This function tests whether a given mean (with a specific precision) can
result from a sample of a given size based on integer responses to one or more
items. The test is based on Brown & Heathers (2017).
If return_values = TRUE and if there is more than one precise mean compatible
with the given parameters, all possible means are returned. In that case, if the
given mean is not consistent, the closest consistent mean is returned with a
warning.
GRIM_test(mean, n_obs, m_prec = NULL, n_items = 1, return_values = FALSE)The mean of the distribution
The number of observations (sample size)
The precision of the mean, as number of digits after the decimal point.
If not provided, taken based on the significant digits of mean - so only needed if reported mean ends in 0
Number of items in scale, if distribution represents scale averages. Defaults to 1, which represents any single-item measure.
Should all means consistent with the given parameters be returned?
Either TRUE/FALSE, or all possible means (if test passes)/closest consistent mean (if test fails)
Brown NJ, Heathers JA (2017). “The GRIM test: A simple technique detects numerous anomalies in the reporting of results in psychology.” Social Psychological and Personality Science, 8(4), 363--369.
# A sample of 28 integers cannot result in a mean of 5.19. This is shown by
GRIM_test(5.19, 28)
#> [1] FALSE
# To find the closest possible mean, set return_values to TRUE
GRIM_test(5.19, 28, return_values = TRUE)
#> Warning: Mean 5.19 fails GRIM test - closest consistent value: 5.18
#> [1] 5.18