nps_moe_test: Calculate whether NPS score has significantly changed between...

In MarkMc1089/FIC: Shiny Dashboard For FIC

Calculate whether NPS score has significantly changed between 2 samples.

Description

A margin of error is calculated for each sample, from the number of promoters, neutrals (i.e. passives) and detractors. The standard error of their difference is estimated using the Pythagorean formula, and the absolute difference of the two samples is compared to this multiplied by the critical value (aka z*-value).

The return value is in (-1, 0, +1), according to whether a significant decrease is found, no significant change, or a significant increase, respectively. If the total for a sample is 0, then 0 is returned.

Formula is based on the one found in this [blog post] (https://www.genroe.com/blog/how-to-calculate-margin-of-error-and-other-stats- for-nps/5994).

Usage

nps_moe_test(p_0, n_0, d_0, p_1, n_1, d_1, z_val = 1.96)

Arguments

p_0	Number of Promoters in latest sample
n_0	Number of Neutrals in latest sample
d_0	Number of Detractors in latest sample
p_1	Number of Promoters in oldest sample
n_1	Number of Neutrals in oldest sample
d_1	Number of Detractors in oldest sample
z_val	Critical value multiplier; 1.96 by default for a 95 interval. See [this table] (http://www.ltcconline.net/greenl/courses/201/estimation/smallConfLevelTable.htm) for further values of z_val for common confidence intervals.

Value

A value in (-1, 0, +1); see notes above.

Examples

# Test with a 99% confidence interval
## Not run: 
nps_moe_test(123, 456, 789, 321, 654, 987, z_val = 2.58)

## End(Not run)

MarkMc1089/FIC documentation built on April 3, 2022, 6:05 p.m.