Sample Size in Behavioral Research: A Systematic Review of JEAB and JABA from 2009 to 2018

Resumo

The current research conducted a systematic review on sample size and the use of inferential statistics in basic and applied behavioral research by assessing all experimental studies from 2009 to 2018 in the Journal of the Experimental Analysis of Behavior (JEAB) and the Journal of Applied Behavior Analysis (JABA) which was 1,155 articles. The use or non-use of inferential statistics in behavioral research remains controversial as visual inspection has deep historical roots. JEAB had a median number of eight subjects and JABA had a median number of four subjects, which was statistically different using Mood’s median test. In addition, articles in JEAB were more likely to use inferential statistics compared to JABA.  In general, inferential statistics were used in the presence of larger sample sizes, however, the use of animal subjects was associated with smaller sample sizes. Although patterns of the use of inferential statistics varied across journal, sample size, and species, this does not preclude the use of statistical methods by applied behavioral researchers, which should be used to support and confirm visual inspections of data. 

Keywords: sample size, inferential statistics, behavioral research, systematic review.