Difference score statistics
Usage
difference_score(
x,
y,
r_xx = 0.9,
r_yy = 0.9,
r_xy = 0,
mu = 100,
sigma = 15,
ci = 0.95,
mu_x = mu,
mu_y = mu,
sigma_x = sigma,
sigma_y = sigma,
tails = 2
)
Arguments
- x
first score
- y
second score
- r_xx
reliability of x
- r_yy
reliability of y
- r_xy
correlation between x and y
- mu
population mean of both x and y
- sigma
population standard deviation of both x and y
- ci
confidence interval of difference score
- mu_x
population mean of x (defaults to mu)
- mu_y
population mean of y (defaults to mu)
- sigma_x
population standard deviation of x (defaults to sigma)
- sigma_y
population standard deviation of y (defaults to sigma)
- tails
for significance and prevalence of difference scores
Examples
difference_score(
x = 120,
y = 110,
r_xx = .95,
r_yy = .92,
r_xy = .65,
mu = 100,
sigma = 15)
#> $d
#> [1] 10
#>
#> $d_ci_lb
#> [1] -1.422461
#>
#> $d_ci_ub
#> [1] 17.70818
#>
#> $d_sig
#> [1] 0.06445772
#>
#> $d_prevalence
#> [1] 0.4255561
#>
#> $sd_d
#> [1] 12.5499
#>
#> $r_dd
#> [1] 0.8142857
#>
#> $x
#> [1] 120
#>
#> $y
#> [1] 110
#>
#> $r_xx
#> [1] 0.95
#>
#> $r_yy
#> [1] 0.92
#>
#> $r_xy
#> [1] 0.65
#>
#> $mu_x
#> [1] 100
#>
#> $mu_y
#> [1] 100
#>
#> $sigma_x
#> [1] 15
#>
#> $sigma_y
#> [1] 15
#>
#> $tails
#> [1] 2
#>
#> $ci
#> [1] 0.95
#>