icc.de.boot {iccde}

Bootstrap Confidence Intervals for Double-Entry Intraclass Correlations)

Description

For different applications, confidence intervals for the double- entry intraclass correlation can be useful. Bootstrap confidence intervals are computed by means of repeated resampling from the original data at hand.

Usage

icc.de.boot(data, n.sim = 1000, alpha = .05, digit = 3)

Arguments

data

A data frame with participants in rows and variables in columns. Users should restrict the data set to the variables of concrete interest because the all available information in the data frame will be used to compute bootstrapped confidence intervals. Thus, it is advisable to create a new data frame that entails only the variables on which the matrix of ICCDEs should be based. Computations in a data set with 56 variables on a Windows device (i5 CPU, 8 GB RAM) and a Macbook Pro (M3 Pro CPU, 36 GB RAM), for instance, took about two minutes, given 1,000 iterations.

n.sim

The number of iterations to be carried out. Default is 1,000.

alpha

Type I error. Default is .05.

digit

Number of digits in the output. Default is 3.

Value

The output provides a list of matrices of double-entry intraclass correlations.

Mean

A matrix of the bootstrapped point estimates of the bootstrapped double-entry intraclass correlations.

LL

A matrix providing the lower limits of the bootstrap confidence intervals, given the desired alpha level.

UL

A matrix providing the upper limits of the bootstrap confidence intervals, given the selected alpha level.

Author(s)

Christian Blötner, Michael Paul Grosz [e-mail]

References

Furr, R. M. (2010). The Double-Entry Intraclass Correlation as an index of profile similarity: Meaning, limitations, and alternatives. Journal of Personality Assessment, 92(1), 1–15. <https://doi.org/10.1080/00223890903379134>

McCrae, R. R. (2008). A note on some measures of profile agreement. Journal of Personality Assessment, 90(2), 105–109. <https://doi.org/10.1080/00223890701845104>

Examples

df <- data.frame(a = rnorm(100), b = rnorm(100), c = rnorm(100),
                 x = rnorm(100), y = rnorm(100), z = rnorm(100))
icc.de.boot(data = df,
            n.sim = 1000,
            alpha = .01,
            digit = 2)