lfast {LikertMakeR}

Synthesise rating-scale data with predefined mean and standard deviation

Description

lfast() applies a simple Evolutionary Algorithm to find a vector that best fits the desired moments.

lfast() generates random discrete values from a scaled Beta distribution so the data replicate a rating scale - for example, a 1-5 Likert scale made from 5 items (questions) or 0-10 likelihood-of-purchase scale.

Usage

lfast(n, mean, sd, lowerbound, upperbound, items = 1, precision = 0)

Arguments

n

(positive, int) number of observations to generate

mean

(real) target mean, between upper and lower bounds

sd

(positive, real) target standard deviation

lowerbound

(positive, int) lower bound (e.g. '1' for a 1-5 rating scale)

upperbound

(positive, int) upper bound (e.g. '5' for a 1-5 rating scale)

items

(positive, int) number of items in the rating scale. Default = 1

precision

(positive, real) can relax the level of accuracy required. (e.g. '1' generally generates a vector with moments correct within '0.025', '2' generally within '0.05') Default = 0

Value

a vector approximating user-specified conditions.

Examples

## six-item 1-7 rating scale
x <- lfast(
  n = 256,
  mean = 4.0,
  sd = 1.25,
  lowerbound = 1,
  upperbound = 7,
  items = 6
)

## four-item 1-5 rating scale with medium variation
x <- lfast(
  n = 128,
  mean = 3.0,
  sd = 1.00,
  lowerbound = 1,
  upperbound = 5,
  items = 4,
  precision = 5
)

## eleven-point 'likelihood of purchase' scale
x <- lfast(256, 3, 3.0, 0, 10)