response_time_review: Review response-time patterns outside the MFRM likelihood

In mfrmr: Estimation and Diagnostics for Many-Facet Measurement Models

Review response-time patterns outside the MFRM likelihood

Description

Build a descriptive response-time review table from the same long-format rating-event data used by fit_mfrm(). This helper does not fit a joint response-time model and does not change MFRM estimates. It summarizes response-time distributions, distributional rapid/slow flags, and person / facet / score-level response-time patterns for screening and reporting context.

Usage

response_time_review(
  data,
  person,
  facets = NULL,
  time,
  score = NULL,
  time_unit = "seconds",
  min_time = 0,
  rapid_threshold = NULL,
  slow_threshold = NULL,
  rapid_quantile = 0.05,
  slow_quantile = 0.95,
  rapid_rate_warn = 0.25,
  slow_rate_warn = 0.25,
  min_n_flag = 3L
)

Arguments

data	A data.frame in long format with one row per observed rating event.
person	Column name for the person identifier.
facets	Optional character vector of facet columns to summarize.
time	Column name containing positive response times.
score	Optional ordered-score column. When supplied, score-level response-time summaries are returned.
time_unit	Label for the response-time unit, such as "seconds".
min_time	Minimum valid response time. Values must be strictly greater than this threshold; default 0.
rapid_threshold	Optional numeric response-time cutoff for rapid responses. When NULL, it is estimated from rapid_quantile.
slow_threshold	Optional numeric response-time cutoff for slow responses. When NULL, it is estimated from slow_quantile.
rapid_quantile	Quantile used when rapid_threshold = NULL.
slow_quantile	Quantile used when slow_threshold = NULL.
rapid_rate_warn	Group-level rapid-response rate that creates a descriptive flag; default 0.25.
slow_rate_warn	Group-level slow-response rate that creates a descriptive flag; default 0.25.
min_n_flag	Minimum group size before rapid/slow rates are flagged; default 3.

Value

An object of class mfrm_response_time_review, a list with overview, thresholds, observations, person_summary, facet_summary, score_summary, flags, and notes.

See Also

plot_response_time_review(), fit_mfrm(), mfrmr_visual_diagnostics, mfrmr_output_guide()

Examples

toy <- load_mfrmr_data("example_core")
toy$ResponseTime <- 12 + as.numeric(factor(toy$Person)) * 0.4 +
  as.numeric(toy$Score)
rt <- response_time_review(
  toy,
  person = "Person",
  facets = c("Rater", "Criterion"),
  score = "Score",
  time = "ResponseTime"
)
summary(rt)
plot_response_time_review(rt, draw = FALSE)

mfrmr documentation built on June 13, 2026, 1:07 a.m.