The probability of each answer category, and its first derivative, at an ability
value, under the rating scale model of Andrich (1978):
$$P(X = k \mid \theta) = \frac{\exp \sum_{t=1}^{k} D(\theta - (\lambda_j + \delta_t))}
{\sum_{h=0}^{m} \exp \sum_{t=1}^{h} D(\theta - (\lambda_j + \delta_t))}$$
with the empty sum of category 0 equal to zero. It returns the same result as
catR::Pi(model = "RSM").
Arguments
- theta
Ability.
- bank
Matrix of item parameters, one row per item: column 1 is the location of the item \(\lambda_j\), the other columns the thresholds \(\delta_1\) to \(\delta_m\), the same in every row (which makes it the rating scale and not the partial credit model). A 5-point Likert item, scored 0 to 4, needs 5 columns.
- D
Metric constant, 1 or 1.702.
Value
A list: Pi, the probabilities, one row per item and one column per
category, and dPi, their first derivatives.
See also
Other rating scale model:
Ii_rsm(),
density_function(),
eap_est_rsm(),
eap_se_rsm(),
integrate_cat(),
likelihood_rsm(),
next_item_rsm()
Examples
bank <- matrix(c(-0.560, -0.230, 1.559, 0.071, 0.129,
rep(1.715, 5), rep(0.461, 5), rep(-1.265, 5), rep(-0.687, 5)),
nrow = 5,
dimnames = list(1:5, c("lambdaj", "delta1", "delta2", "delta3", "delta4")))
Pi_rsm(theta = 0, bank)$Pi
#> cat0 cat1 cat2 cat3 cat4
#> Item1 0.08827106 0.02781046 0.030704594 0.190454305 0.662759579
#> Item2 0.23732018 0.05375358 0.042666335 0.190263548 0.475996359
#> Item3 0.95403130 0.03611457 0.004790796 0.003570467 0.001492864
#> Item4 0.45604320 0.07644622 0.044906724 0.148203679 0.274400176
#> Item5 0.50098691 0.07924783 0.043929232 0.136808204 0.239027827
rowSums(Pi_rsm(theta = 0, bank)$Pi) # each item's probabilities add up to 1
#> Item1 Item2 Item3 Item4 Item5
#> 1 1 1 1 1