Count binary responses by group

Aggregates trial-level binary responses into the number of signal responses (y) and the number of trials (n) for every combination of the grouping columns. A binomial model fitted to these counts has the same likelihood, and therefore the same estimates and posterior, as a Bernoulli model fitted to the trials, and it runs much faster. Use it to prepare custom models fitted with brms or lme4 (see usdt_hypotheses).

Usage

usdt_aggregate(data, response, by)

Arguments

data

A data frame with one row per trial.

response

Name of the binary response column, coded 0/1 or FALSE/TRUE.

by

Names of the grouping columns. Every predictor of the model, the subject included, must be among them, so that all the trials of a row share their predictors.

Value

A data frame with one row per combination of the by columns that occurs in data, those columns, and the counts y and n.

See also

usdt_hypotheses

Examples

# Awareness task of Vadillo et al. (2025): "old" judgements by subject,
# condition and set size
counts <- usdt_aggregate(vadillo_awareness, response = "judged.old",
                         by = c("subj", "condition", "set.size"))
head(counts)
#>   subj condition    set.size  y  n
#> 1 2001       new set size 16  9 16
#> 2 2002       new set size 16 14 16
#> 3 2003       new set size 16 14 16
#> 4 2004       new set size 16  8 16
#> 5 2005       new set size 16 11 16
#> 6 2006       new set size 16 11 16

# The counts keep every trial
sum(counts$n) == nrow(vadillo_awareness)
#> [1] TRUE