">CPS</abbr></td>
|
|
<td>Current Population Survey</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">DB</abbr></td>
|
|
<td>defined benefit</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">DC</abbr></td>
|
|
<td>defined contribution</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">DER</abbr></td>
|
|
<td>Detailed Earnings Record</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">DI</abbr></td>
|
|
<td>Disability Insurance</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">HRS</abbr></td>
|
|
<td>Health and Retirement Study</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">IRS</abbr></td>
|
|
<td>Internal Revenue Service</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">MBR</abbr></td>
|
|
<td>Master Beneficiary Record</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">NHANES</abbr></td>
|
|
<td>National Health and Nutrition Examination Survey</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">NHIS</abbr></td>
|
|
<td>National Health Interview Survey</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">NSCF</abbr></td>
|
|
<td>National Survey of <abbr class="spell">SSI</abbr> Children and Families</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">OASDI</abbr></td>
|
|
<td>Old-Age, Survivors, and Disability Insurance</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">PHUS</abbr></td>
|
|
<td>Payment History Update System</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr>SIPP</abbr></td>
|
|
<td>Survey of Income and Program Participation</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">SSA</abbr></td>
|
|
<td>Social Security Administration</td>
|
|
</tr>
|
|
<tr>
|
|
<td><abbr class="spell">SSI</abbr></td>
|
|
<td>Supplemental Security Income</td>
|
|
</tr>
|
|
</table>
|
|
</div>
|
|
<p>Researchers using survey data matched with administrative data benefit from the best of both worlds—the rich demographic and economic detail available from survey data combined with detailed programmatic data from administrative records. Indeed, researchers at the Social Security Administration (<abbr class="spell">SSA</abbr>) have been using matched survey and administrative data for years, addressing topics spanning policy evaluation, economic research, program statistics, and microsimulation modeling.</p>
|
|
<p>The original use of matched survey and administrative data was to assess the accuracy of the survey data and use that information to adjust for error in statistics produced from survey data. <abbr class="spell">SSA</abbr> and the Census Bureau have a history of matching Census surveys with Social Security administrative data and limited tax return information from the Internal Revenue Service (<abbr class="spell">IRS</abbr>). The earliest matches with the decennial censuses and periodically with the March Current Population Survey (<abbr class="spell">CPS</abbr>) from 1964 through 1972 were limited in scope and sample size because of computing constraints. The earliest matched file still being used is the 1973 <abbr class="spell">CPS</abbr>/<abbr class="spell">SSA</abbr>/<abbr class="spell">IRS</abbr> Exact Match Study, which greatly expanded the sample being matched to <abbr class="spell">SSA</abbr> and <abbr class="spell">IRS</abbr> data compared with previous matched data sets (Aziz, Kilss, and Scheuren 1978; Kilss and Scheuren 1978). This file provides researchers with rich survey data matched with longitudinal earnings histories that were not available elsewhere, and thus greatly expanded the potential scope of research on many topics in labor economics and public policy. Since the 1973 match, these data also have been used as inputs to Social Security's simulation models (Scheuren and Herriot 1975).</p>
|
|
<p>In response to limitations in the <abbr class="spell">CPS</abbr> with respect to analyzing government transfer programs, which required detailed data on income sources, program participation, and assets, the Income Survey Development Program was initiated in the mid-1970s (Ycas and Lininger 1981; Vaughan, Whiteman, and Lininger 1984). This program effectively served as the pilot study for the Census Bureau's Survey of Income and Program Participation (<abbr>SIPP</abbr>), for which the initial design called for matched administrative data on benefits and earnings from <abbr class="spell">SSA</abbr> (Lininger 1981). Pioneering work by Vaughan (1979) and others on errors in survey reports of program participation and type of beneficiary, some of which used the <abbr>SIPP</abbr> matched to <abbr class="spell">SSA</abbr> administrative data (Vaughan 1989), paved the way for a wide variety of uses of matched survey and administrative data by researchers at <abbr class="spell">SSA</abbr>.</p>
|
|
<p>Currently, researchers are using the <abbr>SIPP</abbr><sup><a href="#mn1" id="mt1">1</a></sup> (1984, <span class="nobr">1990–1993,</span> 1996, 2001, and 2004 panels) and the <abbr class="spell">CPS</abbr><sup><a href="#mn2" id="mt2">2</a></sup> (most years from the 1990s through the 2000s) matched to <abbr class="spell">SSA</abbr> administrative data and limited <abbr class="spell">IRS</abbr> earnings data. The matched data are accessed on a restricted basis subject to the terms of interagency agreements between the Census Bureau and <abbr class="spell">SSA</abbr> and of <abbr class="spell">IRS</abbr> laws and regulations. The use of matched administrative data as a tool to assess survey data is still a primary function, but other Census and <abbr class="spell">IRS</abbr>-approved uses of matched data have evolved. Other surveys that have been matched to <abbr class="spell">SSA</abbr> administrative data include the University of Michigan's Health and Retirement Study (<abbr class="spell">HRS</abbr>),<sup><a href="#mn3" id="mt3">3</a></sup> <abbr class="spell">SSA</abbr>'s National Survey of <abbr class="spell">SSI</abbr> Children and Families (<abbr class="spell">NSCF</abbr>),<sup><a href="#mn4" id="mt4">4</a></sup> and the National Center for Health Statistics' National Health Interview Survey (<abbr class="spell">NHIS</abbr>)<sup><a href="#mn5" id="mt5">5</a></sup> and National Health and Nutrition Examination Survey (<abbr class="spell">NHANES</abbr>).<sup><a href="#mn6" id="mt6">6</a></sup> <abbr class="spell">SSA</abbr>'s data are incomplete with respect to demographics and nonprogram oriented measures of income and wealth. The survey data on these elements supplement the administrative data, enabling the agency to produce a wide variety of research and statistical products about the Old-Age, Survivors, and Disability Insurance (<abbr class="spell">OASDI</abbr>, or Social Security) and Supplemental Security Income (<abbr class="spell">SSI</abbr>) programs. These products include detailed and complex microsimulation models that are used to assess the distributional implications of potential <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> policy changes, basic economic research on <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> beneficiaries, and statistics about beneficiaries and recipients of both programs.</p>
|
|
<p>The research benefits of using these matched data are too numerous to mention. But there are drawbacks as well, and those drawbacks have received less systematic attention from researchers. For example, in cases where disability diagnoses are available in both the survey and administrative data, which source is more accurate? In cases where program participation and benefit amounts are available in both the survey and administrative data, which source is correct? By and large, the answer to such questions is, "It depends." It depends on the research questions to be addressed. It depends on the data sources in question. It depends on the analytical techniques to be used. To complicate matters further, different administrative data sources can lead to different values for the same concept.</p>
|
|
<p>In this article, we do not attempt to provide definitive answers as to which sources are preferred in which situations. Rather, we attempt to draw together the available evidence on a number of important areas in which researchers using matched survey and administrative data must decide on the appropriate data source and definition for the concept in question. Specifically, in the next four sections of the article we examine and discuss the available evidence in the following areas.</p>
|
|
<ul>
|
|
<li><abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> participation and benefits</li>
|
|
<li>Disability diagnosis, health, and functional limitations</li>
|
|
<li>Earnings</li>
|
|
<li>Deferred compensation</li>
|
|
</ul>
|
|
<p>Some concluding observations are then offered on these measurement issues and the importance of matched survey and administrative data for research, policy evaluation, and program statistics. Finally, we highlight several areas for future research.</p>
|
|
<h2><abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> Participation and Benefits</h2>
|
|
<p>The most basic area of comparison between survey and administrative data is program participation and benefit amounts. Several <abbr class="spell">SSA</abbr> researchers have addressed this issue using data from the <abbr>SIPP</abbr> and <abbr class="spell">CPS</abbr> matched with <abbr class="spell">SSA</abbr> administrative data on the receipt and amount of <abbr class="spell">OASDI</abbr> benefits and <abbr class="spell">SSI</abbr> payments. Survey data may differ from administrative records for three main reasons: (1) survey error, (2) administrative record error, or (3) error in matching survey and administrative records (Huynh, Rupp, and Sears 2002). Although <abbr class="spell">SSA</abbr> records on program participation and benefit amounts are generally regarded to be more reliable than survey reports, this is not always the case. Before the availability of the Payment History Update System (<abbr class="spell">PHUS</abbr>), the administrative records for <abbr class="spell">OASDI</abbr> came only from the Master Beneficiary Record (<abbr class="spell">MBR</abbr>), which reflected program eligibility, as opposed to the actual benefit amount that was paid in a given month.<sup><a href="#mn7" id="mt7">7</a></sup> Since 2003, however, the match has included <abbr class="spell">PHUS</abbr> data with actual payment amounts from 1984 to the present, which is thought to be more consistent with the benefit amount that would be reported by survey respondents.<sup><a href="#mn8" id="mt8">8</a></sup> The Supplemental Security Record, which provides data on <abbr class="spell">SSI</abbr> applicants and recipients, has always captured data on both program eligibility and actual payment amounts.</p>
|
|
<p>Huynh, Rupp, and Sears (2002) assessed discrepancies in reports of benefit receipt and benefit amounts between <abbr class="spell">SSA</abbr>'s administrative records (Master Beneficiary Record and Supplemental Security Record) and the 1993 and 1996 panels of the <abbr>SIPP</abbr>.<sup><a href="#mn9" id="mt9">9</a></sup> They found that there is confusion among survey respondents as to whether an <abbr class="spell">OASDI</abbr> benefit or <abbr class="spell">SSI</abbr> payment was received. Table 1 shows that for the sample months analyzed by those authors, a nontrivial proportion of <abbr class="spell">SSI</abbr> recipient survey respondents (receiving <abbr class="spell">SSI</abbr> only or concurrent with <abbr class="spell">OASDI</abbr>) reported receiving <abbr class="spell">OASDI</abbr> only; respondents misreported receiving <abbr class="spell">OASDI</abbr> as <abbr class="spell">SSI</abbr>, but much less frequently. The authors offered a number of explanations for this pattern.</p>
|
|
<ul>
|
|
<li>Both <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> benefits are administered by <abbr class="spell">SSA</abbr>.</li>
|
|
<li>The <abbr class="spell">OASDI</abbr> program has greater visibility.</li>
|
|
<li>Stigma may be attached to the receipt of <abbr class="spell">SSI</abbr> payments.</li>
|
|
<li>The receipt of <abbr class="spell">SSI</abbr> for a few months often precedes the receipt of Disability Insurance (<abbr class="spell">DI</abbr>) for working-age individuals with disabilities.</li>
|
|
</ul>
|
|
<div class="table" id="table1">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 1. </span><abbr>SIPP</abbr> report of <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> benefit receipt, by <abbr class="spell">SSA</abbr> administrative record of benefit receipt status and observation period for adults with matched <abbr>SIPP</abbr> records (in percent)</caption>
|
|
<colgroup span="1" style="width:18em"></colgroup>
|
|
<colgroup span="4" style="width:6em"></colgroup>
|
|
<colgroup span="1" style="width:6em"></colgroup>
|
|
<colgroup class="shaded" span="1" style="width:6em"></colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" scope="colgroup">Administrative record receipt status and observation period</th>
|
|
<th class="spanner" colspan="4" scope="colgroup"><abbr>SIPP</abbr> report of receipt </th>
|
|
<th rowspan="2" scope="colgroup">Total </th>
|
|
<th rowspan="2" scope="colgroup">N</th>
|
|
</tr>
|
|
<tr>
|
|
<th scope="col">Both</th>
|
|
<th scope="col">Neither</th>
|
|
<th scope="col"><abbr class="spell">OASDI</abbr> only</th>
|
|
<th scope="col"><abbr class="spell">SSI</abbr> only</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">Both <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr></th>
|
|
<td colspan="5"></td>
|
|
<td></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1993 (January)</th>
|
|
<td>76.08</td>
|
|
<td>3.49</td>
|
|
<td>14.52</td>
|
|
<td>5.91</td>
|
|
<td>100.00</td>
|
|
<td>372</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1995 (August)</th>
|
|
<td>80.75</td>
|
|
<td>2.48</td>
|
|
<td>10.87</td>
|
|
<td>5.90</td>
|
|
<td>100.00</td>
|
|
<td>322</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1996 (March)</th>
|
|
<td>74.71</td>
|
|
<td>4.89</td>
|
|
<td>12.40</td>
|
|
<td>7.99</td>
|
|
<td>99.99</td>
|
|
<td>613</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1998 (October)</th>
|
|
<td>80.06</td>
|
|
<td>3.81</td>
|
|
<td>12.02</td>
|
|
<td>4.11</td>
|
|
<td>100.00</td>
|
|
<td>341</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="rowgroup">Neither <abbr class="spell">OASDI</abbr> nor <abbr class="spell">SSI</abbr></th>
|
|
<td colspan="5"></td>
|
|
<td></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1993 (January)</th>
|
|
<td>0.06</td>
|
|
<td>98.32</td>
|
|
<td>1.25</td>
|
|
<td>0.37</td>
|
|
<td>100.00</td>
|
|
<td>25,704</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1995 (August)</th>
|
|
<td>0.07</td>
|
|
<td>97.99</td>
|
|
<td>1.44</td>
|
|
<td>0.50</td>
|
|
<td>100.00</td>
|
|
<td>22,436</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1996 (March)</th>
|
|
<td>0.04</td>
|
|
<td>98.81</td>
|
|
<td>0.97</td>
|
|
<td>0.17</td>
|
|
<td>99.99</td>
|
|
<td>33,545</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1998 (October)</th>
|
|
<td>0.05</td>
|
|
<td>98.66</td>
|
|
<td>1.07</td>
|
|
<td>0.23</td>
|
|
<td>100.01</td>
|
|
<td>16,677</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="rowgroup"><abbr class="spell">OASDI</abbr> only</th>
|
|
<td colspan="5"></td>
|
|
<td></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1993 (January)</th>
|
|
<td>0.30</td>
|
|
<td>3.38</td>
|
|
<td>95.95</td>
|
|
<td>0.38</td>
|
|
<td>100.00</td>
|
|
<td>6,068</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1995 (August)</th>
|
|
<td>0.37</td>
|
|
<td>4.35</td>
|
|
<td>94.73</td>
|
|
<td>0.55</td>
|
|
<td>100.00</td>
|
|
<td>5,632</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1996 (March)</th>
|
|
<td>0.41</td>
|
|
<td>4.31</td>
|
|
<td>94.46</td>
|
|
<td>0.82</td>
|
|
<td>100.00</td>
|
|
<td>7,886</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1998 (October)</th>
|
|
<td>0.65</td>
|
|
<td>3.77</td>
|
|
<td>94.78</td>
|
|
<td>0.81</td>
|
|
<td>100.01</td>
|
|
<td>4,328</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="rowgroup"><abbr class="spell">SSI</abbr> only</th>
|
|
<td colspan="5"></td>
|
|
<td></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1993 (January)</th>
|
|
<td>6.01</td>
|
|
<td>6.56</td>
|
|
<td>8.74</td>
|
|
<td>78.69</td>
|
|
<td>100.00</td>
|
|
<td>366</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1995 (August)</th>
|
|
<td>3.60</td>
|
|
<td>9.14</td>
|
|
<td>6.09</td>
|
|
<td>81.16</td>
|
|
<td>99.99</td>
|
|
<td>361</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1996 (March)</th>
|
|
<td>4.81</td>
|
|
<td>8.94</td>
|
|
<td>7.70</td>
|
|
<td>78.54</td>
|
|
<td>99.99</td>
|
|
<td>727</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">1998 (October)</th>
|
|
<td>3.02</td>
|
|
<td>9.32</td>
|
|
<td>7.81</td>
|
|
<td>79.85</td>
|
|
<td>100.00</td>
|
|
<td>397</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="onlyNote" colspan="7">SOURCE: Huynh, Rupp, and Sears (2002, Table 2). Data are tabulated from the 1993 and 1996 panels of the <abbr>SIPP</abbr> matched to <abbr class="spell">SSA</abbr>'s Master Beneficiary Record and Supplemental Security Record.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<p>Huynh, Rupp, and Sears (2002) also found that accuracy of <abbr class="spell">SSI</abbr> reports improved between their observation points within the 1993 and 1996 <abbr>SIPP</abbr> panels. In addition they evaluated the discrepancies between reported <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> benefits and administrative amounts. The authors confirmed that after wave 1 of the 1993 <abbr>SIPP</abbr>, respondents were reporting their <abbr class="spell">OASDI</abbr> benefits net of the Medicare Part B premium, consistent with the revised questionnaire wording. They noted that use of these reported benefit amounts without adjusting for the Part B premium could substantially bias estimates of total income and poverty status. Also, they concluded that self-reported <abbr class="spell">SSI</abbr> payments in the <abbr>SIPP</abbr> reflect the sum of federal and federally administered state <abbr class="spell">SSI</abbr> payments, which are provided to recipients in a single payment (check or direct deposit). In addition, the authors found that reporting errors for <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> differed dramatically by imputation status, and that errors may be systematically related to sample attrition and interview status. Finally Huynh, Rupp, and Sears (2002) found evidence of selectivity with respect to the survey respondents who were unable to be matched to administrative records.</p>
|
|
<p>Koenig (2003) followed a framework similar to that of Huynh, Rupp, and Sears (2002) by assessing the accuracy of self-reported <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> data in the 1996 <abbr>SIPP</abbr> and the March 1997 Annual Demographic Supplement to the <abbr class="spell">CPS</abbr>. She compared the accuracy of reported <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> receipt and benefit amounts in the two surveys relative to matched <abbr class="spell">SSA</abbr> administrative records and assessed the effect on poverty estimates when administrative benefit information is used with the survey data. Koenig (2003) found that although both surveys reflected aggregate benefits well, the <abbr>SIPP</abbr> overestimated the percentages of individuals who received <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr>, and the <abbr class="spell">CPS</abbr> underestimated them. The <abbr>SIPP</abbr> was better able than the <abbr class="spell">CPS</abbr> to identify both <abbr class="spell">OASDI</abbr> beneficiaries (99 percent versus 95 percent) and <abbr class="spell">SSI</abbr> recipients (93 percent versus 69 percent). For the sample of respondents receiving <abbr class="spell">OASDI</abbr> and/or <abbr class="spell">SSI</abbr> in both the survey and administrative records, the <abbr>SIPP</abbr>-reported benefit amount was within $100 of the benefit amount in the administrative records twice as often as the <abbr class="spell">CPS</abbr>-reported benefit amount for <abbr class="spell">OASDI</abbr> (47 percent versus 24 percent), but slightly less frequently than the <abbr class="spell">CPS</abbr>-reported benefit amount for <abbr class="spell">SSI</abbr> (47 percent versus 55 percent). The impact on total income and poverty estimates of using administrative data in place of self-reported survey data was largest for the group with imputed records (Table 2). The overall poverty estimates were slightly lower in both surveys when administrative data were used in place of self-reported survey data; respondents in the <abbr class="spell">CPS</abbr> were more likely to exhibit a change in poverty status because of the use of administrative data.</p>
|
|
<div class="table" id="table2">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 2. </span>Percentage distribution of persons aged 65 or older with poverty status change after substituting self-reported survey data with administrative data, by imputation status </caption>
|
|
<colgroup span="1" style="width:18em"></colgroup>
|
|
<colgroup span="2" style="width:6em"></colgroup>
|
|
<colgroup span="2" style="width:6em"></colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" scope="colgroup">Poverty status </th>
|
|
<th class="spanner" colspan="2" scope="colgroup"><abbr class="spell">CPS</abbr></th>
|
|
<th class="spanner" colspan="2" scope="colgroup"><abbr>SIPP</abbr></th>
|
|
</tr>
|
|
<tr>
|
|
<th scope="col">Imputed benefits</th>
|
|
<th scope="col">No imputed benefits</th>
|
|
<th scope="col">Imputed benefits</th>
|
|
<th scope="col">No imputed benefits</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" scope="row">Poverty status does not change</th>
|
|
<td>89.9</td>
|
|
<td>95.8</td>
|
|
<td>95.7</td>
|
|
<td>98.1</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Change from in poverty to not in poverty</th>
|
|
<td>5.7</td>
|
|
<td>2.2</td>
|
|
<td>2.5</td>
|
|
<td>1.1</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Change from not in poverty to in poverty</th>
|
|
<td>4.4</td>
|
|
<td>2.0</td>
|
|
<td>1.8</td>
|
|
<td>0.8</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">Total</th>
|
|
<td>100.0</td>
|
|
<td>100.0</td>
|
|
<td>100.0</td>
|
|
<td>100.0</td>
|
|
</tr>
|
|
<tr class="shaded">
|
|
<th class="stub0" scope="row">Unweighted N</th>
|
|
<td>2,097</td>
|
|
<td>8,956</td>
|
|
<td>2,322</td>
|
|
<td>6,513</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="onlyNote" colspan="5">SOURCE: Koenig (2003, Table 9). Data are tabulated from the 1996 <abbr>SIPP</abbr> and March 1997 <abbr class="spell">CPS</abbr> matched to the <abbr class="spell">SSA</abbr>'s Master Beneficiary Record and Supplemental Security Record.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<p>Nicholas and Wiseman (2009) developed a detailed method for replacing self-reported survey data from the March 2003 Annual Social and Economic Supplement to the <abbr class="spell">CPS</abbr> with administrative data on <abbr class="spell">SSI</abbr> payments, <abbr class="spell">OASDI</abbr> benefits, and earnings. The authors also implemented a propensity scoring system to reweight <abbr class="spell">CPS</abbr> families in the matched <abbr class="spell">CPS</abbr>/<abbr class="spell">SSA</abbr> sample to reflect the <abbr>U.S.</abbr> population as a whole. Using a "high" and a "low" version of their matching and data replacement system, the authors then examined the implications of using the matched administrative data for measuring poverty among the general population and among <abbr class="spell">SSI</abbr> recipients. Their findings for absolute poverty were quite dramatic, especially among <abbr class="spell">SSI</abbr> recipients, as illustrated in Table 3. Based on public-use <abbr class="spell">CPS</abbr> data, 44.3 percent of all <abbr class="spell">SSI</abbr> recipients were in poverty in 2002. Depending on the exact definitions used, the poverty rate was reduced from 44.3 percent to between 38.0 percent and 40.9 percent when <abbr class="spell">SSA</abbr> administrative data on benefits and earnings were used in place of <abbr class="spell">CPS</abbr> self-reported data. The effects were the strongest for elderly <abbr class="spell">SSI</abbr> recipients, whose "official" poverty rate derived from public-use <abbr class="spell">CPS</abbr> data fell from 48.0 percent to between 38.6 percent and 40.6 percent based on <abbr class="spell">CPS</abbr>/<abbr class="spell">SSA</abbr> matched data. The effects were much more modest for the <abbr>U.S.</abbr> population in general, which confirms the authors' finding that <abbr class="spell">SSI</abbr> participation and benefits were substantially underreported in the <abbr class="spell">CPS</abbr> relative to <abbr class="spell">SSA</abbr> administrative data.</p>
|
|
<div class="table" id="table3">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 3. </span>Poverty rates for the <abbr>U.S.</abbr> population and <abbr class="spell">SSI</abbr> recipients, by age group, source of data, and income-adjustment method, 2002 (in percent)</caption>
|
|
<colgroup span="1" style="width:18em"></colgroup>
|
|
<colgroup span="5" style="width:6em"></colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" id="c1">Population and age group</th>
|
|
<th rowspan="2" id="c2">Public-use <abbr class="spell">CPS</abbr> data</th>
|
|
<th class="spanner" colspan="2" id="c3"><abbr class="spell">CPS</abbr> income adjusted based on <abbr class="spell">SSA</abbr> data; matched plus unmatched individuals</th>
|
|
<th class="spanner" colspan="2" id="c4"><abbr class="spell">CPS</abbr> income adjusted based on <abbr class="spell">SSA</abbr> data; matched individuals only</th>
|
|
</tr>
|
|
<tr>
|
|
<th id="c5" headers="c3">"Lower" income adjustment</th>
|
|
<th id="c6" headers="c3">"Higher" income adjustment</th>
|
|
<th id="c7" headers="c4">"Lower" income adjustment</th>
|
|
<th id="c8" headers="c4">"Higher" income adjustment</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" id="r1" headers="c1"><abbr>U.S.</abbr> population</th>
|
|
<td colspan="5"></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1 nobr" id="r2" headers="r1 c1">0–17</th>
|
|
<td headers="r1 r2 c2">16.7</td>
|
|
<td headers="r1 r2 c3 c5">16.4</td>
|
|
<td headers="r1 r2 c3 c6">13.3</td>
|
|
<td headers="r1 r2 c4 c7">16.3</td>
|
|
<td headers="r1 r2 c4 c8">13.0</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1 nobr" id="r3" headers="r1 c1">18–64</th>
|
|
<td headers="r1 r3 c2">10.6</td>
|
|
<td headers="r1 r3 c3 c5">10.5</td>
|
|
<td headers="r1 r3 c3 c6">8.4</td>
|
|
<td headers="r1 r3 c4 c7">10.5</td>
|
|
<td headers="r1 r3 c4 c8">7.9</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" id="r4" headers="r1 c1">65 or older</th>
|
|
<td headers="r1 r4 c2">10.4</td>
|
|
<td headers="r1 r4 c3 c5">9.1</td>
|
|
<td headers="r1 r4 c3 c6">8.9</td>
|
|
<td headers="r1 r4 c4 c7">8.4</td>
|
|
<td headers="r1 r4 c4 c8">8.1</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub2" id="r5" headers="r1 r4 c1">Total</th>
|
|
<td headers="r1 r4 r5 c2">12.1</td>
|
|
<td headers="r1 r4 r5 c3 c5">11.8</td>
|
|
<td headers="r1 r4 r5 c3 c6">9.7</td>
|
|
<td headers="r1 r4 r5 c4 c7">11.8</td>
|
|
<td headers="r1 r4 r5 c4 c8">9.3</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" id="r6" headers="c1"><abbr class="spell">SSI</abbr> recipients</th>
|
|
<td colspan="5"></td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1 nobr" id="r7" headers="r6 c1">0–17</th>
|
|
<td headers="r6 r7 c2">36.2</td>
|
|
<td headers="r6 r7 c3 c5">26.5</td>
|
|
<td headers="r6 r7 c3 c6">21.8</td>
|
|
<td headers="r6 r7 c4 c7">26.5</td>
|
|
<td headers="r6 r7 c4 c8">21.8</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1 nobr" id="r8" headers="r6 c1">18–64</th>
|
|
<td headers="r6 r8 c2">43.9</td>
|
|
<td headers="r6 r8 c3 c5">42.3</td>
|
|
<td headers="r6 r8 c3 c6">40.9</td>
|
|
<td headers="r6 r8 c4 c7">44.6</td>
|
|
<td headers="r6 r8 c4 c8">43.0</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" id="r9" headers="r6 c1">65 or older</th>
|
|
<td headers="r6 r9 c2">48.0</td>
|
|
<td headers="r6 r9 c3 c5">40.6</td>
|
|
<td headers="r6 r9 c3 c6">39.4</td>
|
|
<td headers="r6 r9 c4 c7">39.9</td>
|
|
<td headers="r6 r9 c4 c8">38.6</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub2" id="r10" headers="r6 r9 c1">Total</th>
|
|
<td headers="r6 r9 r10 c2">44.3</td>
|
|
<td headers="r6 r9 r10 c3 c5">39.8</td>
|
|
<td headers="r6 r9 r10 c3 c6">38.0</td>
|
|
<td headers="r6 r9 r10 c4 c7">40.9</td>
|
|
<td headers="r6 r9 r10 c4 c8">39.0</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="onlyNote" colspan="6">SOURCE: Derived by authors from Nicholas and Wiseman (2009, Table 7). Data are from the 2003 <abbr class="spell">CPS</abbr> Annual Social and Economic Supplement and matched <abbr class="spell">SSA</abbr> administrative records.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<p>Huynh, Rupp, and Sears (2002) and Koenig (2003), among others, questioned the extent to which selectivity in the ability to match administrative records to <abbr>SIPP</abbr> and <abbr class="spell">CPS</abbr> survey records resulted in a match bias. Attrition bias in the <abbr>SIPP</abbr> was another prominent concern. To address these issues, <abbr class="spell">SSA</abbr> awarded a contract to Mathematica Policy Research, <abbr title="Incorporated">Inc.</abbr> to determine the extent to which attrition and match selectivity influence estimates of income receipt and amounts. After calibrating their sample from the 2001 <abbr>SIPP</abbr> to Census demographic controls, Czajka, Mabli, and Cody (2008) found little evidence of bias in estimates of a wide range of characteristics. They also found that although the proportion of <abbr>SIPP</abbr> respondents who could be matched with administrative records dropped substantially between the 1996 and 2001 panels of the <abbr>SIPP</abbr>, bias in the matched sample did not appear to have increased. Their more limited evaluation of match bias in the <abbr class="spell">CPS</abbr> focused on retired workers, with results similar to those for the <abbr>SIPP</abbr>. Personal, family, and household demographics among the matched sample mirrored the full <abbr class="spell">CPS</abbr> sample, although matched cases had slightly more income and were slightly less reliant on Social Security benefits.</p>
|
|
<p>Fisher (2005, 2008) examined the impact of survey choice and the use of administrative data in place of survey data on estimates of the importance of Social Security relative to total income for the elderly. In particular, she examined the proportion of the elderly receiving all of their income from Social Security. Using the 1996 <abbr>SIPP</abbr> and the March 1997 <abbr class="spell">CPS</abbr>, Fisher (2005) estimated that in 1996, 19.4 percent of the elderly in the <abbr class="spell">CPS</abbr> and 9.4 percent of the elderly in the <abbr>SIPP</abbr> received all of their income from Social Security. The author found that among those receiving all income from Social Security benefits, either in reported or administrative data, the <abbr>SIPP</abbr> had a lower rate of beneficiary misclassification than the <abbr class="spell">CPS</abbr>, as shown in Table 4. In particular, respondents in the <abbr class="spell">CPS</abbr> were more likely to omit <abbr class="spell">SSI</abbr> and were also five times as likely to report having no income at all, despite being <abbr class="spell">OASDI</abbr> (Social Security) beneficiaries. The substitution of administrative data for self-reported survey data had a negligible effect on the estimates, however, because receipt of sources of income other than Social Security is what is essentially being measured.</p>
|
|
<div class="table" id="table4">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 4. </span>Misclassification of beneficiary status of person observations 65 or older with an administrative record match</caption>
|
|
<colgroup span="1" style="width:32em"></colgroup>
|
|
<colgroup span="2" style="width:5em"></colgroup>
|
|
<colgroup span="2" style="width:5em"></colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" scope="colgroup">Misclassification status</th>
|
|
<th class="spanner" colspan="2" scope="colgroup"><abbr>SIPP</abbr></th>
|
|
<th class="spanner" colspan="2" scope="colgroup"><abbr class="spell">CPS</abbr></th>
|
|
</tr>
|
|
<tr>
|
|
<th scope="col">Number</th>
|
|
<th scope="col">Percent</th>
|
|
<th scope="col">Number</th>
|
|
<th scope="col">Percent</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" scope="row">Persons showing all income from <abbr class="spell">OASDI</abbr> benefits</th>
|
|
<td>902</td>
|
|
<td>100</td>
|
|
<td>2,169</td>
|
|
<td>100.0</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="row">No beneficiary misclassification</th>
|
|
<td>827</td>
|
|
<td>91.7</td>
|
|
<td>1,813</td>
|
|
<td>83.6</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="row">100 percent reliance on self-report, but not on administrative records</th>
|
|
<td>52</td>
|
|
<td>5.8</td>
|
|
<td>196</td>
|
|
<td>9.0</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Self-report omitted <abbr class="spell">SSI</abbr> income</th>
|
|
<td>29</td>
|
|
<td>3.2</td>
|
|
<td>138</td>
|
|
<td>6.4</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Not an <abbr class="spell">OASDI</abbr> beneficiary</th>
|
|
<td>38</td>
|
|
<td>4.2</td>
|
|
<td>106</td>
|
|
<td>4.9</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Both self-report omitted <abbr class="spell">SSI</abbr> income and not an <abbr class="spell">OASDI</abbr> beneficiary</th>
|
|
<td>15</td>
|
|
<td>1.7</td>
|
|
<td>48</td>
|
|
<td>2.2</td>
|
|
</tr>
|
|
<tr class="topPad1">
|
|
<th class="stub0" scope="row">100 percent reliance on administrative records, but not on self-report</th>
|
|
<td>23</td>
|
|
<td>2.5</td>
|
|
<td>160</td>
|
|
<td>7.4</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Self-report included <abbr class="spell">SSI</abbr> income not in administrative records</th>
|
|
<td>15</td>
|
|
<td>1.7</td>
|
|
<td>41</td>
|
|
<td>1.9</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row"><abbr class="spell">OASDI</abbr> beneficiary in administrative records, but not in self-report</th>
|
|
<td>11</td>
|
|
<td>1.2</td>
|
|
<td>128</td>
|
|
<td>5.9</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="onlyNote" colspan="5">SOURCE: Fisher (2005, Table 5). Data are tabulated from the 1996 <abbr>SIPP</abbr> and March 1997 <abbr class="spell">CPS</abbr> matched to the <abbr class="spell">SSA</abbr>'s Payment History Update System and Supplemental Security Record.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<p>Fisher (2008) found that the large differences in estimates of the elderly receiving all of their income from Social Security in the <abbr class="spell">CPS</abbr> and <abbr>SIPP</abbr> for 1996 is most likely the result of underreporting the receipt of asset income in the <abbr class="spell">CPS</abbr>, although most sources of income are significantly more likely to be reported in the <abbr>SIPP</abbr> than the <abbr class="spell">CPS</abbr>. To determine the extent to which these sources of income are underreported in the <abbr class="spell">CPS</abbr>, particularly asset income and pensions, <abbr class="spell">SSA</abbr>, the Census Bureau, and the <abbr class="spell">IRS</abbr> entered into an agreement to match a limited set of variables from individual income tax returns (Form 1040) and informational returns (Form 1099-R) to the March 2007 <abbr class="spell">CPS</abbr>. Research using these data will begin soon.</p>
|
|
<p>These articles and others in this same line of research suggest that self-reported data in the <abbr class="spell">CPS</abbr> slightly underreport <abbr class="spell">OASDI</abbr> receipt and significantly underreport <abbr class="spell">SSI</abbr> receipt. Self-reported data in the <abbr>SIPP</abbr> slightly overreport receipt of <abbr class="spell">OASDI</abbr>; however, the picture is more complicated for receipt of <abbr class="spell">SSI</abbr> depending on the year of analysis and whether the data are analyzed from a monthly or annual perspective. Estimates from both surveys indicate some confusion among respondents between the two sources of income. Misreporting of income is unlikely to be limited to the <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> programs; other sources of income should be assessed in a similar fashion. Confusion between <abbr class="spell">OASDI</abbr> benefits and <abbr class="spell">SSI</abbr> payments, which are administered by <abbr class="spell">SSA</abbr>, is probably not unique; reported data on other programs that are also administered by the same agency, such as Medicare and Medicaid, may also benefit from examining administrative data. Additional research in these areas should lead to improvements in survey measurement of program participation and benefits, which in turn should lead to more accurate estimates of total income, poverty status, and well-being.</p>
|
|
<h2>Disability Diagnosis, Health, and Functional Limitations</h2>
|
|
<p>Although similar labels often are applied to the disability and health information available from surveys and administrative data sources, the concepts being measured may be fundamentally different. The <abbr>SIPP</abbr>, <abbr class="spell">HRS</abbr>, <abbr class="spell">NSCF</abbr>, <abbr class="spell">NHIS</abbr>, and <abbr class="spell">NHANES</abbr> contain detailed data on disabling conditions, health status, and functional impairments. These data reflect the respondent's (or the respondent's proxy) subjective perceptions of his or her health and disability status at the time the survey was administered.<sup><a href="#mn10" id="mt10">10</a></sup> The data reported by the respondent typically are recoded in various ways by the survey administrator before being released to researchers. Social Security administrative records contain data on primary and secondary impairments for disability beneficiaries, which reflect the medical conditions considered in the medical decision about disability or blindness (initial application or continuing disability review). Those administrative records do not contain data on the general health status of disability beneficiaries, their functional limitations, or the severity of their disabling condition(s). For denied disability applicants, <abbr class="spell">SSA</abbr>'s administrative records systems generally do not contain impairment codes. Moreover, <abbr class="spell">SSA</abbr> disability data document the condition that supports the medical decision regarding eligibility for disability benefits, which is not necessarily the same as the condition that is most disabling from the individual's perspective.</p>
|
|
<p>Given this limited background information, consider the data in Table 5 on the disabling conditions of children receiving <abbr class="spell">SSI</abbr>, which are derived from the <abbr class="spell">NSCF</abbr> and <abbr class="spell">SSA</abbr> administrative records and are reproduced from Rupp and others (2005/2006). The distribution of disability types (left side of table) differs greatly between <abbr class="spell">NSCF</abbr> data reported by the respondent and <abbr class="spell">SSA</abbr> administrative data. Nearly 44 percent of <abbr class="spell">NSCF</abbr> respondents report a physical disability, compared with 25.4 percent in <abbr class="spell">SSA</abbr> administrative data. Only 8 percent of <abbr class="spell">NSCF</abbr> respondents report mental retardation, compared with 32.5 percent in <abbr class="spell">SSA</abbr> administrative data. However, if individuals identified by <abbr class="spell">SSA</abbr> administrative data as being mentally retarded are removed from the sample, the distribution of disabilities in the <abbr class="spell">NSCF</abbr> more closely matches the distribution of disabilities in <abbr class="spell">SSA</abbr> administrative data (right side of table). This supports the hypothesis that some respondents are reluctant to report that their child is mentally retarded or that they did not consider mental retardation to be a health condition.</p>
|
|
<div class="table" id="table5">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 5. </span>Type of disability among children receiving <abbr class="spell">SSI</abbr>, by source of disability data (in percent)</caption>
|
|
<colgroup span="1" style="width:12em"></colgroup>
|
|
<colgroup span="2" style="width:8em"></colgroup>
|
|
<colgroup span="2" style="width:10em"></colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" scope="colgroup">Type of disability</th>
|
|
<th class="spanner" colspan="2" scope="colgroup">All children receiving <abbr class="spell">SSI</abbr></th>
|
|
<th class="spanner" colspan="2" scope="colgroup">Children receiving <abbr class="spell">SSI</abbr> who are not identified<br>as mentally retarded in <abbr class="spell">SSA</abbr> records</th>
|
|
</tr>
|
|
<tr>
|
|
<th scope="col"><abbr class="spell">NSCF</abbr> <sup>a</sup></th>
|
|
<th scope="col"><abbr class="spell">SSA</abbr> records</th>
|
|
<th scope="col"><abbr class="spell">NSCF</abbr> <sup>a</sup></th>
|
|
<th scope="col"><abbr class="spell">SSA</abbr> records</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">Physical</th>
|
|
<td>43.5</td>
|
|
<td>25.4</td>
|
|
<td>52.0</td>
|
|
<td>37.7</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">Mental</th>
|
|
<td>50.4</td>
|
|
<td>61.8</td>
|
|
<td>42.3</td>
|
|
<td>43.3</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">Mental retardation</th>
|
|
<td>7.9</td>
|
|
<td>32.5</td>
|
|
<td>3.9</td>
|
|
<td>. . .</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub1" scope="row">Other mental</th>
|
|
<td>44.2</td>
|
|
<td>29.2</td>
|
|
<td>39.2</td>
|
|
<td>43.3</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">Other</th>
|
|
<td>14.8</td>
|
|
<td>7.7</td>
|
|
<td>14.3</td>
|
|
<td>11.5</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">None reported</th>
|
|
<td>0.4</td>
|
|
<td>. . .</td>
|
|
<td>0.3</td>
|
|
<td>. . .</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="rowgroup">Missing</th>
|
|
<td>2.8</td>
|
|
<td>5.1</td>
|
|
<td>2.6</td>
|
|
<td>7.6</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="firstNote" colspan="5">SOURCE: Rupp and others (2006, Table 3 and note 15) and unpublished tabulations of <abbr class="spell">NSCF</abbr> data and <abbr class="spell">SSA</abbr> administrative data.</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="note" colspan="5">NOTES: <abbr class="spell">NSCF</abbr> interviews were conducted from July 2001 through June 2002.</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="note" colspan="5">. . . = not applicable.</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="lastNote" colspan="5">a. Up to three health problems or conditions were coded in the <abbr class="spell">NSCF</abbr>. Because sample members can have more than one health problem or condition, the disability categories and subcategories are not mutually exclusive. Therefore, the percentages do not add to 100.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<p>We conclude that the choice to use self-reported survey data on disabilities and health conditions or administrative disability data should depend on the specific application of the data. For studies that seek to understand the relationship between individual behavior and disabilities, self-reported survey data on disabilities may be more appropriate, whereas administrative disability data may be the better choice for programmatic studies or tabulations of disability beneficiaries. Both survey and administrative measures of disability and health are very complex. Survey data reflect the respondent's perception of his or her disability status and also may be influenced by proxy respondents, coding choices by survey administrators, social norms, and the quality of training provided to survey interviewers. Administrative data tend to be driven by programmatic requirements and complexities. Self-reported disability measures have been criticized in the literature as subjective, inconsistent, and endogenous (Sickles and Taubman 1997; Bound and Waidmann 1992; Kreider 1999). However, it is important to note that survey respondents may have much more detailed information about their own health and functional status than other more objective sources based on limited information. In addition, research has shown that self-reported disability measures at the time of the survey interview are highly correlated with long-term measures of mortality and disability program participation, even after controlling for a variety of demographic and economic characteristics (Rupp and Davies 2004).</p>
|
|
<h2>Earnings</h2>
|
|
<p>The earliest benefit of matching administrative earnings records with survey data was to expand the scope and quality of research in labor economics and public policy. Earnings records derived from <abbr class="spell">IRS</abbr> W-2 Forms also are used to evaluate the accuracy of survey data, particularly in the <abbr>SIPP</abbr>. Bridges, Del Bene, and Leonesio (2003) used the Detailed Earnings Record (<abbr class="spell">DER</abbr>), which is an extract of <abbr class="spell">SSA</abbr>'s Master Earnings File, matched to the 1992 and 1993 panels of the <abbr>SIPP</abbr> to study the accuracy of calendar year 1993 wage and self-employment income in the <abbr>SIPP</abbr>. Gottschalk and Huynh (2005) used the <abbr class="spell">DER</abbr> matched to the 1996 <abbr>SIPP</abbr> to determine the effect of measurement error on the mean and dispersion of the distributions of earnings for people of different ages and on the correlation in earnings across years. Individual earnings reported in the <abbr>SIPP</abbr> may differ from those in the <abbr class="spell">DER</abbr> for reasons other than error. Respondents may report on a maximum of two jobs in the survey, and the administrative records report all jobs. Administrative records exclude pretax health care premiums paid by the employee or contributions to <span class="nobr">401(k)</span> plans out of earnings that may be accurately reported in the survey as prededuction earnings.<sup><a href="#mn11" id="mt11">11</a></sup></p>
|
|
<p>Gottschalk and Huynh (2005) found that the <abbr class="spell">DER</abbr> had consistently higher employment rates than those in the <abbr>SIPP</abbr>. Respondents with missing <abbr>SIPP</abbr> data on earnings tend to have lower earnings in the <abbr class="spell">DER</abbr> than respondents with observed earnings in both data sets. Similarly, respondents with positive <abbr>SIPP</abbr> earnings and no <abbr class="spell">DER</abbr> earnings had lower earnings than respondents with observed earnings in both data sets, possibly reflecting informal work arrangements. Bridges, Del Bene, and Leonesio (2003) obtained qualitatively similar results from their 1993 <abbr>SIPP</abbr>/<abbr class="spell">DER</abbr> earnings comparisons. Gottschalk and Huynh (2005) found that the number of individuals with positive <abbr>SIPP</abbr> earnings and no <abbr class="spell">DER</abbr> earnings was smaller than the number with positive <abbr class="spell">DER</abbr> earnings and no <abbr>SIPP</abbr> earnings. However, Bridges, Del Bene, and Leonesio (2003) found the opposite pattern. Gottschalk and Huynh (2005) also found that lifetime earnings patterns were similar in the two data sources. Men <span class="nobr">aged 25–59</span> had higher earnings in the <abbr class="spell">DER</abbr> than in the <abbr>SIPP</abbr>, but there were no systematic differences in earnings between the two data sources for older men or for women. Finally, correlations between <abbr>SIPP</abbr> nonimputed earnings and <abbr class="spell">DER</abbr> earnings are approximately 0.75 for men and women <span class="nobr">aged 25–59</span> and 65 or older. Bridges, Del Bene, and Leonesio (2003) found substantial measurement error in <abbr>SIPP</abbr> wage and salary data, with mean <abbr>SIPP</abbr> wages understated by 7.5 percent relative to <abbr class="spell">DER</abbr> wages. The absolute relative error in wage and salary income was 18 percent overall, but 28 percent for those with imputed earnings.</p>
|
|
<p>Measurement error for wage and salary income is an important and complex area for future research. Survey data on earnings are reported for different time periods (weekly, monthly, annual), different concepts (gross or net of income taxes), and different sources (primary job, secondary job, wage and salary income, self-employment income). Likewise, administrative earnings records may record different concepts depending on the programmatic purpose for which they are collected. Comparisons of survey data on earnings and matched administrative data on earnings may lead to improvements in survey imputations of missing earnings data, more accurate analyses of individual well-being, and improved policy estimates of the distributional effects of <abbr class="spell">OASDI</abbr> (Social Security) and <abbr class="spell">SSI</abbr> reform proposals.</p>
|
|
<h2>Deferred Compensation</h2>
|
|
<p>Many researchers have documented the dramatic shift in the employer-provided pension environment from defined benefit (<abbr class="spell">DB</abbr>) pensions to defined contribution (<abbr class="spell">DC</abbr>) pensions (Munnell and Sunden 2004; Costo 2006; Buessing and Soto 2006; Poterba and others 2006; Dushi and Iams 2007). Traditional <abbr class="spell">DB</abbr> pensions are funded by the employer and provide retirement benefits based on a formula that usually considers final salary, years of service, and age. All employees typically are included in the plan. Upon retirement, monthly benefits are generally paid in the form of a life annuity. Defined contribution plans (for example, <span class="nobr">401(k)</span> and <span class="nobr">403(b)</span> plans), on the other hand, place more risks and responsibilities on employees, and enrollment often is not automatic. After enrolling, employees must make decisions about contribution amounts and investment allocations. Employee contributions to <abbr class="spell">DC</abbr> pension plans are treated as deferred compensation, meaning that contributions are made on a pretax basis. Taxes are usually paid when funds are withdrawn. Upon retirement, employees face many options for withdrawing their <abbr class="spell">DC</abbr> account balances, including lump-sum withdrawals, the purchase of whole- or partial-life annuities, and rollover of funds into a tax-preferred individual retirement account from which withdrawals may be made.</p>
|
|
<p>The <abbr class="spell">HRS</abbr> has become a premier source of data for studying changes in the pension environment, pension plan participation by employees, and pension income of retirees, among other important topics related to retirement and older Americans. Importantly, on a restricted basis, researchers can access <abbr class="spell">HRS</abbr> data matched to <abbr class="spell">SSA</abbr> administrative data on benefits and earnings. The earnings records are derived from <abbr class="spell">IRS</abbr> W-2 records submitted by employers on behalf of their employees. These records provide data on annual tax-deferred contributions by employees to <abbr class="spell">DC</abbr> pension accounts. Dushi and Honig (2008) compared the deferred compensation data from <abbr class="spell">IRS</abbr> W-2 tax records with the self-reported pension type and pension contributions of <abbr class="spell">HRS</abbr> respondents to determine the accuracy of the self-reports and to assess employee understanding of the mechanics of <abbr class="spell">DB</abbr> and <abbr class="spell">DC</abbr> pension plans.</p>
|
|
<p>Table 6 provides some estimates from Dushi and Honig (2008) on the accuracy of self-reported <abbr class="spell">DB</abbr> and <abbr class="spell">DC</abbr> pension plan participation among <abbr class="spell">HRS</abbr> respondents born in the period from 1931 through 1941 <span class="nobr">(aged 51–61</span> in 1992). Thirty percent of individuals who reported having a <abbr class="spell">DB</abbr>-only pension plan had positive contributions to a <abbr class="spell">DC</abbr> pension plan on their W-2 record, which suggests that these individuals misreported their pension plan type in the <abbr class="spell">HRS</abbr>. Thirty-nine percent of individuals who reported having a <abbr class="spell">DC</abbr>-only pension plan had zero contributions to a <abbr class="spell">DC</abbr> pension plan on their W-2 record. This may reflect misreporting of <abbr class="spell">DB</abbr> pension plans as <abbr class="spell">DC</abbr> pension plans, or it may reflect actual lack of contributions to the <abbr class="spell">DC</abbr> plan during the year in question. Finally, 6 percent of individuals who reported that they were not included in a pension plan had positive contributions to a <abbr class="spell">DC</abbr> pension plan on their W-2 record, again suggesting a nontrivial amount of misreporting of pension plan type in the <abbr class="spell">HRS</abbr>. This is clearly an important area for future research.</p>
|
|
<div class="table" id="table6">
|
|
<table>
|
|
<caption><span class="tableNumber">Table 6. </span>Mismatch between self-reported pension type in the <abbr class="spell">HRS</abbr> and pension contributions from matched W-2 data among the <abbr class="spell">HRS</abbr> cohort <span class="nobr">aged 51–61</span> in 1992 (in percent)</caption>
|
|
<colgroup span="1" style="width:18em"></colgroup>
|
|
<colgroup>
|
|
<col style="width:6em">
|
|
<col style="width:6em">
|
|
<col style="width:6em">
|
|
<col class="shaded" style="width:6em">
|
|
</colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th class="stubHeading" rowspan="2" scope="colgroup">Self-reported pension type in the <abbr class="spell">HRS</abbr></th>
|
|
<th class="spanner" colspan="4" scope="colgroup">Amount of contribution to <abbr class="spell">DC</abbr> pension from W-2 record</th>
|
|
</tr>
|
|
<tr>
|
|
<th scope="col">Zero </th>
|
|
<th scope="col">Greater<br>than zero </th>
|
|
<th scope="col">Total </th>
|
|
<th scope="col">N</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<th class="stub0" scope="row"><abbr class="spell">DB</abbr> only </th>
|
|
<td>70</td>
|
|
<td>30</td>
|
|
<td>100</td>
|
|
<td>1,084</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row"><abbr class="spell">DC</abbr> only </th>
|
|
<td>39</td>
|
|
<td>61</td>
|
|
<td>100</td>
|
|
<td>1,406</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Both <abbr class="spell">DB</abbr> and <abbr class="spell">DC</abbr> </th>
|
|
<td>44</td>
|
|
<td>56</td>
|
|
<td>100</td>
|
|
<td>85</td>
|
|
</tr>
|
|
<tr>
|
|
<th class="stub0" scope="row">Not included in a pension plan</th>
|
|
<td>94</td>
|
|
<td>6</td>
|
|
<td>100</td>
|
|
<td>1,333</td>
|
|
</tr>
|
|
</tbody>
|
|
<tfoot>
|
|
<tr>
|
|
<td class="firstNote" colspan="5">SOURCE: Dushi and Honig (2008, Table 3).</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="lastNote" colspan="5">NOTE: Percentages are weighted. Sample counts (N) are unweighted. Forty-two <abbr class="spell">HRS</abbr> observations with a missing pension plan type were excluded from the table.</td>
|
|
</tr>
|
|
</tfoot>
|
|
</table>
|
|
</div>
|
|
<h2>Concluding Observations</h2>
|
|
<p>The ability to use survey data matched with administrative data is tremendously beneficial for a wide variety of research applications, from policy evaluation to economic research and program statistics to microsimulation modeling. A fundamental use of matched survey and administrative data by researchers at <abbr class="spell">SSA</abbr> has been to assess the accuracy of the survey data and to adjust for error in research and statistics produced from survey data. The primary surveys used in these types of analyses are the <abbr>SIPP</abbr>, <abbr class="spell">CPS</abbr>, and <abbr class="spell">HRS</abbr>, which may be accessed only on a restricted basis, subject to the terms and conditions specified by their parent entities and the agencies with authority over the matched administrative data files. This article reports on some important findings from these surveys with respect to survey measurement in the areas of <abbr class="spell">OASDI</abbr> (Social Security) and <abbr class="spell">SSI</abbr> participation and benefit amounts, disability diagnosis, earnings, and deferred compensation. The general findings regarding <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> participation and benefit amounts appear to be quite robust across data sources and in terms of their implications for analyses of beneficiary well-being and poverty status. Research on measuring disability diagnosis, earnings, and deferred compensation using matched survey and administrative data is in its infancy. We summarize the key findings as follows.</p>
|
|
<ul>
|
|
<li>Self-reported data in the <abbr class="spell">CPS</abbr> slightly underreport <abbr class="spell">OASDI</abbr> receipt and significantly underreport <abbr class="spell">SSI</abbr> receipt. Self-reported data in the <abbr>SIPP</abbr> slightly overreport receipt of <abbr class="spell">OASDI</abbr>; however, the picture is more complicated for receipt of <abbr class="spell">SSI</abbr> depending on the year of analysis and whether the data are analyzed from a monthly or annual perspective. Estimates from both surveys indicate some confusion among respondents between the two sources of income. When administrative data are used in place of self-reported survey data, estimated poverty rates fall, especially among <abbr class="spell">SSI</abbr> recipients.</li>
|
|
<li>For disability research, both survey and administrative data have appreciable strengths depending on the specific application of the data. Survey data are more likely to better reflect the perspective of the individual and often contain measures of functional limitations and severity that are not available from administrative records. The disability information in matched administrative records may better reflect the concepts of interest for more programmatically oriented studies.</li>
|
|
<li>There appears to be substantial misreporting of pension type based on comparisons between self-reported pension type and administrative data on annual contributions to <abbr class="spell">DC</abbr> pension accounts. Matched administrative data from <abbr class="spell">IRS</abbr> W-2 records and other sources hold great promise for improving the measurement of pension plan participation and contribution amounts.</li>
|
|
</ul>
|
|
<h2>Future Research</h2>
|
|
<p>One area that is ripe for future research is the extent to which self-reported earnings in the <abbr>SIPP</abbr>, <abbr class="spell">CPS</abbr>, and <abbr class="spell">HRS</abbr> agree with earnings captured in <abbr class="spell">SSA</abbr>'s administrative records systems. This is an important measurement issue, especially for the working-age population. It is also a complex measurement issue. Survey data on earnings are captured in many forms (weekly, monthly, annual—gross or net of income taxes) and for different sources (primary job, secondary job, wage and salary income, self-employment income). In <abbr class="spell">SSA</abbr>'s administrative records systems, earnings may be recorded differently depending on whether they are counted when earned or when received, or whether they are actual or countable, estimated or verified, monthly or annual. A systematic comparison of survey-based earnings measures and matched administrative data on earnings may lead to improvements in survey imputations of missing earnings data and more accurate analyses of individual well-being and the distributional implications of <abbr class="spell">OASDI</abbr> and <abbr class="spell">SSI</abbr> policies.</p>
|
|
<p>Finally, although they were not addressed in this article, some studies on mortality also have used <abbr class="spell">SSA</abbr> administrative records matched to survey data. Age-specific death rates typically are constructed by combining vital statistics on the number of deaths (numerator) with Census data on the size of the at-risk population (denominator). Administrative records provide these data from a single source (Lauderdale and Kestenbaum 2002), but do not necessarily contain the socioeconomic variables needed to compute subgroup-specific death rates that may be of interest to researchers. Survey data matched with administrative data provide a broader picture of the population; however, very few surveys were conducted long enough ago and have a sufficiently high match rate to administrative data to support detailed analyses.</p>
|
|
<div id="notes">
|
|
<h2>Notes</h2>
|
|
<p> <a href="#mt1" id="mn1">1</a> See the <abbr>SIPP</abbr> home page for additional details (<a href="https://www.census.gov/programs-surveys/sipp.html">www.census.gov/sipp/</a>).</p>
|
|
<p> <a href="#mt2" id="mn2">2</a> See the <abbr class="spell">CPS</abbr> home page for additional details (<a href="https://www.census.gov/programs-surveys/cps.html">www.census.gov/CPS/</a>).</p>
|
|
<p> <a href="#mt3" id="mn3">3</a> See the <abbr class="spell">HRS</abbr> home page for additional details (<a href="https://hrs.isr.umich.edu/">hrsonline.isr.umich.edu/</a>).</p>
|
|
<p> <a href="#mt4" id="mn4">4</a> See the <abbr class="spell">NSCF</abbr> home page for additional details (<a href="/disabilityresearch/nscf.htm">www.socialsecurity.gov/disabilityresearch/nscf.htm</a>). See also Davies and Rupp (2005/2006) and Rupp and others (2005/2006).</p>
|
|
<p> <a href="#mt5" id="mn5">5</a> See the <abbr class="spell">NHIS</abbr> home page for additional details (<a href="https://www.cdc.gov/nchs/nhis.htm">www.cdc.gov/nchs/nhis.htm</a>).</p>
|
|
<p> <a href="#mt6" id="mn6">6</a> See the <abbr class="spell">NHANES</abbr> home page for additional details (<a href="https://www.cdc.gov/nchs/nhanes.htm">www.cdc.gov/nchs/nhanes.htm</a>).</p>
|
|
<p> <a href="#mt7" id="mn7">7</a> Sizeable differences between the <abbr class="spell">MBR</abbr> and <abbr class="spell">PHUS</abbr> would arise predominantly for Social Security Disability Insurance (<abbr class="spell">DI</abbr>) beneficiaries who went through the appeals process. Upon the award of the <abbr class="spell">DI</abbr> benefit, the <abbr class="spell">MBR</abbr> would be updated to reflect benefits paid retroactively to the date of entitlement, whereas the <abbr class="spell">PHUS</abbr> would show one large lump-sum payment for the month of award and zero payments before award.</p>
|
|
<p> <a href="#mt8" id="mn8">8</a> Sears and Rupp (2003) compared results using the <abbr class="spell">MBR</abbr> and <abbr class="spell">PHUS</abbr> with Huynh, Rupp, and Sears (2002) and found the differences to be negligible. They found that the percentage of March 1996 respondents who reported the exact amount of the administrative <abbr class="spell">OASDI</abbr> benefit improved to 51 percent with the <abbr class="spell">PHUS</abbr> compared with 46 percent in the earlier study using the <abbr class="spell">MBR</abbr>, but there was no corresponding improvement in the estimated mean error between the survey and administrative benefit amounts. This suggests that large lump-sum payments to <abbr class="spell">DI</abbr> awardees occurred relatively rarely among <abbr>SIPP</abbr> respondents. However, Huynh, Rupp, and Sears (2002) did not disaggregate by age or type of <abbr class="spell">OASDI</abbr> benefit, so we can only speculate without further research.</p>
|
|
<p> <a href="#mt9" id="mn9">9</a> Olson (2002) analyzed the consistency between Social Security benefit amounts for May 1990 in the <abbr>SIPP</abbr> and the <abbr class="spell">MBR</abbr>.</p>
|
|
<p><a href="#mt10" id="mn10">10</a> Beginning in 2006, the <abbr class="spell">HRS</abbr> also collects detailed data on physical performance measures, biomarkers, and psychological topics through enhanced face-to-face interviews with selected respondents. These data are not addressed in this article.</p>
|
|
<p><a href="#mt11" id="mn11">11</a> Abowd and Stinson (2004) developed a procedure that allows for potential measurement error in both data sources.</p>
|
|
</div>
|
|
<div id="references">
|
|
<h2>References</h2>
|
|
<p>Abowd, John, and Martha Stinson. 2004. Estimating measurement error in <abbr>SIPP</abbr> annual job earnings: A comparison of Census survey and <abbr class="spell">SSA</abbr> administrative data. Mimeo (July).</p>
|
|
<p>Aziz, Faye, Beth Kilss, and Frederick Scheuren. 1978. <i>1973 Current Population Survey: Administrative record exact match file codebook, part I—code counts and item definitions</i>. Studies from Interagency Data Linkages, Report <abbr title="Number">No.</abbr> 8. Washington, <abbr class="spell">DC</abbr>: Department of Health, Education, and Welfare, Publication <abbr title="Number">No.</abbr> (<abbr class="spell">SSA</abbr>) 79-11750.</p>
|
|
<p>Bound, John, and Timothy Waidmann. 1992. Disability transfers, self-reported health and the labor force attachment of older men: Evidence from the historical record. <i>Quarterly Journal of Economics</i> 107(4): <span class="nobr">1393–1420.</span></p>
|
|
<p>Bridges, Benjamin, Linda Del Bene, and Michael V. Leonesio. 2003. Evaluating the accuracy of 1993 <abbr>SIPP</abbr> earnings through the use of matched Social Security Administrative data. 2002 Proceedings of the American Statistical Association, Survey Research Methods Section. Alexandria, <abbr title="Virginia">VA</abbr>: American Statistical Association, <span class="nobr">306–311.</span></p>
|
|
<p>Buessing, Marric, and Mauricio Soto. 2006. The state of private pensions: Current 5500 data. Issue Brief <abbr title="Number">No.</abbr> 42. Chestnut Hill, <abbr title="Massachusetts">MA</abbr>: Center for Retirement Research at Boston College (February).</p>
|
|
<p>Costo, Stephanie L. 2006. Trends in retirement plan coverage over the last decade. <i>Monthly Labor Review</i> 129(2): <span class="nobr">59–64</span>.</p>
|
|
<p>Czajka, John L., James Mabli, and Scott Cody. 2008. <i>Sample loss and survey bias in estimates of Social Security beneficiaries: A tale of two surveys</i>. Washington, <abbr class="spell">DC</abbr>: Mathematica Policy Research, <abbr title="Incorporated">Inc.</abbr></p>
|
|
<p>Davies, Paul S., and Kalman Rupp. 2005/2006. An overview of the National Survey of <abbr class="spell">SSI</abbr> Children and Families and related products. <i>Social Security Bulletin</i> 66(2): <span class="nobr">7–20.</span></p>
|
|
<p>Dushi, Irena, and Marjorie Honig. 2008. How much do respondents in the Health and Retirement Study know about their tax-deferred contribution plans? A cross-cohort comparison. Working Paper <abbr title="Number">No.</abbr> 2008-201. Ann Arbor, <abbr title="Michigan">MI</abbr>: Retirement Research Center at the University of Michigan.</p>
|
|
<p>Dushi, Irena, and Howard Iams. 2007. Cohort differences in wealth and pension participation of near-retirees. Paper presented at the Population Association of America Annual Meeting, New York, <abbr title="New York">NY</abbr> (<span class="nobr">March 29–31).</span></p>
|
|
<p>Fisher, T. Lynn. 2005. Measurement of reliance on Social Security benefits. Paper presented at the Federal Committee on Statistical Methodology Research Conference, Washington, <abbr class="spell">DC</abbr> (November 15).</p>
|
|
<p>———. 2008. The impact of survey choice on measuring the relative importance of Social Security benefits to the elderly. <i>Social Security Bulletin</i> 67(2): <span class="nobr">55–64.</span></p>
|
|
<p>Gottschalk, Peter, and Minh Huynh. 2005. Validation study of earnings data in the <abbr>SIPP</abbr>—Do older workers have larger measurement error? Working Paper <abbr title="Number">No.</abbr> 2005-07. Chestnut Hill, <abbr title="Massachusetts">MA</abbr>: Center for Retirement Research at Boston College.</p>
|
|
<p>Huynh, Minh, Kalman Rupp, and James Sears. 2002. <i>The assessment of Survey of Income and Program Participation benefit data using longitudinal administrative records</i>. Survey of Income and Program Participation Report <abbr title="Number">No.</abbr> 238. Washington, <abbr class="spell">DC</abbr>: Census Bureau.</p>
|
|
<p>Kilss, Beth, and Frederick J. Scheuren. 1978. The 1973 <abbr class="spell">CPS</abbr>-<abbr class="spell">IRS</abbr>-<abbr class="spell">SSA</abbr> exact match study. <i>Social Security Bulletin</i> 41(10): <span class="nobr">14–22.</span></p>
|
|
<p>Koenig, Melissa. 2003. An assessment of the Current Population Survey and the Survey of Income and Program Participation using Social Security Administrative data. Paper presented at the Federal Committee on Statistical Methodology Research Conference, Washington, <abbr class="spell">DC</abbr> (November 18).</p>
|
|
<p>Kreider, Brent. 1999. Latent work disability and reporting bias. <i>Journal of Human Resources</i> 34(4): <span class="nobr">734–769.</span></p>
|
|
<p>Lauderdale, Diane S., and Bert Kestenbaum. 2002. Mortality rates of elderly Asian American populations based on Medicare and Social Security data. <i>Demography</i> 39(2): <span class="nobr">529–540.</span></p>
|
|
<p>Lininger, Charles A. 1981. The goals and objectives of the Survey of Income and Program Participation. 1980 Proceedings of the American Statistical Association, Survey Research Methods Section. Alexandria, <abbr title="Virginia">VA</abbr>: American Statistical Association, <span class="nobr">480–485.</span></p>
|
|
<p>Munnell, Alicia H., and Annika Sunden. 2004. <i>Coming up short: The challenge of <span class="nobr">401(k)</span> plans</i>. Washington, <abbr class="spell">DC</abbr>: Brookings Institution Press.</p>
|
|
<p>Nicholas, Joyce, and Michael Wiseman. 2009. Elderly poverty and Supplemental Security Income. <i>Social Security Bulletin</i> 69(1): <span class="nobr">45–73.</span></p>
|
|
<p>Olson, Janice A. 2002. Social Security benefit reporting in the Survey of Income and Program Participation and in Social Security administrative records. <abbr class="spell">ORES</abbr> Working Paper Series <abbr title="Number">No.</abbr> 96. Washington, <abbr class="spell">DC</abbr>: Social Security Administration.</p>
|
|
<p>Poterba, M. James, Steven F. Venti, Joshua Rauh, and David A. Wise. 2006. Defined contribution plans, defined benefit plans, and the accumulation of retirement wealth. <abbr class="spell">NBER</abbr> Working Paper <abbr title="Number">No.</abbr> 12597. Cambridge, <abbr title="Massachusetts">MA</abbr>: National Bureau of Economic Research.</p>
|
|
<p>Rupp, Kalman, and Paul S. Davies. 2004. A long-term view of health status, disabilities, mortality, and participation in the <abbr class="spell">DI</abbr> and <abbr class="spell">SSI</abbr> disability programs. In <i>Research in Labor Economics: Accounting for Worker Well-Being</i>, <abbr title="Volume">Vol.</abbr> 23, Solomon Polachek, <abbr title="editor">ed.</abbr>, <span class="nobr">119–183.</span> Amsterdam, Netherlands: Elsevier, <abbr class="spell">JAI</abbr> Press.</p>
|
|
<p>Rupp, Kalman, Paul S. Davies, Chad Newcomb, Howard Iams, Carrie Becker, Shanti Mulpuru, Stephen Ressler, Kathleen Romig, and Baylor Miller. 2005/2006. A profile of children with disabilities receiving <abbr class="spell">SSI</abbr>: Highlights from the National Survey of <abbr class="spell">SSI</abbr> Children and Families. <i>Social Security Bulletin</i> 66(2): <span class="nobr">21–48.</span></p>
|
|
<p>Scheuren, Frederick, and Roger Herriot. 1975. <i>Chapter 1: General introduction and background, exact match research using the March 1973 Current Population Survey</i>—<i>initial states</i>. Studies from Interagency Data linkages, Report <abbr title="Number">No.</abbr> 4. Washington, <abbr class="spell">DC</abbr>: Department of Health, Education, and Welfare, Publication <abbr title="Number">No.</abbr> <abbr class="spell">SSA</abbr> 76-11750.</p>
|
|
<p>Sears, James, and Kalman Rupp. 2003. Exploring Social Security payment history matched with the Survey of Income and Program Participation. Paper presented at the Federal Committee on Statistical Methodology Research Conference, Washington, <abbr class="spell">DC</abbr> (November 18).</p>
|
|
<p>Sickles, Robin C., and Paul Taubman. 1997. Mortality and morbidity among adults and the elderly. In <i>Handbook of Population and Family Economics</i>, <abbr title="Volume">Vol.</abbr> 1A, Mark R. Rosenzweig and Oded Stard, <abbr title="editors">eds.</abbr>, <span class="nobr">559–643.</span> Amsterdam, Netherlands: Elsevier.</p>
|
|
<p>Vaughan, Denton R. 1979. Errors in reporting Supplemental Security Income recipiency in a pilot household survey. 1978 Proceedings of the American Statistical Association, Survey Research Methods Section. Alexandria, <abbr title="Virginia">VA</abbr>: American Statistical Association, <span class="nobr">288–293.</span></p>
|
|
<p>———. 1989. Development and evaluation of a survey-based type of benefit classification for the Social Security program. <i>Social Security Bulletin</i> 52(1): <span class="nobr">12–26.</span></p>
|
|
<p>Vaughan, Denton R., T. Cameron Whiteman, and Charles A. Lininger. 1984. The quality of income and program data in the 1979 <abbr class="spell">ISDP</abbr> research panel: Some preliminary findings. <i>Review of Public Data Use</i>, <abbr title="Volume">Vol.</abbr> 12, <span class="nobr">107–131.</span></p>
|
|
<p>Ycas, Martynas A., and Charles A. Lininger. 1981. The income survey development program: A review. 1980 Proceedings of the American Statistical Association, Survey Research Methods Section. Alexandria, <abbr title="Virginia">VA</abbr>: American Statistical Association, <span class="nobr">486–490.</span></p>
|
|
</div>
|
|
</div>
|
|
</article>
|
|
<footer><div id="footer">
|
|
<div class="important-info"><h4>Important Information:</h4>
|
|
<ul><li><a href="/agency/">About Us</a></li>
|
|
<li><a href="/accessibility/">Accessibility</a></li>
|
|
<li><a href="/foia/">FOIA</a></li>
|
|
<li><a href="/open/">Open Government</a></li>
|
|
<li><a href="/agency/glossary/">Glossary</a></li>
|
|
<li><a href="/privacy/">Privacy</a></li>
|
|
<li><a href="https://oig.ssa.gov/report/">Report Fraud, Waste or Abuse</a></li>
|
|
<li><a href="/agency/websitepolicies.html">Website Policies</a></li></ul>
|
|
</div>
|
|
<p class="align-center margin-top">This website is produced and published at U.S. taxpayer expense.</p>
|
|
</div></footer>
|
|
<!-- SSA INTERNET BODY SCRIPTS -->
|
|
<script src="/policy/js/rspa.doc.js"></script>
|
|
<script src="/policy/js/rspa-shared.js"></script>
|
|
<script src="/framework/js/ssa.internet.body.js"></script>
|
|
</body></html> |