Automated Psychometric Workflows and Reporting Tools

Automates common psychometric workflows for applied researchers, including item descriptives, inter-item correlations, exploratory and confirmatory factor analysis, reliability, multi-group measurement invariance, and alignment optimization. Decision heuristics are informed by procedures such as parallel analysis (Horn, 1965, ), multivariate normality diagnostics (Mardia, 1970, ), measurement-invariance fit-change rules (Chen, 2007, ), and alignment optimization (Asparouhov and Muthen, 2014, ), among others. Results can be returned as structured R objects and exported as bilingual reports for transparent analytical documentation.

PsychoMatic

Automated psychometric analysis with minimal syntax

PsychoMatic is an R package for automated psychometric workflows. Its main goal is to let applied researchers run common psychometric analyses with the least possible syntax, while still returning transparent results, structured tables, bilingual reports, and literature-informed decision support.

The package is designed for workflows where the researcher wants to move from item screening to factor analysis, reliability, scoring, measurement invariance, IRT/DIF screening with interpretation, and reporting without writing long procedural code for each step.

Design Philosophy

PsychoMatic prioritizes:

• Minimal syntax for common psychometric analyses.
• Conservative automated defaults based on psychometric practice.
• English-facing API and documentation for CRAN-readiness.
• Bilingual output through language = "esp" or language = "eng".
• Structured return objects that can be inspected, tested, summarized, and exported.
• Optional advanced arguments for researchers who need more control.

Installation

install.packages("remotes")
remotes::install_github("gmoncayoj/PsychoMatic")

Minimal Workflow

library(PsychoMatic)

data(psychomatic_ordinal)
data(psychomatic_continuous)

# Screen item quality
screen_items(psychomatic_ordinal)

# Compute a total score with minimal syntax
scores <- score_scale(psychomatic_ordinal)

# Run automated EFA
efa_result <- efa_auto(psychomatic_ordinal, verbose = FALSE, language = "eng")
summary(efa_result)

# Compare CFA models
models <- list(
  one_factor = "general =~ item1 + item2 + item3 + item4 + item5 + item6",
  two_factor = "
    factor1 =~ item1 + item2 + item3
    factor2 =~ item4 + item5 + item6
  "
)
compare_models_auto(psychomatic_continuous, models = models)

Core Functions

Example Datasets

PsychoMatic includes small simulated datasets for reproducible examples:

Descriptives And Item Screening

data(psychomatic_ordinal)

desc_auto(psychomatic_ordinal, language = "eng")
screen_items(psychomatic_ordinal)

Scoring

keys <- list(
  total = names(psychomatic_ordinal)
)

score_scale(
  psychomatic_ordinal,
  keys = keys,
  method = "mean",
  min_valid = 0.80
)

Exploratory Factor Analysis

efa_result <- efa_auto(
  psychomatic_ordinal,
  rotation = "oblique",
  verbose = FALSE,
  language = "eng"
)

summary(efa_result)

Confirmatory Factor Analysis

data(psychomatic_continuous)

model <- "
factor1 =~ item1 + item2 + item3
factor2 =~ item4 + item5 + item6
"

cfa_result <- cfa_auto(
  psychomatic_continuous,
  model = model,
  estimator = "ML",
  language = "eng"
)

summary(cfa_result)

Measurement Invariance

data(psychomatic_multigroup)

model <- "
factor1 =~ mg1 + mg2 + mg3
factor2 =~ mg4 + mg5 + mg6
"

invariance <- factorial_invariance_auto(
  psychomatic_multigroup,
  group = "group",
  model = model,
  estimator = "ML",
  language = "eng",
  report = FALSE
)

summary(invariance)

IRT And DIF Screening

data(psychomatic_ordinal)
data(psychomatic_multigroup)

irt_result <- irt_auto(psychomatic_ordinal[, 1:4], model = "grm", language = "eng")
irt_result$model_summary
irt_result$interpretation
irt_result$references

dif_result <- dif_auto(
  psychomatic_multigroup[paste0("mg", 1:6)],
  group = psychomatic_multigroup$group,
  language = "eng"
)
dif_result$summary
dif_result$interpretation
dif_result$references

Reporting

Reports are explicit actions. In examples and tests, use temporary files so the package does not write uncontrolled files to the working directory.

report_path <- tempfile("psychomatic-report-", fileext = ".xlsx")

psychomatic_report(
  screening = screen_items(psychomatic_ordinal),
  scores = score_scale(psychomatic_ordinal),
  format = "excel",
  file_name = report_path,
  language = "eng"
)

CRAN-Readiness

Version 0.3.0 focuses on CRAN-readiness:

• English documentation and manual pages.
• Smaller, focused R files for EFA/CFA helpers, methods, and exports.
• Optional dependencies loaded only when needed.
• Executable tests for success and error paths.
• Minimal vignettes using bundled datasets.
• Clean R CMD check --as-cran on the release candidate tarball.

Development Status

PsychoMatic is in active development. The package aims to support applied psychometric research while keeping the interface simple enough for teaching, graduate research, and reproducible analytical reports.

Author

Jose Gamarra-Moncayo
Psychology professor, Faculty of Medicine, Universidad Catolica Santo Toribio de Mogrovejo
Email: [email protected]

Suggestions and bug reports are welcome through the GitHub issue tracker.

License

MIT