each respondent by partitioning the respondent’s item responses into parallel half tests, computing the
variance between the halves, and then averaging the result over all respondents at the given score level.
Lord (1955) suggested that conditional standard errors of measurement could be derived by
application of the binomial error model. Qualls-Payne (1992) detailed how Keats (1957), Feldt (1984), and
Jarjoura (1986) refined Lord’s application of the binomial model. Lord himself has suggested another
refinement (Lord, 1984). By and large these refinements involve a compound binomial model; in general
they result in estimates of conditional measurement errors that are smaller than those obtained by applying
the simple binomial model.