The Use of Control Groups in Music Therapy Research: A Content Analysis of Articles in the Journal of Music Therapy

Abstract

The use of a control group is fundamental to experimental research design, though the use with clinical populations must be carefully considered. The purpose of this research was to examine the use of control groups in research with clinical and nonclinical populations published in Journal of Musical Therapy from 1964 through 2004. Criteria for inclusion were music or music therapy as an independent variable applied to one or more groups and at least one control group that did not receive a music treatment. Control groups were qualified as alternative treatment, placebo, no contact, and treatment as usual. Of the 692 articles, 94 met these criteria, 62 clinical and 32 nonclinical, representing 13.5% of the publications. Results indicated that research with clinical populations involved a mean of 38.1 subjects typically divided into two groups, an experimental and a control group. The pretest-posttest design was the most common (55%) as was a treatment as usual control group (45%). These design methods maximized the impact of the experimental music treatment on outcome. Experimental music groups significantly improved over control groups in the vast majority of studies identified. Undoubtedly, the foundation for evidence-based clinical practice is firm.