equate (version 0.1-0)

freqavg: Frequency Averaging

Description

This function replaces frequencies falling below a specified threshold with an average of adjacent values. Frequency averaging is a simple method for smoothing a score distribution, typically one with unobserved scores (for a description see Moses & Holland, 2008).

Usage

freqavg(freqtab, jmin = 1)

Arguments

freqtab

two-column matrix of class freqtab, where column 1 specifies the score scale and column 2 the frequencies for each score point

jmin

the minimum frequency, as an integer, below which frequencies will be replaced

Value

• Returns a matrix with 3 columns: the score scale, the original frequencies, and the adjusted frequencies. If no values fall below jmin the adjusted frequencies will simply be the original frequencies.

References

Moses, T., & Holland, P. (2008). Notes on a general framework for observed score equating (ETS Research Rep. No. RR-08-59). Princeton, NJ: ETS.

See Also

freqbump, loglinear

Examples

# generic frequency table:
set.seed(2005)
x <- round(rnorm(1000,100,10))
xscale <- 70:130
xtab <- freqtab(x,xscale)

# compare freqavg to freqbump:
xavg <- freqavg(xtab)
xbump <- freqbump(xtab[,2],Kx=60)
cbind(xscale,xavg,bump=round(xbump*sum(xtab[,2]),6))