Methods for the Behavioral, Educational, and Social Sciences: An R package

Abstract

Methods for the Behavioral, Educational, and Social Sciences (MBESS;Kelley, 2007b) is an open source package for R (R Development Core Team, 2007b), an open source statistical programming language and environment. MBESS implements methods that are not widely available elsewhere, yet are especially helpful for the idiosyncratic techniques used within the behavioral, educational, and social sciences. The major categories of functions are those that relate to confidence interval formation for noncentral t, F, and 2 parameters, confidence intervals for standardized effect sizes (which require noncentral distributions), and sample size planning issues from the power analytic and accuracy in parameter estimation perspectives. In addition, MBESS contains collections of other functions that should be helpful to substantive researchers and methodologists. MBESS is a long-term project that will continue to be updated and expanded so that important methods can continue to be made available to researchers in the behavioral, educational, and social sciences.

Mixed-format exams in higher education: Assessment of internal consistency reliability

“…We used the MBESS program (Dunn et al, 2014;Revelle & Zinbarg, 2009;Kelley, 2007) written for the R platform for statistical computing (Field et al, 2012) to estimate reliability coefficients alpha and omega. We used the normal bootstrapping method of estimating reliabilities, as it is known to be appropriate for small sample sizes (Padilla & Divers, 2013a, 2013b.…”

An R package for single-case randomization tests

“…Although this may seem a completely subjective procedure, the data are visually inspected according to a variety of UNIX platforms, as well as on Windows and MacOS (Hornik, 2007). It is a powerful tool for statistical modeling and is extremely flexible, which enables it to cope with difficult and unusual data sets and problems, making it ideal for tailoring calculations to one's own specific statistical requirements (Crawley, 2005;Kelley, 2007). This makes R the perfect software environment for devising one's own randomization tests adapted to the design of a particular study.…”

“…For instance, R can be used to easily calculate effect size measures, percentages of nonoverlapping data, and other indicators of the importance of observed systematic effects. Also, researchers can draw graphical indicators of the magnitude of an effect, thanks to the very well-developed graphical possibilities of the system (Dalgaard, 2002;Kelley, 2007).…”

Randomization tests for multiple-baseline designs: An extension of the SCRT-R package

“…As an open-source implementation of the S-PLUS language, it can be downloaded at no cost from the Comprehensive R Archive Network Web site (CRAN; cran.r-project.org). R is extremely flexible and can be used for statistical modeling as well as for graphical applications (Crawley, 2005;Dalgaard, 2002;Kelley, 2007).…”