View source: R/estimate_valid_function.R
estimate_valid | R Documentation |
estimate_valid
estimates the nomination validity of an identification system
given the observed scores of the identified students, the confirmatory test
cutoff, the nomination cutoff, and the proportion of students that are
identified. The test cutoff, if not known, is inferred from the minimum score.
The nomination cutoff is inferred from the proportion of students who are nominated
under the assumption that the nomination x follow a standard normal distribution.
estimate_valid(x, nom.rate, id.rate, pop.mean = 0, pop.sd = 1, adjust = 1)
x |
Numeric vector of observed scores. |
nom.rate |
The proportion of students who have been nominated. Range (0, 1). Used to calculate the nomination cutoff. |
id.rate |
The proportion of students who have been identified. Range (0, 1). Must
be less than or equal to |
pop.mean |
The known general population mean of the x. Defaults to 0. |
pop.sd |
The known general population standard deviation of the x. Defaults to 1. |
adjust |
Controls the bandwidth of the density estimator. Defaults to 1.0, which has been found to perform well in simulation. |
The function uses the Levenburg-Marquardt algorithm to minimize the discrepancy
between the density of the x and the theoretical unnormalized density of the
data. See d_identified
for details.
# generate some observed scores # (note the lack of a relyt argument) # true validity is .6 set.seed(1) x <- r_identified( n = 500, test.cutoff = .9, valid = .6, nom.cutoff = .85 ) # calculate the identification rate implied by the system parameters id.rate <- marginal_psychometrics( test.cutoff = .9, valid = .6, nom.cutoff = .85 )$identification.rate # calculate the nomination rate implied by the system parameters nom.rate <- marginal_psychometrics( test.cutoff = .9, valid = .6, nom.cutoff = .85 )$nom.rate # estimate the system parameters from the data estimate_valid(x = x, id.rate = id.rate, nom.rate = nom.rate)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.