Skip to main content
v2026.11,986 entries · CC-BY 4.0

Confirmatory Factor Analysis (CFA): A Complete Guide

Confirmatory factor analysis tests a pre-specified measurement model against new data — unlike exploratory factor analysis, which discovers structure. This guide covers specification, identification, estimation, fit indices, modification-index risk, and measurement invariance.

Written and maintained by CASRAI Editorial Board

Published · Last updated

On this page

Confirmatory factor analysis (CFA) is a statistical technique used to test whether a set of observed variables (survey items, test scores, indicators) fit a measurement model that was specified in advance, based on theory or prior research. That single feature — the model is fixed before the data are analyzed — is what separates CFA from exploratory factor analysis (EFA), and it is the source of most confusion between the two. In EFA, the number of factors and which items load on which factor are discovered from the data. In CFA, the researcher specifies which items belong to which factor, how many factors exist, and (often) which factors correlate with one another, and then tests whether the data are consistent with that structure.

CFA is a special case of structural equation modeling (SEM) restricted to the measurement part of a model — the relationships between latent variables and their observed indicators — without the structural (regression) paths between latent variables that a full SEM would add. It is the standard tool for validating a scale or questionnaire’s factor structure in a new sample, confirming that a previously developed instrument still behaves the way it did in the original development study, and establishing that a set of items can legitimately be treated as a single composite score before that score is used in further analysis.

What is confirmatory factor analysis? Confirmatory factor analysis (CFA, sometimes written “cfa analysis”) is a statistical method that tests whether observed items fit a factor structure specified in advance from theory or earlier research. The researcher fixes how many factors exist and which items load on each, then judges the fit with indices such as CFI, RMSEA and SRMR. A good fit supports the proposed measurement model.

CFA vs. EFA: Why the Distinction Matters

The core difference is confirmatory versus exploratory logic, not a difference in the underlying factor-analytic machinery:

  • EFA is used when there is no strong prior theory about how many factors underlie a set of items or which items belong together. It lets the data suggest a structure — typically during initial scale development, when a researcher has generated a pool of candidate items and needs to see how they group empirically.
  • CFA is used when a structure already exists — from a prior EFA, from established theory, or from a previously published and validated instrument — and the question is whether that specific structure holds in a new sample, population, language, or context.

A well-established methodological rule follows directly from this: EFA and CFA should not be run on the same sample. Using EFA to find a structure and then immediately running CFA on the same data to “confirm” it is circular — the CFA is almost guaranteed to fit well because the factor structure was extracted from that exact data, capitalizing on sample-specific noise rather than testing anything independent. The standard practice is to split the sample (commonly via random split-half) and run EFA on one half and CFA on the other, or to run the EFA on one sample and reserve CFA for a genuinely separate sample or a later replication study. If only one sample is available and no prior theory exists, that is a signal to run EFA and stop there, being explicit that the structure is exploratory and needs confirmation elsewhere.

Specifying a CFA Model

A CFA model consists of several building blocks:

  • Latent variables (factors) — the unobserved constructs the researcher is trying to measure (e.g., “anxiety,” “job satisfaction,” “instrument usability”). They cannot be measured directly, which is the entire reason a measurement model is needed.
  • Observed indicators — the actual items, test scores, or ratings collected from participants, each hypothesized to reflect one (or occasionally more than one) latent factor.
  • Factor loadings — the estimated strength of the relationship between each indicator and the latent factor it belongs to, analogous to a regression coefficient predicting the observed item from the unobserved factor.
  • Error (residual) terms — the portion of each indicator’s variance not explained by the latent factor, capturing measurement error and anything specific to that one item.
  • Correlated factors — in most applied CFA models, latent factors are allowed to correlate with one another rather than being forced orthogonal (uncorrelated), since psychological and organizational constructs are rarely independent in reality.
  • Correlated errors — occasionally two indicators’ error terms are allowed to correlate, usually because they share wording, method, or some other source of overlap not accounted for by the latent factor. This should be theoretically justified, not added purely to improve fit (see the section on modification indices below).

Identification

Before a CFA model can be estimated, it must be identified — there must be enough known information (the observed variances and covariances) to solve for the unknown parameters (loadings, factor variances, error variances) uniquely. Two mechanical requirements govern this:

  • The t-rule: the number of freely estimated parameters cannot exceed the number of unique elements in the observed variance-covariance matrix (p(p+1)/2 for p indicators). This is a necessary but not sufficient condition for identification.
  • The three-indicator rule: a standard, sensible practice is that every latent factor should have at least three observed indicators (two-indicator factors can be identified but are fragile and more sensitive to specification error; single-indicator factors require fixing measurement error externally).

A further, separate requirement is scaling the latent variable, which has no inherent metric of its own. This is done one of two equivalent ways: fixing one indicator’s loading to 1 (the “marker variable” or “reference indicator” method, which gives the factor the scale of that one item), or fixing the latent factor’s variance to 1 (the “standardized” method, which puts all loadings on a correlation-like metric). Most CFA software defaults to the marker-variable approach; both give mathematically equivalent model fit.

Estimation: ML, MLR, and WLSMV

The default estimator in most CFA software is maximum likelihood (ML), which assumes the observed indicators are continuous and multivariate normal. In practice, this assumption is routinely violated — and two robust alternatives are standard practice depending on why it fails:

  • MLR (robust/Huber-White maximum likelihood) — used when indicators are continuous but non-normally distributed (skewed or kurtotic). MLR produces the same point estimates as ML but corrects the standard errors and the chi-square test statistic (the Satorra-Bentler correction) for non-normality, without requiring a different estimation approach.
  • WLSMV (weighted least squares with mean- and variance-adjustment) — used when indicators are ordinal or categorical, which is the typical case for Likert-type survey items. WLSMV treats each item as a discretized version of an underlying continuous variable and estimates polychoric (rather than Pearson) correlations among items before fitting the model.

A common and consequential mistake is running ordinary ML on Likert-scale data treated as if it were continuous. With five or more roughly symmetric response categories the practical distortion is often modest, but with fewer categories, skewed distributions, or ordinal data with only two or three levels, ML on Likert data understates factor loadings and distorts fit indices, typically making a genuinely acceptable model look worse than it is. WLSMV (or a robust ML variant applied to polychoric correlations) is the methodologically preferred choice for ordinal indicators, and is what most published psychometric validation studies now use for Likert-based instruments.

Evaluating Model Fit

No single statistic determines whether a CFA model “fits.” The convention is to report a small set of complementary indices together:

  • Chi-square (χ²) test — tests the null hypothesis that the model-implied covariance matrix equals the population covariance matrix exactly. A non-significant χ² technically indicates good fit, but the test is well known to over-reject at large sample sizes: with a large enough N, even trivial, substantively unimportant discrepancies between the model and the data become statistically significant. For this reason χ² is reported (often alongside its degrees of freedom, and sometimes the χ²/df ratio) but is rarely treated as the deciding criterion on its own in applied work.
  • CFI (Comparative Fit Index) and TLI (Tucker-Lewis Index) — incremental fit indices that compare the fit of the specified model to a baseline (null) model in which all variables are uncorrelated. Conventional guidance treats values at or above .95 as indicating good fit.
  • RMSEA (Root Mean Square Error of Approximation) — a “badness of fit” index that estimates how far the model is from the population, per degree of freedom, with a built-in penalty for model complexity. Conventional guidance treats values at or below .06 as indicating good fit, and RMSEA is typically reported with a 90% confidence interval.
  • SRMR (Standardized Root Mean Square Residual) — the average standardized discrepancy between the observed and model-implied correlations. Conventional guidance treats values at or below .08 as indicating good fit.

These specific cutoffs (CFI ≥ .95, RMSEA ≤ .06, SRMR ≤ .08) trace back to Hu and Bentler’s widely cited 1999 simulation study in Structural Equation Modeling (Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55), which recommended a “two-index strategy” — reporting CFI together with either RMSEA or SRMR — based on which combinations best distinguished correctly specified from misspecified models across their simulated conditions.

It is important to be explicit about what these numbers are and are not. Hu and Bentler’s simulations used continuous, normally distributed indicators, specific sample sizes, and small-to-moderate models; the cutoffs do not automatically transfer to every combination of estimator, indicator type, model size, and sample size a researcher might encounter. They are widely used conventions born from one influential simulation study, not universal laws of model fit, and methodologists have repeatedly cautioned against treating them as a rigid pass/fail checklist. Our guide to fit indices and the cut-off controversy in structural equation modeling traces exactly where the commonly cited thresholds diverge from the combinational rules Hu and Bentler actually recommended. Chasing a specific cutoff by mechanically adding parameters until CFI crosses .95 is a well-documented way to produce a model that fits the sample but does not reflect a defensible theoretical structure — the same problem addressed directly in the next section.

Why fixed cutoffs are only a starting point

The Hu and Bentler cutoffs were derived from simulations of particular models, so a cutoff that works for one model can be too lenient or too strict for another. McNeish and Wolf (2023, Psychological Methods 28(1), 61–88) addressed this with dynamic fit index cutoffs: instead of one fixed number, the researcher simulates data from the model they actually fitted (same number of factors, items, loadings and sample size) and derives cutoffs specific to that model. They released a Shiny app and an R package named dynamic for this purpose. In practice, report the conventional values for comparability, but justify the verdict with model-specific evidence: standardized loadings, residual correlations and, where feasible, dynamic cutoffs, rather than a single pass/fail comparison against .95, .06 and .08.

Modification Indices and the Risk of Data-Driven Respecification

Most CFA software reports modification indices (MIs) — an estimate of how much the model χ² would drop if a currently fixed parameter (most often a correlated error, or a cross-loading of an item onto a factor it was not specified to belong to) were freed and estimated instead. Because a large MI tells you precisely which single change would most improve fit, it is tempting to add that parameter, re-run the model, and repeat until the fit indices clear their conventional thresholds.

This practice carries a real risk that CASRAI treats as a genuine research-integrity concern, not just a technical caveat: respecifying a model purely because a modification index suggests it, with no theoretical justification for why that specific error correlation or cross-loading should exist, is a form of overfitting. The revised model is now partly fit to sampling-specific noise in that one dataset rather than to a structure that will replicate. A model modified this way and then presented as if it had been confirmatory from the start misrepresents what was actually tested — the same underlying problem as HARKing (hypothesizing after results are known) in other areas of quantitative research.

The defensible practice is: (1) only add a modification that has a substantive, a priori justification (for example, two items that share near-identical wording, which plausibly correlate for reasons beyond the latent factor); (2) disclose every modification made, and why, in the reporting; and (3) treat a model that required substantial post hoc modification as exploratory rather than confirmatory, ideally cross-validated in an independent sample before being relied upon.

Measurement Invariance

Before comparing latent factor means, variances, or structural relationships across groups — different countries, languages, time points, demographic groups, or experimental conditions — a researcher first has to establish that the instrument is measuring the construct the same way in each group. This is measurement invariance (also called measurement equivalence) testing, and it proceeds through a standard, increasingly restrictive sequence of nested models, each compared to the one before it:

  1. Configural invariance — the same factor structure (same number of factors, same pattern of which items load on which factor) holds across groups, with all parameters estimated freely in each group. This is the baseline model against which every later step is compared.
  2. Metric (weak) invariance — factor loadings are constrained to be equal across groups. Establishing metric invariance justifies comparing relationships involving the latent variable (e.g., correlations, regression paths) across groups.
  3. Scalar (strong) invariance — item intercepts are additionally constrained to be equal across groups, on top of the equal loadings from metric invariance. Scalar invariance is the level required to justify comparing latent factor means across groups — without it, an apparent group difference in the observed scores could reflect a difference in how the items function rather than a genuine difference in the underlying construct.
  4. Strict (residual) invariance — item residual (error) variances are additionally constrained to be equal. This is the most restrictive level and is not required for most substantive group comparisons; it is more relevant when the researcher wants to treat observed and latent scores as fully interchangeable across groups.

Each step is typically evaluated by comparing model fit to the previous, less restrictive step, using a χ² difference test and/or a change in CFI (a commonly used rough guideline is a drop in CFI of no more than .01 between nested models, alongside corresponding changes in RMSEA), rather than judging each model’s absolute fit in isolation. This nested-model logic for testing invariance is most closely associated with methodological syntheses such as Vandenberg and Lance’s (2000) review of the practice in organizational research, building on earlier multi-group CFA work. Comparing latent means across groups without first establishing at least scalar invariance is a common and consequential error — it risks attributing a measurement artifact to a substantive group difference.

Change-in-fit criteria used to judge invariance

Published studies decide whether each step holds by looking at how much fit worsens when constraints are added. The most used criteria, as summarized in the Putnick and Bornstein (2016) review of measurement invariance conventions, are:

  • Cheung and Rensvold (2002): a decrease in CFI of more than .01 between nested models signals non-invariance. Later work criticized this rule as liberal.
  • Chen (2007), metric invariance: a CFI decrease of .01 combined with an RMSEA increase of .015 or an SRMR increase of .030 signals non-invariance.
  • Chen (2007), scalar and residual invariance: a CFI decrease of .01 combined with an RMSEA increase of .015 or an SRMR increase of .015. These were proposed for adequate power, equal group sizes and a mix of invariant and non-invariant parameters.
  • Other proposals: Meade and colleagues (2008) suggested a CFI change of .002 together with McDonald's noncentrality index, and Rutkowski and Svetina (2014) proposed more lenient values (CFI .02, RMSEA .03 for metric invariance) when groups are large.

The data table below lists the numeric criteria alongside the Monte Carlo sample sizes and the lavaan reference output used in this guide. When a step fails, partial invariance is the usual fallback: free the loadings or intercepts that differ for specific items and keep the rest constrained, then report which items were freed and why.

How often invariance is actually reached

Putnick and Bornstein reviewed 126 articles containing 269 invariance tests. About 75% of the tests compared two groups, and the median sample size was 725. Full invariance was reached in 94.4% of configural tests, 82.3% of metric tests, 57.8% of scalar tests and 43.6% of residual tests. Scalar invariance, the level needed to compare latent means, therefore failed in roughly four of ten tests, so it should be tested rather than assumed. Reporting was uneven: ΔCFI was reported in 73.2% of tests, a χ² difference test alone in 16.7%, alternative fit indices alone in 34.1% and both in 45.9%, and about 20% of tests omitted the model degrees of freedom. A minimum invariance report gives the total and per-group sample sizes, how missing data were handled, the fit criteria used, and a table listing each model, its fit and the decision taken.

Reliability and Validity from a CFA

A CFA model provides more precise reliability and validity estimates than simpler approaches like Cronbach’s alpha, because it accounts for each item’s actual estimated relationship to the latent factor rather than treating all items as equally good indicators. See CASRAI’s guide to Cronbach’s alpha for how the two approaches differ and why alpha assumes essential tau-equivalence that a CFA does not require.

  • Composite reliability (CR), sometimes called construct reliability, and McDonald’s omega are both computed from the standardized factor loadings and error variances of a CFA model, and are generally preferred over Cronbach’s alpha when a scale’s items are not assumed to be equally strong indicators of the underlying factor.
  • Average variance extracted (AVE) — the average amount of variance a latent factor explains in its own indicators, relative to the variance due to measurement error. AVE is used as a convergent-validity check (a common informal benchmark is AVE ≥ .50, meaning the factor explains more variance in its indicators than error does) and feeds into discriminant-validity tests — CASRAI’s guide to calculating average variance extracted and applying the 0.50 threshold works through the arithmetic and the discriminant-validity tests that consume it.
  • Discriminant validity asks whether factors that are supposed to be distinct constructs are, in fact, statistically distinguishable from one another. The traditional test is the Fornell-Larcker criterion (Fornell & Larcker, 1981): a factor’s AVE should exceed the squared correlation between that factor and any other factor in the model. A more recently developed and increasingly used alternative is the heterotrait-monotrait ratio of correlations (HTMT) (Henseler, Ringle, & Sarstedt, 2015), which several methodological studies have found detects discriminant validity problems the Fornell-Larcker criterion misses, particularly with highly correlated factors.

See CASRAI’s guide to types of validity in research for how construct validity, established in part through CFA, relates to the broader family of validity concepts a study needs to address.

Sample Size

CFA, like other SEM techniques, is a large-sample method, and no single sample size is correct for every model. Commonly cited heuristics include 100 or 200 cases, 5 or 10 cases per estimated parameter, or 10 cases per variable. Wolf, Harrington, Clark and Miller (Educational and Psychological Measurement) examined these rules with Monte Carlo simulations of power, bias and solution propriety and found that they are not model-specific: the minimum sample size they found ranged from 30 to 460 depending on the model. Actual power to detect misspecification and the stability of estimates depend on the number of indicators per factor, the size of the loadings, the number of factors, the estimator and the amount of missing data.

For the models in that study, the minimum sample sizes were as follows (the data table below repeats these values with their source):

  • One factor, four indicators: 190, 90 and 60 cases for loadings of .50, .65 and .80.
  • One factor, six indicators: 90, 60 and 40 cases for the same three loading sizes.
  • One factor, eight indicators: 90 cases at loadings of .50 and 50 cases at .65.
  • Two factors, three indicators each: 460, 200 and 120 cases for loadings of .50, .65 and .80.

Two patterns matter for planning. Weak loadings are costly: on average, models with loadings of .50 needed roughly 2.5 times the sample of models with loadings of .80. And more indicators per factor help, though the benefit levels off at about six to eight. Where your model differs from those simulated, run a Monte Carlo analysis for your own model rather than citing a generic N or N-to-parameter ratio.

Worked Example: A Three-Factor CFA in R (lavaan)

The lavaan package documentation walks through a classic three-factor CFA on the Holzinger and Swineford (1939) mental-ability data, where nine test scores are hypothesized to reflect three latent abilities. The model is written in lavaan syntax, with =~ meaning “is measured by”:

HS.model <- ' visual  =~ x1 + x2 + x3
              textual =~ x4 + x5 + x6
              speed   =~ x7 + x8 + x9 '
fit <- cfa(HS.model, data = HolzingerSwineford1939)
summary(fit, fit_measures = TRUE)

According to the lavaan CFA tutorial, cfa() scales each factor by fixing its first indicator's loading to 1, adds residual variances automatically, and lets the latent factors correlate by default. The tutorial's output for this model uses maximum likelihood on 301 observations and reports the following:

Statistic Value Reading against common conventions
χ² (df = 24) 85.306, p < .001 Significant, which is common at this sample size
CFI 0.931 Below the .95 convention
TLI 0.896 Below the .95 convention
RMSEA (90% CI) 0.092 (0.071–0.114) Above the .06 convention
SRMR 0.065 Within the .08 convention

Read together, these indices show a mixed picture: SRMR is acceptable, while CFI, TLI and RMSEA fall short of the conventional cutoffs. That is a realistic outcome. It signals that the three-factor structure only approximately reproduces the data, and the researcher should examine the standardized loadings and residuals and consider a theoretically justified alternative, rather than mechanically freeing parameters as warned in the modification-index section above.

Other lines of the same lavaan output help check the table above: the model has 21 freely estimated parameters, the baseline (null) model has χ² = 918.852 on 36 degrees of freedom (the comparison model behind CFI and TLI), and the test of close fit gives p(RMSEA ≤ .05) = .001, which rejects close fit in agreement with the RMSEA of .092.

How to Read CFA Output: A Checklist

  1. Check that the model ran properly. Confirm the estimator, the number of observations used, the degrees of freedom (above zero) and that the software reports convergence without negative variances.
  2. Read the fit indices together. Compare CFI, TLI, RMSEA (with its confidence interval) and SRMR with the conventions above, and note when they disagree, as in the lavaan example where SRMR passes and CFI and RMSEA do not.
  3. Inspect the standardized loadings. Weak loadings reduce reliability and, as the sample-size section shows, raise the sample needed.
  4. Look at residuals and modification indices. Use them to understand where the model fails, and change the model only for substantive reasons you disclose.
  5. Test invariance before group comparisons, using the change-in-fit criteria above.
  6. Report everything listed in the reporting section, including the estimator, sample size and missing-data handling.

What Does CFA Mean in Research and Statistics?

In research methodology, CFA is the validation step of the measurement workflow: items are drafted, structure is explored (see exploratory factor analysis), and CFA then checks that structure in an independent sample. In statistics, it is the measurement part of structural equation modeling, and it works from the item covariance matrix. When the researcher goes on to add regression paths between latent variables, the model becomes a full SEM, a step related to path analysis. CFA is also the usual tool for the ordinal items produced by a Likert scale, which is why the estimator choice (WLSMV or robust ML) matters.

Reporting a CFA

A complete, reproducible CFA report typically includes:

  • The full model specification — which items load on which factors, which factors are allowed to correlate, and any correlated error terms — ideally as a path diagram or an explicit list, not just prose.
  • The estimator used (ML, MLR, WLSMV, etc.) and the software and version.
  • Sample size and how missing data were handled.
  • All conventional fit indices together (χ² and df, CFI, TLI, RMSEA with its confidence interval, SRMR), not a single favorable statistic selected after the fact.
  • Standardized factor loadings for every indicator, and composite reliability/AVE if reported.
  • Any modification made to the originally specified model, with the justification for it, reported transparently as noted above — not silently incorporated into a “final” model presented as if it had been the a priori specification.

Software

  • R (lavaan) — a free, widely used, and well-documented package for CFA and SEM in R, with syntax that closely mirrors standard SEM notation and native support for ML, MLR, and WLSMV estimation.
  • Mplus — a commercial, specialist SEM program with broad support for categorical, count, and mixture modeling extensions beyond standard CFA, widely used in psychology and education research.
  • SPSS AMOS — a commercial add-on to SPSS with a graphical path-diagram interface, commonly used where a point-and-click SEM tool is preferred over syntax-based specification.

Frequently Asked Questions

What is the difference between confirmatory factor analysis and exploratory factor analysis?

EFA discovers a factor structure from the data with no fixed prior specification; CFA tests whether a specific, previously specified factor structure fits new data. EFA is typically used earlier, during scale development; CFA is used to confirm a structure in a new sample or context, and the two should not be run on the same sample.

What are good CFA fit indices?

The most commonly cited conventions, from Hu and Bentler (1999), are CFI at or above .95, RMSEA at or below .06, and SRMR at or below .08 — reported together rather than relying on any single index, and treated as guidelines rather than fixed pass/fail rules.

Why shouldn’t I chase modification indices until my model fits?

Adding parameters purely because a modification index flags them, with no theoretical justification, fits the model to noise specific to that one sample rather than to a structure likely to replicate. Any such change should be theoretically justified, disclosed, and treated as exploratory rather than confirmatory.

What is measurement invariance and why does it matter?

Measurement invariance testing establishes whether an instrument measures a construct the same way across groups, through a sequence of increasingly restrictive models (configural, metric, scalar, strict). At minimum, scalar invariance is required before comparing latent factor means across groups, since without it an apparent group difference could reflect how the items function rather than a real difference in the underlying construct.

Can I use Cronbach’s alpha instead of a CFA-based reliability estimate?

Alpha is a simpler, widely reported statistic, but it assumes all items are equally strong indicators of the same construct (essential tau-equivalence) and does not itself confirm unidimensionality. Composite reliability and omega, both computed from a CFA’s factor loadings, do not require that assumption and are generally preferred once a CFA has been run. See CASRAI’s Cronbach’s alpha guide for the full comparison.

What sample size do I need for a CFA?

Common heuristics suggest a minimum of roughly 200 observations, or 10-20 observations per estimated parameter, but these are rough starting points, not formal power calculations. Actual requirements depend on the number of factors and indicators, loading strength, and estimator used.

Are Hu and Bentler's cutoffs still valid?

They remain the most cited conventions, but they came from simulations of specific models. McNeish and Wolf proposed dynamic, model-specific cutoffs (R package dynamic) as a more defensible alternative.

What ΔCFI shows measurement invariance?

A CFI decrease of no more than .01 (Cheung and Rensvold, 2002), ideally with ΔRMSEA of .015 or less and ΔSRMR of .030 or less for metric invariance and .015 or less for scalar invariance (Chen, 2007).

How often is scalar invariance achieved?

In the 269 tests reviewed by Putnick and Bornstein (2016), full scalar invariance was reached in 57.8% of tests, compared with 82.3% for metric invariance.

Can I proceed with partial invariance?

Yes: free the specific loadings or intercepts that differ, keep the others constrained, and report which items were freed and why.

What is the difference between CFA and SEM?

CFA is the measurement part of SEM; a full SEM adds regression paths between the latent variables.

Confirmatory factor analysis sits within CASRAI’s broader research methods coverage. See also Cronbach’s alpha and types of validity in research for the reliability and validity concepts a CFA is often used to establish; correlation coefficient and descriptive statistics for the foundational statistics a CFA builds on; regression analysis, which shares much of its underlying logic with the factor model; and questionnaire design for how the items that go into a CFA are constructed in the first place.

Sources

Sources (accessed 2026-11-14):

Data: CFA benchmarks from simulation and review studies: minimum sample sizes, invariance cutoffs and invariance attainment

Compiled from three sources: a Monte Carlo sample-size study (Wolf et al.), a review of 269 published measurement-invariance tests (Putnick and Bornstein 2016, which also reports the Chen 2007 and Cheung and Rensvold 2002 change-in-fit criteria), and the lavaan CFA tutorial. Each row is one benchmark with its value and the study it comes from.

Period covered: 2002/2023.

CFA benchmarks from simulation and review studies: minimum sample sizes, invariance cutoffs and invariance attainment
BenchmarkConditionValueSource
Minimum N, one factor (cases)4 indicators, loadings .50190Wolf et al.
Minimum N, one factor (cases)4 indicators, loadings .6590Wolf et al.
Minimum N, one factor (cases)4 indicators, loadings .8060Wolf et al.
Minimum N, one factor (cases)6 indicators, loadings .5090Wolf et al.
Minimum N, one factor (cases)6 indicators, loadings .6560Wolf et al.
Minimum N, one factor (cases)6 indicators, loadings .8040Wolf et al.
Minimum N, two factors (cases)3 indicators each, loadings .50460Wolf et al.
Minimum N, two factors (cases)3 indicators each, loadings .65200Wolf et al.
Minimum N, two factors (cases)3 indicators each, loadings .80120Wolf et al.
Change-in-fit cutoff, CFI (change in CFI)Nested invariance models (Cheung and Rensvold 2002)-0.01Putnick and Bornstein 2016
Change-in-fit cutoff, RMSEA (change in RMSEA)Metric invariance (Chen 2007)0.015Putnick and Bornstein 2016
Change-in-fit cutoff, SRMR (change in SRMR)Metric invariance (Chen 2007)0.03Putnick and Bornstein 2016
Change-in-fit cutoff, SRMR (change in SRMR)Scalar or residual invariance (Chen 2007)0.015Putnick and Bornstein 2016
Full invariance reached (% of tests)Configural, 269 tests94.4Putnick and Bornstein 2016
Full invariance reached (% of tests)Metric, 220 tests82.3Putnick and Bornstein 2016
Full invariance reached (% of tests)Scalar, 232 tests57.8Putnick and Bornstein 2016
Full invariance reached (% of tests)Residual, 110 tests43.6Putnick and Bornstein 2016
Reference output, CFI (index)lavaan HolzingerSwineford1939, N=301, ML0.931lavaan tutorial
Reference output, RMSEA (index)lavaan HolzingerSwineford1939, N=301, ML0.092lavaan tutorial
Reference output, SRMR (index)lavaan HolzingerSwineford1939, N=301, ML0.065lavaan tutorial

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about Confirmatory Factor Analysis (CFA): A Complete Guide

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →