chen.thissen.1997: Computes local dependence indices for all pairs of items

Description

Item Factor Analysis makes two assumptions: (1) that the latent distribution is reasonably approximated by the multivariate Normal and (2) that items are conditionally independent. This test examines the second assumption. The presence of locally dependent items can inflate the precision of estimates causing a test to seem more accurate than it really is.

Usage

chen.thissen.1997(grp, data = NULL, inames = NULL, qwidth = 6,
  qpoints = 49, method = "rms")

Arguments

grp

a list with the spec, param, mean, and cov describing the group

data

inames

a subset of items to examine

qwidth

quadrature width

qpoints

number of equally spaced quadrature points

method

method to use to calculate P values. The default ("rms") uses the root mean square statistic (see ptw2011.gof.test). To obtain the traditional Pearson X^2 statistic, use method="pearson".

Value

a lower triangular matrix of log P values with the sign determined by relative association between the observed and expected tables (see ordinal.gamma)

Details

Statically significant entries suggest that the item pair has local dependence. Since log(.01)=-4.6, an absolute magitude of 5 is a reasonable cut-off. Positive entries indicate that the two items are more correlated than expected. These items may share an unaccounted for latent dimension. Consider a redesign of the items or the use of testlets for scoring. Negative entries indicate that the two items are less correlated than expected.

References

Chen, W.-H. & Thissen, D. (1997). Local dependence indexes for item pairs using Item Response Theory. Journal of Educational and Behavioral Statistics, 22(3), 265-289.

Wainer, H. & Kiely, G. L. (1987). Item clusters and computerized adaptive testing: A case for testlets. Journal of Educational measurement, 24(3), 185--201.