Skip to contents

Purpose

Scale development has a long methodological history. Many techniques that graduate students meet in published articles were reasonable responses to the computing and theory of their time, and several have since been refined or replaced. This article maps each nomologR workflow stage from historical practice to contemporary practice, states what nomologR does, and notes what is planned.

How to read the labels:

  • Historical practice — widely used earlier and still common in published work. nomologR shows these methods as context when learners need to recognize them, labels them, and never silently substitutes them for contemporary evidence. “Historical” is not the same as “always wrong.”
  • Contemporary practice — what the current methodological literature recommends. These methods provide nomologR’s primary evidence.
  • Planned — work accepted for a future release, linked to its issue.

Full references appear at the end of this article and in each function’s help page.

The methods registry

The same mapping is available as data. nomo_methods() returns one row per method the package implements, with its stage, lineage, role, estimand, key assumptions, implementing function, computational engine, and references.

methods <- nomo_methods()
nrow(methods)
#> [1] 89
table(methods$stage, methods$lineage)
#>              
#>               contemporary emerging historical
#>   cfa                   10        0          3
#>   compare                6        0          0
#>   efa                    3        0          2
#>   factors                7        1          4
#>   invariance             6        0          0
#>   network                4        0          1
#>   reliability            9        0          2
#>   scores                 4        0          3
#>   screen                10        0          1
#>   validity               3        0          2
#>   workflow               6        0          2

The role column records how the evidence is used. A context method is shown so that a reader can recognize it in published work, and it takes no part in any synthesis or decision:

nomo_methods(lineage = "historical")[, c("stage", "method", "role")]
#> # A tibble: 20 × 3
#>    stage       method                                               role      
#>    <chr>       <chr>                                                <chr>     
#>  1 screen      Fixed item-total correlation reference (about .30)   context   
#>  2 factors     Eigenvalue-greater-than-one rule                     context   
#>  3 factors     Scree test                                           context   
#>  4 factors     Kaiser-Meyer-Olkin sampling adequacy                 supporting
#>  5 factors     Bartlett's test of sphericity                        supporting
#>  6 efa         Orthogonal (varimax) rotation                        context   
#>  7 efa         Fixed loading cutoff                                 context   
#>  8 cfa         Chi-square exact-fit test                            supporting
#>  9 cfa         Fixed fit-index cutoffs                              context   
#> 10 cfa         Modification indices                                 context   
#> 11 reliability Coefficient alpha                                    supporting
#> 12 reliability Schmid-Leiman decomposition                          supporting
#> 13 validity    Standardized loadings and average variance extracted supporting
#> 14 validity    Fornell-Larcker comparison                           context   
#> 15 scores      Unit-weighted sum or mean score                      primary   
#> 16 scores      Regression (Thurstone) factor scores                 primary   
#> 17 scores      Bartlett factor scores                               primary   
#> 18 network     Nomological network of construct relations           primary   
#> 19 workflow    Listwise deletion                                    context   
#> 20 workflow    Pairwise deletion                                    context

The registry lists only what the package actually computes. Methods that are planned but not yet implemented appear in this article and in the issue tracker, not in the registry, so that nomo_methods() never describes a capability the package lacks.

Given a workflow object, nomo_methods(run) returns only the methods that workflow used. For any selection, references = TRUE expands the result into a reference list:

nomo_methods(stage = "reliability", references = TRUE)[, c("method", "citation")]
#> # A tibble: 24 × 2
#>    method                                         citation                      
#>    <chr>                                          <chr>                         
#>  1 Model-based coefficient omega                  Dunn, T. J., Baguley, T., & B…
#>  2 Model-based coefficient omega                  McNeish, D. (2018). Thanks co…
#>  3 Model-based coefficient omega                  Flora, D. B. (2020). Your coe…
#>  4 Model-based coefficient omega                  Bell, S. M., Chalmers, R. P.,…
#>  5 Reliability on the ordered-score scale         Green, S. B., & Yang, Y. (200…
#>  6 Coefficient alpha                              Cronbach, L. J. (1951). Coeff…
#>  7 Coefficient alpha                              Sijtsma, K. (2009). On the us…
#>  8 Bootstrap confidence intervals for reliability Kelley, K., & Pornprasertmani…
#>  9 Omega hierarchical                             Reise, S. P. (2012). The redi…
#> 10 Omega hierarchical                             Reise, S. P., Bonifay, W. E.,…
#> # ℹ 14 more rows

This is what the Methods and citations section of nomo_report() is built from, so an archived report cites the methods a run used rather than every method the package knows about. Every DOI in the registry is checked against the DOI registries before each release, including that the registered author, year, and title match the citation.

How practice changed

The registry also records when each method entered the literature, and, for a historical method, which contemporary methods now address the question it answered. The tables below are drawn from nomo_methods(), so they cannot drift from what the package computes.

A method is dated only from the publication that introduced it, and only when the registry cites that publication. 43 of the 89 methods are not yet dated, because the registry cites a later review or critique instead of the original. Adding those originating references is ongoing work.

dated <- methods[!is.na(methods$introduced), , drop = FALSE]
dated <- dated[order(dated$introduced, dated$stage), , drop = FALSE]
knitr::kable(
  dated[, c("introduced", "method", "stage", "lineage")],
  col.names = c("Year", "Method", "Stage", "Lineage"),
  row.names = FALSE
)
Year Method Stage Lineage
1937 Bifactor measurement model cfa contemporary
1950 Bartlett’s test of sphericity factors historical
1951 Coefficient alpha reliability historical
1954 Eigenvalue-greater-than-one rule factors historical
1955 Nomological network of construct relations network historical
1957 Schmid-Leiman decomposition reliability historical
1958 Orthogonal (varimax) rotation efa historical
1965 Common-factor parallel analysis factors contemporary
1966 Scree test factors historical
1969 Maximum-likelihood confirmatory factor analysis cfa contemporary
1969 Chi-square exact-fit test cfa historical
1970 Kaiser-Meyer-Olkin sampling adequacy factors historical
1971 Multiple-group confirmatory factor analysis invariance contemporary
1974 Information criteria (AIC, BIC) compare contemporary
1976 Velicer original minimum average partial (MAP) factors contemporary
1980 Incremental fit indices (CFI, TLI) cfa contemporary
1981 Standardized loadings and average variance extracted validity historical
1981 Fornell-Larcker comparison validity historical
1987 Equivalence testing against a smallest effect size of interest network contemporary
1988 Measurement-then-structure modeling network contemporary
1989 Partial invariance with documented releases invariance contemporary
1992 RMSEA with confidence interval cfa contemporary
1993 Configural, metric, scalar, and strict sequence invariance contemporary
1999 Fixed fit-index cutoffs cfa historical
2000 Scaled-and-shifted difference test compare contemporary
2000 Velicer revised MAP (fourth powers) factors contemporary
2001 Satorra-Bentler scaled difference test compare contemporary
2001 Construct replicability (H) reliability contemporary
2002 Change-in-fit indices compare contemporary
2007 Change-in-fit evidence across levels invariance contemporary
2009 Reliability on the ordered-score scale reliability contemporary
2010 Formal model-nesting check compare contemporary
2011 Hull method factors contemporary
2012 Comparison data factors contemporary
2012 Percentage of uncontaminated correlations reliability contemporary
2013 Omega hierarchical subscale reliability contemporary
2015 Inter-item standard deviation screen contemporary
2015 Heterotrait-monotrait ratio validity contemporary
2016 Identification-aware sequence for ordered indicators invariance contemporary
2016 Bootstrap confidence intervals for reliability reliability contemporary
2017 Empirical Kaiser criterion factors contemporary
2017 Next Eigenvalue Sufficiency Test (NEST) factors emerging
2018 A-priori versus post-hoc prediction provenance network contemporary
2020 Test of the constraints unit weighting assumes scores contemporary
2021 HTMT2 validity contemporary
2022 Latent correlations with confidence intervals validity contemporary

Read by era, the timeline tells a story of methods shaped by the computing of their time and then revisited.

  • 1930s to 1950s. The field laid its foundations: the bifactor model (Holzinger & Swineford, 1937), coefficient alpha (Cronbach, 1951), the nomological network (Cronbach & Meehl, 1955), and the Schmid-Leiman transformation (1957). Factor analysis was computed by hand, and rules that were easy to apply, such as retaining factors with eigenvalues greater than one (Guttman, 1954; Kaiser, 1960), became habits.
  • 1960s to 1980s. Structural modeling arrived. Horn (1965) proposed parallel analysis and Cattell (1966) the scree test. Jöreskog made confirmatory factor analysis practical (1969) and extended it to several groups (1971). Researchers gained information criteria (Akaike, 1974), incremental fit (Bentler & Bonett, 1980), the Fornell-Larcker comparison (1981), and the two-step logic of measurement before structure (Anderson & Gerbing, 1988).
  • 1990s and 2000s. Fit and invariance were formalized: an interval for the RMSEA (Browne & Cudeck, 1992), the invariance hierarchy (Meredith, 1993), the fit-index cutoffs that became conventions (Hu & Bentler, 1999), scaled difference tests (Satorra, 2000; Satorra & Bentler, 2001), and change-in-fit criteria (Cheung & Rensvold, 2002; Chen, 2007).
  • 2010s to the present. Many defaults were reconsidered: better factor-retention criteria (Hull, comparison data, the empirical Kaiser criterion, NEST), bifactor indices (Reise, 2012), heterotrait-monotrait ratios (Henseler et al., 2015; Roemer et al., 2021), identification for ordered indicators (Wu & Estabrook, 2016), the cost of sum scores (McNeish & Wolf, 2020), and latent correlations with intervals as discriminant evidence (Rönkkö & Cho, 2022).

What took over from each historical method:

old <- methods[!is.na(methods$contemporary_practice), , drop = FALSE]
now <- vapply(
  strsplit(old$contemporary_practice, "; ", fixed = TRUE),
  function(ids) paste(methods$method[match(ids, methods$id)], collapse = "; "),
  character(1)
)
knitr::kable(
  data.frame(Year = ifelse(is.na(old$introduced), "", old$introduced),
             Historical = old$method, Now = now),
  row.names = FALSE
)
Year Historical Now
Fixed item-total correlation reference (about .30) Corrected item-rest correlation
1954 Eigenvalue-greater-than-one rule Common-factor parallel analysis; Velicer revised MAP (fourth powers); Empirical Kaiser criterion
1966 Scree test Common-factor parallel analysis
1958 Orthogonal (varimax) rotation Oblique (oblimin) rotation
Fixed loading cutoff Cross-loading, communality, and residual diagnostics
1969 Chi-square exact-fit test Incremental fit indices (CFI, TLI); RMSEA with confidence interval; Standardized root mean square residual; Localized residual correlations
1999 Fixed fit-index cutoffs Localized residual correlations
Modification indices Localized residual correlations; Recorded revision lineage
1951 Coefficient alpha Model-based coefficient omega
1957 Schmid-Leiman decomposition Bifactor measurement model
1981 Fornell-Larcker comparison HTMT2; Latent correlations with confidence intervals
Unit-weighted sum or mean score Test of the constraints unit weighting assumes
1955 Nomological network of construct relations Measurement-then-structure modeling; A-priori versus post-hoc prediction provenance
Listwise deletion Full-information maximum likelihood for missing data; Missing-data sensitivity comparison
Pairwise deletion Full-information maximum likelihood for missing data; Missing-data sensitivity comparison

“Now” does not mean that the historical method is wrong. It means that contemporary work answers the same question differently, usually with evidence the historical method could not provide. nomologR shows the historical method as labeled context where a reader will meet it, and bases its own evidence on the contemporary one.

1. Item and data audit — nomo_screen()

Historical practice. Item analysis often meant screening corrected item–total correlations against a fixed value (a reference near .30 is widely taught; see Nunnally & Bernstein, 1994) and deleting items that fell below it, with little attention to response distributions, missing data, or response quality.

Contemporary practice. Item statistics are one strand of evidence, interpreted alongside content, construct definition, and later structural evidence (Clark & Watson, 2019). Audits describe missingness, category use, and response concentration before modeling. Case-level screening for careless or insufficient-effort responding uses several complementary indices rather than a single exclusion rule (Meade & Craig, 2012; Huang et al., 2012; Curran, 2016).

What nomologR does. Reports item and case missingness, response distributions, unused categories, response concentration, near-zero variance (using heuristics described by Kuhn & Johnson, 2013), corrected item–rest and inter-item correlations, and descriptive shape statistics. Flags are review prompts; no item or case is removed or recoded. With effort = TRUE it adds the careless-responding indices those sources describe (long-string, inter-item standard deviation, Mahalanobis distance, even-odd consistency, and psychometric antonyms and synonyms). They are reported side by side because they detect different failures, and a case is flagged only where a source states a rule.

Items can arrive from content review: a handoff from contentvalidR supplies the carried items and their construct mapping. Held-back items are reported in that package’s own words, and the log records that item membership came from content review rather than from these data (#46).

2. Dimensionality — nomo_factors()

Historical practice. Retaining factors with eigenvalues greater than one (Guttman, 1954; Kaiser, 1960) and visually inspecting the scree plot (Cattell, 1966). The “Little Jiffy” routine — principal components, eigenvalues greater than one, and varimax rotation — became a default in software (discussed by Kaiser, 1970).

Contemporary practice. Simulation research favors parallel analysis (Horn, 1965; Crawford et al., 2010) and Velicer’s minimum average partial criterion (Velicer, 1976; Velicer et al., 2000), with newer criteria such as the empirical Kaiser criterion (Braeken & van Assen, 2017), comparison data (Ruscio & Roche, 2012), NEST (Achim, 2017), and the Hull method (Lorenzo-Seva et al., 2011). Criteria are triangulated, and the correlation model matches the measurement level of the items.

What nomologR does. Common-factor parallel analysis is the primary evidence, complemented by original and revised MAP and EKC; NEST, Hull, and comparison data are optional. The eigenvalue-greater-than-one rule is displayed only as labeled historical context and is excluded from the synthesis. KMO (Kaiser, 1974) and Bartlett’s test (Bartlett, 1950) are supporting diagnostics, not decision rules. Pearson, polychoric, tetrachoric, or mixed correlations are chosen explicitly.

3. Exploratory structure — nomo_efa()

Historical practice. Principal components analysis reported as “factor analysis,” orthogonal varimax rotation by default (Kaiser, 1958), and deleting items below loading cutoffs. Reviews documented how common these choices were (Fabrigar et al., 1999; Conway & Huffcutt, 2003).

Contemporary practice. Common-factor extraction, oblique rotation whenever factors may plausibly correlate (Browne, 2001; Fabrigar et al., 1999), attention to cross-loadings, communalities, and residual correlations, and item decisions that combine numerical evidence with content (Costello & Osborne, 2005; Watkins, 2018).

What nomologR does. MINRES common-factor extraction with oblimin rotation by default; pattern and structure matrices; communalities; cross-loading and residual diagnostics; KEEP, REVIEW, or STRONG REVIEW item guidance with explanations. Orthogonal rotation is allowed but recorded as a choice that needs justification.

4. Confirmatory measurement model — nomo_cfa()

Historical practice. Treating the chi-square exact-fit test as the sole criterion (Jöreskog, 1969); judging incremental fit indices against .90 (Bentler & Bonett, 1980); later, treating the cutoffs of Hu and Bentler (1999) — CFI and TLI near .95, RMSEA near .06, SRMR near .08 — as universal golden rules; and freeing parameters suggested by modification indices until a model “fits.”

Contemporary practice. Fit indices are interpreted in the context of model size, estimator, sample size, and indicator type rather than as universal rules (Marsh et al., 2004). RMSEA is reported with its confidence interval (Browne & Cudeck, 1992). Local strain in residuals is inspected. Estimators match the indicators — for example WLSMV for ordered categories (Flora & Curran, 2004; Rhemtulla et al., 2012). Data-driven respecification is recognized as capitalization on chance (MacCallum et al., 1992). Improper solutions such as Heywood cases are treated as possible signs of misspecification (Kolenikov & Bollen, 2012). Model-specific (“dynamic”) fit-index cutoffs are an emerging alternative to fixed thresholds (McNeish & Wolf, 2023).

What nomologR does. Retains the unchanged lavaan fit (Rosseel, 2012) and reports chi-square, CFI (Bentler, 1990), TLI, RMSEA with its interval, and SRMR against teaching references; localized residual correlations; Heywood and convergence diagnostics; and modification indices quarantined as post-hoc diagnostics that never change the model. Declared ordered indicators request WLSMV. nomo_split() supports calibration and validation samples because evaluating an exploratory solution on the same data it came from overstates confirmation (Fokkema & Greiff, 2017).

Comparing models. Historically, competing models were compared by subtracting chi-square values and degrees of freedom, even for robust estimators, or by applying one fixed change-in-fit cutoff. Contemporary practice uses the difference test that matches the estimator — scaled differences for robust ML (Satorra & Bentler, 2001, 2010) and scaled-and-shifted tests for categorical estimators (Satorra, 2000) — verifies nesting rather than assuming it (Bentler & Satorra, 2010), reports several changes in fit (Cheung & Rensvold, 2002; Chen, 2007), and compares non-nested models of the same data with information criteria (Akaike, 1974; Schwarz, 1978; Raftery, 1995; Burnham & Anderson, 2004). nomo_compare() implements this: it requires a recorded rationale, labels post-hoc comparisons, refuses comparisons across different estimators, cases, or data, and never selects a model automatically.

Revising a model. Historically, a model that fit poorly was respecified in place, and the published model was reported as though it had been the plan. nomo_revise() instead creates a child workflow from the parent run, records what changed, the researcher’s rationale, and whether the change was prespecified or post hoc, compares the parent and revised models, and recommends confirming a post-hoc revision in independent data.

Planned. Model-specific fit cutoffs remain a research proposal (#23).

5. Reliability — nomo_reliability()

Historical practice. Reporting coefficient alpha (Cronbach, 1951) as “the” reliability of a scale regardless of whether its assumptions hold.

Contemporary practice. Model-based omega for congeneric measurement (Dunn et al., 2014; McNeish, 2018; Flora, 2020), recognition of alpha’s limits (Sijtsma, 2009), reliability defined on the score scale actually used for ordered items (Green & Yang, 2009), interval estimates (Kelley & Pornprasertmanit, 2016), and awareness that measurement misspecification biases omega (Bell et al., 2024).

What nomologR does. Model-based omega from semTools is primary; alpha is a qualified secondary statistic and is reported as unavailable rather than silently redefined for ordered-score estimands; bootstrap intervals are optional.

Total and subscale scores. Historically, a scale with subscales was often summarized by coefficient alpha for its multidimensional total score. The bi-factor method (Holzinger & Swineford, 1937) and the Schmid-Leiman orthogonalization of higher-order exploratory solutions (Schmid & Leiman, 1957) long predate their wide use. Contemporary practice fits confirmatory bifactor and higher-order models and evaluates them with model-based indices: omega hierarchical for the total score, omega hierarchical subscale for what each subscale adds beyond the general factor, explained common variance, and the percentage of uncontaminated correlations (Reise, 2012; Reise, Bonifay, & Haviland, 2013; Rodriguez, Reise, & Haviland, 2016). A bifactor model will usually fit at least as well as the alternatives even when it did not generate the data (Reise, 2012), and a higher-order model is a constrained bifactor model (Yung, Thissen, & McLeod, 1999), so fit alone cannot choose between them. nomo_model() writes all three structures, nomo_hierarchical() computes the indices with their estimands stated and no pass/fail thresholds, and nomo_compare() compares the structures; see Total and subscale scores.

6. Convergent and discriminant evidence — nomo_validity()

Historical practice. Multitrait–multimethod correlation matrices (Campbell & Fiske, 1959); later, average variance extracted above .50 and the Fornell–Larcker comparison of AVE with squared correlations (Fornell & Larcker, 1981).

Contemporary practice. Simulation work shows that the Fornell–Larcker comparison can miss discriminant-validity problems (Henseler et al., 2015; Voorhees et al., 2016). The heterotrait–monotrait ratio (Henseler et al., 2015) and its congeneric-appropriate successor HTMT2 (Roemer et al., 2021) are recommended, together with latent factor correlations and their confidence intervals (Rönkkö & Cho, 2022).

What nomologR does. Standardized loadings and AVE as convergent evidence (never as reliability); HTMT2 as primary construct-separation evidence with original HTMT for comparison; latent correlations with uncertainty; the Fornell–Larcker table only on explicit request, labeled legacy/supporting.

7. Measurement invariance — nomo_invariance(), nomo_partial()

Historical practice. Multiple-group factor analysis (Jöreskog, 1971) and the configural–metric–scalar–strict hierarchy (Meredith, 1993; Vandenberg & Lance, 2000), evaluated with chi-square difference tests alone or a single fixed rule such as a CFI decrease of .01 (Cheung & Rensvold, 2002); ordinal items analyzed as if continuous.

Contemporary practice. Several change-in-fit indices interpreted with sample size and model context (Chen, 2007; Putnick & Bornstein, 2016); identification-aware sequences for ordered-categorical indicators (Wu & Estabrook, 2016; Svetina et al., 2020); partial invariance based on substantively justified, transparently reported releases (Byrne et al., 1989).

What nomologR does. Category-aware sequences for continuous, binary, three-category, and four-or-more-category indicators; fit and change evidence without a universal pass/fail rule; score-test diagnostics that locate strain but never free parameters; researcher-specified releases with required rationales carried forward to more restrictive levels.

Planned. Longitudinal invariance is a candidate for a 1.x release after 1.0.0. The reasons for deferring it are recorded in #38 and #113.

8. Nomological network — nomo_hypotheses(), nomo_network()

Historical practice. Construct validity articulated as a nomological network of lawful relations (Cronbach & Meehl, 1955), in practice often examined through tables of observed correlations and significance tests, with a non-significant result taken as evidence of “no relation.”

Contemporary practice. Establish the measurement model before interpreting structural relations (Anderson & Gerbing, 1988); treat validity as an integrated argument about the meaning of scores (Messick, 1995); test negligible predictions with equivalence procedures and a smallest effect size of interest rather than p > .05 (Schuirmann, 1987; Lakens et al., 2018); and distinguish prespecified from post-hoc predictions (Nosek et al., 2018).

What nomologR does. Machine-readable directional, magnitude, and negligible predictions; researcher-specified equivalence regions (the package never invents one); latent SEM with observed outcomes; measurement context kept alongside theory concordance; a-priori versus post-hoc provenance; and replication of the exact same model in validation data. A change in sign between samples is called a reversal only when both confidence intervals exclude zero on opposite sides; otherwise it is reported as a direction that did not replicate, or as a sign change within sampling uncertainty, because estimates scattered around a null relation differ in sign about half the time.

A hypothesis whose endpoints are observed variables, such as sum or factor scores, is disclosed in the decision log, because the measurement error of a composite enters the relationship unmodeled. The disclosure points to modeling the items as indicators, to estimating structural relationships after the measurement model with lavaan::sam() (Rosseel & Loh, 2024), and, for a linear regression among factor scores, to the one scoring design shown to give consistent coefficients: regression-method scores for the predictors and Bartlett scores for the outcome, each block scored from a measurement model of its own (Skrondal & Laake, 2001). None of these is applied automatically.

9. Workflow, scores, and reporting — nomo_run(), nomo_report()

Historical practice. A sequence of undisclosed decisions — dropping items, adding residual covariances, trying estimators — reported as if it had been the plan all along; unit-weighted sum scores assumed to represent the construct; listwise deletion, pairwise deletion, or mean substitution applied without asking whether the data meet the assumption each requires.

Contemporary practice. Transparent reporting of measurement decisions and their justification (Flake et al., 2017; Flake & Fried, 2020), awareness of researcher degrees of freedom (Simmons et al., 2011; Wicherts et al., 2016), staged scale-development guidance (Clark & Watson, 1995, 2019; Hinkin, 1998; Boateng et al., 2018), missing-data methods such as full-information maximum likelihood (Enders & Bandalos, 2001; Schafer & Graham, 2002), and checking whether sum scores are consistent with the measurement model before using them (Grice, 2001; McNeish & Wolf, 2020).

What nomologR does. nomo_run() pauses at consequential decisions and records the decision and rationale; nomo_report() archives methods, evidence, decisions, deviations, and session information. nomo_methods() identifies which methods a workflow actually used, so the report cites those methods and their literature rather than only the software that computed them. nomo_scores() fits the parallel model a sum score assumes and reports Grice’s criteria for the scores it returns, and nomo_apa_table() formats evidence as manuscript tables following APA style (American Psychological Association, 2020).

nomo_missing() refits a prespecified measurement model or network under listwise deletion and FIML, or under listwise and pairwise deletion for ordered indicators, and reports where estimates, reliability, or theory evidence differ. Listwise and pairwise deletion require data missing completely at random; FIML requires data missing at random, and was unbiased under both in Enders and Bandalos’s (2001) simulations while the deletion methods were biased under MAR. A difference beyond half the reference standard error is flagged, following Schafer and Graham’s (2002) rule for when bias becomes practically important, with the caveat that it estimates bias only if the data are MAR. Because MAR cannot in general be tested from the data at hand (Schafer & Graham, 2002), agreement between strategies is reported as insensitivity, not as evidence that either is unbiased. Mean substitution is explained and not offered.

Planned. Multiple-imputation integration is a candidate for a 1.x release after 1.0.0. The reasons for deferring it are recorded in #38 and #113.

References

Every DOI below was verified against Crossref or the DOI registry.

Achim, A. (2017). Testing the number of required dimensions in exploratory factor analysis. The Quantitative Methods for Psychology, 13(1), 64–74. https://doi.org/10.20982/tqmp.13.1.p064

Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19(6), 716–723. https://doi.org/10.1109/TAC.1974.1100705

American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.). https://doi.org/10.1037/0000165-000

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411–423. https://doi.org/10.1037/0033-2909.103.3.411

Bartlett, M. S. (1950). Tests of significance in factor analysis. British Journal of Statistical Psychology, 3(2), 77–85. https://doi.org/10.1111/j.2044-8317.1950.tb00285.x

Bell, S. M., Chalmers, R. P., & Flora, D. B. (2024). The impact of measurement model misspecification on coefficient omega estimates of composite reliability. Educational and Psychological Measurement, 84(1), 5–39. https://doi.org/10.1177/00131644231155804

Beauducel, A. (2011). Indeterminacy of factor score estimates in slightly misspecified confirmatory factor models. Journal of Modern Applied Statistical Methods, 10(2), 583–598. https://doi.org/10.22237/jmasm/1320120900

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238–246. https://doi.org/10.1037/0033-2909.107.2.238

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88(3), 588–606. https://doi.org/10.1037/0033-2909.88.3.588

Bentler, P. M., & Satorra, A. (2010). Testing model nesting and equivalence. Psychological Methods, 15(2), 111–123. https://doi.org/10.1037/a0019625

Boateng, G. O., Neilands, T. B., Frongillo, E. A., Melgar-Quiñonez, H. R., & Young, S. L. (2018). Best practices for developing and validating scales for health, social, and behavioral research: A primer. Frontiers in Public Health, 6, 149. https://doi.org/10.3389/fpubh.2018.00149

Braeken, J., & van Assen, M. A. L. M. (2017). An empirical Kaiser criterion. Psychological Methods, 22(3), 450–466. https://doi.org/10.1037/met0000074

Bonifay, W., Lane, S. P., & Reise, S. P. (2017). Three concerns with applying a bifactor model as a structure of psychopathology. Clinical Psychological Science, 5(1), 184–186. https://doi.org/10.1177/2167702616657069

Browne, M. W. (2001). An overview of analytic rotation in exploratory factor analysis. Multivariate Behavioral Research, 36(1), 111–150. https://doi.org/10.1207/S15327906MBR3601_05

Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods & Research, 21(2), 230–258. https://doi.org/10.1177/0049124192021002005

Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference: Understanding AIC and BIC in model selection. Sociological Methods & Research, 33(2), 261–304. https://doi.org/10.1177/0049124104268644

Byrne, B. M., Shavelson, R. J., & Muthén, B. (1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105(3), 456–466. https://doi.org/10.1037/0033-2909.105.3.456

Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56(2), 81–105. https://doi.org/10.1037/h0046016

Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1(2), 245–276. https://doi.org/10.1207/s15327906mbr0102_10

Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling, 14(3), 464–504. https://doi.org/10.1080/10705510701301834

Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9(2), 233–255. https://doi.org/10.1207/S15328007SEM0902_5

Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. https://doi.org/10.1037/1040-3590.7.3.309

Clark, L. A., & Watson, D. (2019). Constructing validity: New developments in creating objective measuring instruments. Psychological Assessment, 31(12), 1412–1427. https://doi.org/10.1037/pas0000626

Conway, J. M., & Huffcutt, A. I. (2003). A review and evaluation of exploratory factor analysis practices in organizational research. Organizational Research Methods, 6(2), 147–168. https://doi.org/10.1177/1094428103251541

Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research, and Evaluation, 10, Article 7. https://doi.org/10.7275/jyj1-4868

Crawford, A. V., Green, S. B., Levy, R., Lo, W.-J., Scott, L., Svetina, D., & Thompson, M. S. (2010). Evaluation of parallel analysis methods for determining the number of factors. Educational and Psychological Measurement, 70(6), 885–901. https://doi.org/10.1177/0013164410379332

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297–334. https://doi.org/10.1007/BF02310555

Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281–302. https://doi.org/10.1037/h0040957

Curran, P. G. (2016). Methods for the detection of carelessly invalid responses in survey data. Journal of Experimental Social Psychology, 66, 4–19. https://doi.org/10.1016/j.jesp.2015.07.006

Dunn, T. J., Baguley, T., & Brunsden, V. (2014). From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British Journal of Psychology, 105(3), 399–412. https://doi.org/10.1111/bjop.12046

Enders, C. K., & Bandalos, D. L. (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling, 8(3), 430–457. https://doi.org/10.1207/S15328007SEM0803_5

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. https://doi.org/10.1037/1082-989X.4.3.272

Flake, J. K., & Fried, E. I. (2020). Measurement schmeasurement: Questionable measurement practices and how to avoid them. Advances in Methods and Practices in Psychological Science, 3(4), 456–465. https://doi.org/10.1177/2515245920952393

Flake, J. K., Pek, J., & Hehman, E. (2017). Construct validation in social and personality research: Current practice and recommendations. Social Psychological and Personality Science, 8(4), 370–378. https://doi.org/10.1177/1948550617693063

Flora, D. B. (2020). Your coefficient alpha is probably wrong, but which coefficient omega is right? A tutorial on using R to obtain better reliability estimates. Advances in Methods and Practices in Psychological Science, 3(4), 484–501. https://doi.org/10.1177/2515245920951747

Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9(4), 466–491. https://doi.org/10.1037/1082-989X.9.4.466

Fokkema, M., & Greiff, S. (2017). How performing PCA and CFA on the same data equals trouble. European Journal of Psychological Assessment, 33(6), 399–402. https://doi.org/10.1027/1015-5759/a000460

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.2307/3151312

Green, S. B., & Yang, Y. (2009). Reliability of summed item scores using structural equation modeling: An alternative to coefficient alpha. Psychometrika, 74(1), 155–167. https://doi.org/10.1007/s11336-008-9099-3

Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Lawrence Erlbaum.

Grice, J. W. (2001). Computing and evaluating factor scores. Psychological Methods, 6(4), 430–450. https://doi.org/10.1037/1082-989X.6.4.430

Guttman, L. (1954). Some necessary conditions for common-factor analysis. Psychometrika, 19(2), 149–161. https://doi.org/10.1007/BF02289162

Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8

Hinkin, T. R. (1998). A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods, 1(1), 104–121. https://doi.org/10.1177/109442819800100106

Hancock, G. R., & Mueller, R. O. (2001). Rethinking construct reliability within latent variable systems. In R. Cudeck, S. du Toit, & D. Sörbom (Eds.), Structural equation modeling: Present and future (pp. 195–216). Scientific Software International.

Holzinger, K. J., & Swineford, F. (1937). The bi-factor method. Psychometrika, 2(1), 41–54. https://doi.org/10.1007/BF02287965

Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis. Psychometrika, 30(2), 179–185. https://doi.org/10.1007/BF02289447

Hu, L., & 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. https://doi.org/10.1080/10705519909540118

Huang, J. L., Curran, P. G., Keeney, J., Poposki, E. M., & DeShon, R. P. (2012). Detecting and deterring insufficient effort responding to surveys. Journal of Business and Psychology, 27(1), 99–114. https://doi.org/10.1007/s10869-011-9231-8

Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34(2), 183–202. https://doi.org/10.1007/BF02289343

Jöreskog, K. G. (1971). Simultaneous factor analysis in several populations. Psychometrika, 36(4), 409–426. https://doi.org/10.1007/BF02291366

Kaiser, H. F. (1958). The varimax criterion for analytic rotation in factor analysis. Psychometrika, 23(3), 187–200. https://doi.org/10.1007/BF02289233

Kaiser, H. F. (1960). The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20(1), 141–151. https://doi.org/10.1177/001316446002000116

Kaiser, H. F. (1970). A second generation little jiffy. Psychometrika, 35(4), 401–415. https://doi.org/10.1007/BF02291817

Kaiser, H. F. (1974). An index of factorial simplicity. Psychometrika, 39(1), 31–36. https://doi.org/10.1007/BF02291575

Kelley, K., & Pornprasertmanit, S. (2016). Confidence intervals for population reliability coefficients: Evaluation of methods, recommendations, and software for composite measures. Psychological Methods, 21(1), 69–92. https://doi.org/10.1037/a0040086

Kolenikov, S., & Bollen, K. A. (2012). Testing negative error variances: Is a Heywood case a symptom of misspecification? Sociological Methods & Research, 41(1), 124–167. https://doi.org/10.1177/0049124112442138

Kuhn, M., & Johnson, K. (2013). Applied predictive modeling. Springer. https://doi.org/10.1007/978-1-4614-6849-3

Lakens, D., Scheel, A. M., & Isager, P. M. (2018). Equivalence testing for psychological research: A tutorial. Advances in Methods and Practices in Psychological Science, 1(2), 259–269. https://doi.org/10.1177/2515245918770963

Lorenzo-Seva, U., Timmerman, M. E., & Kiers, H. A. L. (2011). The Hull method for selecting the number of common factors. Multivariate Behavioral Research, 46(2), 340–364. https://doi.org/10.1080/00273171.2011.564527

MacCallum, R. C., Roznowski, M., & Necowitz, L. B. (1992). Model modifications in covariance structure analysis: The problem of capitalization on chance. Psychological Bulletin, 111(3), 490–504. https://doi.org/10.1037/0033-2909.111.3.490

Marjanovic, Z., Holden, R., Struthers, W., Cribbie, R., & Greenglass, E. (2015). The inter-item standard deviation (ISD): An index that discriminates between conscientious and random responders. Personality and Individual Differences, 84, 79–83. https://doi.org/10.1016/j.paid.2014.08.021

Marsh, H. W., Hau, K.-T., & Wen, Z. (2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler’s (1999) findings. Structural Equation Modeling, 11(3), 320–341. https://doi.org/10.1207/s15328007sem1103_2

McNeish, D. (2018). Thanks coefficient alpha, we’ll take it from here. Psychological Methods, 23(3), 412–433. https://doi.org/10.1037/met0000144

McNeish, D., & Wolf, M. G. (2020). Thinking twice about sum scores. Behavior Research Methods, 52(6), 2287–2305. https://doi.org/10.3758/s13428-020-01398-0

McNeish, D., & Wolf, M. G. (2023). Dynamic fit index cutoffs for confirmatory factor analysis models. Psychological Methods, 28(1), 61–88. https://doi.org/10.1037/met0000425

Meade, A. W., & Craig, S. B. (2012). Identifying careless responses in survey data. Psychological Methods, 17(3), 437–455. https://doi.org/10.1037/a0028085

Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58(4), 525–543. https://doi.org/10.1007/BF02294825

Messick, S. (1995). Validity of psychological assessment: Validation of inferences from persons’ responses and performances as scientific inquiry into score meaning. American Psychologist, 50(9), 741–749. https://doi.org/10.1037/0003-066X.50.9.741

Nosek, B. A., Ebersole, C. R., DeHaven, A. C., & Mellor, D. T. (2018). The preregistration revolution. Proceedings of the National Academy of Sciences, 115(11), 2600–2606. https://doi.org/10.1073/pnas.1708274114

Murray, A. L., & Johnson, W. (2013). The limitations of model fit in comparing the bi-factor versus higher-order models of human cognitive ability structure. Intelligence, 41(5), 407–422. https://doi.org/10.1016/j.intell.2013.06.004

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.

Putnick, D. L., & Bornstein, M. H. (2016). Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Developmental Review, 41, 71–90. https://doi.org/10.1016/j.dr.2016.06.004

Raftery, A. E. (1995). Bayesian model selection in social research. Sociological Methodology, 25, 111–163. https://doi.org/10.2307/271063

Reise, S. P. (2012). The rediscovery of bifactor measurement models. Multivariate Behavioral Research, 47(5), 667–696. https://doi.org/10.1080/00273171.2012.715555

Reise, S. P., Bonifay, W. E., & Haviland, M. G. (2013). Scoring and modeling psychological measures in the presence of multidimensionality. Journal of Personality Assessment, 95(2), 129–140. https://doi.org/10.1080/00223891.2012.725437

Rhemtulla, M., Brosseau-Liard, P. É., & Savalei, V. (2012). When can categorical variables be treated as continuous? A comparison of robust continuous and categorical SEM estimation methods under suboptimal conditions. Psychological Methods, 17(3), 354–373. https://doi.org/10.1037/a0029315

Rodriguez, A., Reise, S. P., & Haviland, M. G. (2016). Evaluating bifactor models: Calculating and interpreting statistical indices. Psychological Methods, 21(2), 137–150. https://doi.org/10.1037/met0000045

Roemer, E., Schuberth, F., & Henseler, J. (2021). HTMT2—An improved criterion for assessing discriminant validity in structural equation modeling. Industrial Management & Data Systems, 121(12), 2637–2650. https://doi.org/10.1108/IMDS-02-2021-0082

Rönkkö, M., & Cho, E. (2022). An updated guideline for assessing discriminant validity. Organizational Research Methods, 25(1). https://doi.org/10.1177/1094428120968614

Rosseel, Y. (2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

Rosseel, Y., & Loh, W. W. (2024). A structural after measurement approach to structural equation modeling. Psychological Methods, 29(3), 561–588. https://doi.org/10.1037/met0000503

Ruscio, J., & Roche, B. (2012). Determining the number of factors to retain in an exploratory factor analysis using comparison data of known factorial structure. Psychological Assessment, 24(2), 282–292. https://doi.org/10.1037/a0025697

Satorra, A. (2000). Scaled and adjusted restricted tests in multi-sample analysis of moment structures. In Innovations in multivariate statistical analysis (pp. 233–247). Springer. https://doi.org/10.1007/978-1-4615-4603-0_17

Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66(4), 507–514. https://doi.org/10.1007/BF02296192

Satorra, A., & Bentler, P. M. (2010). Ensuring positiveness of the scaled difference chi-square test statistic. Psychometrika, 75(2), 243–248. https://doi.org/10.1007/s11336-009-9135-y

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. https://doi.org/10.1037/1082-989X.7.2.147

Schmid, J., & Leiman, J. M. (1957). The development of hierarchical factor solutions. Psychometrika, 22(1), 53–61. https://doi.org/10.1007/BF02289209

Schuirmann, D. J. (1987). A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. Journal of Pharmacokinetics and Biopharmaceutics, 15(6), 657–680. https://doi.org/10.1007/BF01068419

Schwarz, G. (1978). Estimating the dimension of a model. The Annals of Statistics, 6(2), 461–464. https://doi.org/10.1214/aos/1176344136

Sijtsma, K. (2009). On the use, the misuse, and the very limited usefulness of Cronbach’s alpha. Psychometrika, 74(1), 107–120. https://doi.org/10.1007/s11336-008-9101-0

Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology: Undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychological Science, 22(11), 1359–1366. https://doi.org/10.1177/0956797611417632

Skrondal, A., & Laake, P. (2001). Regression among factor scores. Psychometrika, 66(4), 563–575. https://doi.org/10.1007/BF02296196

Svetina, D., Rutkowski, L., & Rutkowski, D. (2020). Multiple-group invariance with categorical outcomes using updated guidelines: An illustration using Mplus and the lavaan/semTools packages. Structural Equation Modeling, 27(1), 111–130. https://doi.org/10.1080/10705511.2019.1602776

Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3(1), 4–70. https://doi.org/10.1177/109442810031002

Velicer, W. F. (1976). Determining the number of components from the matrix of partial correlations. Psychometrika, 41(3), 321–327. https://doi.org/10.1007/BF02293557

Velicer, W. F., Eaton, C. A., & Fava, J. L. (2000). Construct explication through factor or component analysis: A review and evaluation of alternative procedures for determining the number of factors or components. In R. D. Goffin & E. Helmes (Eds.), Problems and solutions in human assessment (pp. 41–71). Springer. https://doi.org/10.1007/978-1-4615-4397-8_3

Voorhees, C. M., Brady, M. K., Calantone, R., & Ramirez, E. (2016). Discriminant validity testing in marketing: An analysis, causes for concern, and proposed remedies. Journal of the Academy of Marketing Science, 44(1), 119–134. https://doi.org/10.1007/s11747-015-0455-4

Watkins, M. W. (2018). Exploratory factor analysis: A guide to best practice. Journal of Black Psychology, 44(3), 219–246. https://doi.org/10.1177/0095798418771807

Wicherts, J. M., Veldkamp, C. L. S., Augusteijn, H. E. M., Bakker, M., van Aert, R. C. M., & van Assen, M. A. L. M. (2016). Degrees of freedom in planning, running, analyzing, and reporting psychological studies: A checklist to avoid p-hacking. Frontiers in Psychology, 7, 1832. https://doi.org/10.3389/fpsyg.2016.01832

Wu, H., & Estabrook, R. (2016). Identification of confirmatory factor analysis models of different levels of invariance for ordered categorical outcomes. Psychometrika, 81(4), 1014–1045. https://doi.org/10.1007/s11336-016-9506-0

Yung, Y.-F., Thissen, D., & McLeod, L. D. (1999). On the relationship between the higher-order factor model and the hierarchical factor model. Psychometrika, 64(2), 113–128. https://doi.org/10.1007/BF02294531