Learn R Programming

psychmeta (version 2.7.0)

correct_r_coarseness: Correct correlations for scale coarseness

Description

Corrects correlations for scale coarseness.

Usage

correct_r_coarseness(
  r,
  kx = NULL,
  ky = NULL,
  n = NULL,
  dist_x = "norm",
  dist_y = "norm",
  bin_value_x = c("median", "mean", "index"),
  bin_value_y = c("median", "mean", "index"),
  width_x = 3,
  width_y = 3,
  lbound_x = NULL,
  ubound_x = NULL,
  lbound_y = NULL,
  ubound_y = NULL,
  index_values_x = NULL,
  index_values_y = NULL
)

Value

Vector of correlations corrected for scale coarseness (if n is supplied, corrected error variance and adjusted sample size is also reported).

Arguments

r

Observed correlation.

kx, ky

Number of scale points used to measure the x and y variables. Set to NULL to treat as continuously measured.

n

Optional sample size.

dist_x, dist_y

Assumed latent distribution of the x and y variables.

bin_value_x, bin_value_y

Are the scale points used to measure the of the x and y variables assumed to represent bin medians, means, or index values?

width_x, width_y

For symmetrically distributed variables, how many standard deviations above/below the latent mean should be be used for the latent variable range to make the correction? (Note: Setting width > 3 produces erratic results.) The latent variable range can alternatively be set using lbound and ubound.

lbound_x, lbound_y

What lower bound of the range for the latent x and y variables should be used to make the correction? (Note: For normally distributed variables, setting lbound < -3 produces erratic results.)

ubound_x, ubound_y

What upper bound of the range for the latent x and y variables should be used to make the correction? (Note: For normally distributed variables, setting ubound > 3 produces erratic results.)

index_values_x, index_values_y

Optional. If bin_value = "index", the bin index values. If unspecified, values 1:k are used.

References

Aguinis, H., Pierce, C. A., & Culpepper, S. A. (2009). Scale coarseness as a methodological artifact: Correcting correlation coefficients attenuated from using coarse scales. Organizational Research Methods, 12(4), 623–652. tools:::Rd_expr_doi("10.1177/1094428108318065")

Schmidt, F. L., & Hunter, J. E. (2015). Methods of meta-analysis: Correcting error and bias in research findings (3rd ed.). Sage. tools:::Rd_expr_doi("10.4135/9781483398105"). pp. 287-288.

Peters, C. C., & Van Voorhis, W. R. (1940). Statistical procedures and their mathematical bases. New York, NY: Mcgraw-Hill. tools:::Rd_expr_doi("10.1037/13596-000"). pp. 393–399.

Examples

Run this code
correct_r_coarseness(r = .35, kx = 5, ky = 4, n = 100)
correct_r_coarseness(r = .35, kx = 5, n = 100)
correct_r_coarseness(r = .35, kx = 5, ky = 4, n = 100, dist_x="unif", dist_y="norm")

Run the code above in your browser using DataLab