Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
What methodology did the Center for Immigration Studies use to calculate 59% on welfare?
Executive summary
The Center for Immigration Studies (CIS) reported a 59% welfare-use figure for immigrant-headed households with low education based on prior analyses of the Current Population Survey (CPS), while CIS’s later work relies primarily on the Census Bureau’s Survey of Income and Program Participation (SIPP), which they argue gives higher and more complete welfare coverage (CPS showed 59% for immigrant-headed households without a high school diploma; CIS says SIPP understates/overstates comparisons are important) [1]. Critics including Cato, the Center on Budget and Policy Priorities (CBPP), and others say CIS’s methodological choices—such as how it defines “immigrant households,” which programs are counted, and how household-level benefit receipt is attributed—inflate the figures [2] [3].
1. How CIS arrived at “59%”: the datasets and comparisons
CIS’s cited 59% figure comes from prior analysis of the Current Population Survey (CPS) showing that 59 percent of households headed by an immigrant without a high school education used one or more welfare programs; CIS notes that its later SIPP-based work generally finds even higher welfare-use shares and argues the SIPP is superior for measuring program participation [1] [4]. In short: the 59% number is reported as a CPS-based calculation for a specific subgroup (immigrant-headed, no high school) and is presented in CIS publications as comparable context to their SIPP findings [1].
2. CIS’s preferred methodology: why they use the SIPP
CIS emphasizes the Survey of Income and Program Participation (SIPP) because the Census Bureau designs SIPP specifically to capture income and program participation; CIS’s 2015 and later analyses therefore use SIPP to measure household-level receipt of Medicaid, cash assistance, food programs, housing, and other benefits and to compare immigrant- and native-headed households [1] [4]. CIS also reports having its SIPP calculations double-checked by an outside demographic firm, Decision Demographics, and frames SIPP as “the best data available” for program participation [1] [5].
3. Key methodological choices that affect the percent
CIS’s results depend on several definitional and analytic choices: counting welfare at the household level (so if any household member receives benefits the whole household is marked as a welfare-using household); including naturalized citizens when identifying “immigrant-headed” households; and deciding which programs to include (Medicaid, SNAP, WIC, TANF, SSI, EITC, etc.) [1] [3] [4]. Those choices raise the measured share because a household may include U.S.-born members receiving benefits while still being labeled “immigrant-headed,” and including more programs increases the likelihood a household will be classified as using “one or more” programs [3] [4].
4. Critics’ objections and alternative analyses
Independent critics and some think tanks have challenged CIS’s methodology. Cato argued CIS made “an unsound methodological choice” that inflates welfare-use rates and accused CIS of overstating immigrant and noncitizen welfare use [2] [6]. The Center on Budget and Policy Priorities and others faulted CIS for attributing benefits received by U.S.-born household members to immigrant-headed households and for other data-handling choices; CBPP called earlier CIS claims “manipulated” and contested their framing [3]. These critiques highlight that small differences in definitions and which programs are counted can materially change headline percentages.
5. What’s agreed, what’s disputed, and what reporting does not say
There is agreement that the choice of survey (CPS vs. SIPP), program list, and whether benefits are attributed at the household level strongly influence any welfare-use percentage [1] [4]. Disagreement centers on which choices are appropriate: CIS defends SIPP and household attribution and says their numbers reflect broader program coverage, while critics say those choices overstate immigrant-specific use and conflate U.S.-born recipients in immigrant-headed households [1] [3] [2]. Available sources do not mention a step‑by‑step replication file in the public domain that would let outside researchers reproduce the exact 59% CPS calculation reported by CIS (not found in current reporting).
6. Bottom line for readers assessing the 59% claim
The 59% figure is traceable to CIS’s use of Census surveys (CPS in the referenced instance) and to definitional choices that raise household-level program counts; CIS prefers SIPP for completeness and reports even higher shares in its SIPP analyses [1] [4]. Independent analysts dispute CIS’s choices—particularly household attribution and program inclusion—and show that alternative, defensible methodological choices produce materially different, lower estimates [2] [3]. If you need a single next step: compare the CIS report’s methodology section to independent replications (Cato, CBPP) and check whether program inclusion and household-attribution rules match the specific policy question you care about [1] [6] [3].