Compute per-group means and 95% CI bounds

For each group, computes the mean x/y position and a rectangular 95% confidence interval using t(df = n-1, 0.975) * SE on each axis independently. This matches the CI convention used by rENA / Web ENA.

Usage

group_summary(pts, x_col, y_col, group_col, conf_level = 0.95, outlier = FALSE)

Arguments

pts

A data.frame with at least the columns named by x_col, y_col, and group_col.

x_col

Name of the x-coordinate column.

y_col

Name of the y-coordinate column.

group_col

Name of the grouping column.

conf_level

Confidence level for the t interval. Default 0.95.

outlier

Also add <dim>.olow / <dim>.ohigh outlier bounds (Q1 − 1.5·IQR, Q3 + 1.5·IQR). Default FALSE.

Value

A data.frame with the group column, x_col, y_col, and <x_col>.low, <x_col>.high, <y_col>.low, <y_col>.high (the bound names qeviz looks up from the model's dimension names).