Transform Scores to Percent of Maximum Possible (POMPS)

pomps() transforms scores to a 0-100 scale using the Percent of Maximum Possible Scores method. This makes different scales directly comparable regardless of their original range.

This is the R equivalent of the SPSS formula: COMPUTE v81p = ((v81 - 1) / (7 - 1)) * 100.

A score of 0 means the minimum possible score, 100 means the maximum. The transformation preserves all correlations between variables.

Usage

pomps(x, scale_min = NULL, scale_max = NULL)

Arguments

x

A numeric vector to transform (e.g., a column from your data).

scale_min

The theoretical minimum of the scale. If NULL (default), the observed minimum of x is used.

scale_max

The theoretical maximum of the scale. If NULL (default), the observed maximum of x is used.

Value

A numeric vector of the same length as x, with values rescaled to the 0-100 range. The variable label of x is kept; missing values (including the SPSS missing types of imported data) become plain NA, as in an SPSS COMPUTE.

Details

The Formula

POMPS = ((score - scale_min) / (scale_max - scale_min)) * 100

Why Specify scale_min and scale_max?

By default, pomps() uses the observed minimum and maximum of your data. However, for Likert scales you should specify the theoretical range:

• A 1-5 Likert scale: scale_min = 1, scale_max = 5

• A 1-7 Likert scale: scale_min = 1, scale_max = 7

• A 0-10 scale: scale_min = 0, scale_max = 10

Using theoretical values ensures that the transformation is consistent across samples and time points. Values outside the range (e.g. an unrecoded "don't know" = 9 on a 1-5 scale) give scores below 0 or above 100; pomps() warns about them, so recode missing codes first.

When to Use This

• Comparing variables measured on different scales

• Creating profile plots across scales with different ranges

• Reporting scale scores in an intuitive 0-100 format

See also

row_means for creating mean indices across items.

Other scale: efa(), reliability(), row_count(), row_means(), row_sums()

Examples

library(dplyr)
data(survey_data)

# Transform a 1-5 Likert scale to POMPS
survey_data <- survey_data %>%
  mutate(trust_gov_pomps = pomps(trust_government, scale_min = 1, scale_max = 5))

# Transform multiple variables with the same scale
survey_data <- survey_data %>%
  mutate(across(
    c(trust_government, trust_media, trust_science),
    ~ pomps(.x, scale_min = 1, scale_max = 5),
    .names = "{.col}_pomps"
  ))

# Auto-detect range (uses observed min/max)
survey_data <- survey_data %>%
  mutate(age_pomps = pomps(age))