Finds (A, B) of theta_T = A theta_F + B minimizing the summed
squared differences between item characteristic curves of the
common items (plink Eq. 14), with transformed from-scale
parameters a* = a_F/A, b* = A b_F + B, c* = c_F. The symmetric
variant adds the reverse-direction term (a# = A a_T,
b# = (b_T - B)/A).
Usage
morie_linkhae(items_from, items_to, symmetric = FALSE, theta_points = NULL)
Arguments
- items_from, items_to
Matrices (or lists) of common-item 3PL
parameters (a, b, c) on each scale; use c = 0 for 2PL.
- symmetric
Minimize Q1 + Q2 instead of Q1 only.
- theta_points
Optional evaluation grid (default 41 points on
[-4, 4]).
Value
A list with elements A, B, criterion,
symmetric, n_common, method.
References
Haebara, T. (1980). Equating logistic ability scales
by a weighted least squares method. Japanese Psychological
Research, 22, 144-149. Weeks, J. P. (2010). JSS, 35(12).
Examples
M <- matrix(c(1, 2, 3, 4, 5, 6), nrow = 2)
morie_linkhae(M, M)
#> $A
#> [1] 1
#>
#> $B
#> [1] 0
#>
#> $criterion
#> [1] 0
#>
#> $symmetric
#> [1] FALSE
#>
#> $n_common
#> [1] 2
#>
#> $method
#> [1] "Haebara characteristic-curve linking (plink Eq. 14)"
#>
