Defining Experimental Units: Agreement Awaiting Implementation Response to Johnson (2000) Letter to the Editor
Letter to the Editor  |   October 01, 2000
Defining Experimental Units: Agreement Awaiting Implementation
 
Author Affiliations & Notes
  • Ludo Max
    Department of Communication Sciences University of Connecticut Storrs, CT
  • Patrick Onghena
    Department of Educational Sciences Katholieke Universiteit Leuven Leuven, Belgium
Article Information
Hearing & Speech Perception / Research Issues, Methods & Evidence-Based Practice / Speech / Letters to the Editor
Letter to the Editor   |   October 01, 2000
Defining Experimental Units: Agreement Awaiting Implementation
Journal of Speech, Language, and Hearing Research, October 2000, Vol. 43, 1291-1293. doi:10.1044/jslhr.4305.1291
History: Received March 13, 2000 , Accepted May 31, 2000
 
Journal of Speech, Language, and Hearing Research, October 2000, Vol. 43, 1291-1293. doi:10.1044/jslhr.4305.1291
History: Received March 13, 2000; Accepted May 31, 2000
Defining Experimental Units: Agreement Awaiting Implementation: Response to Johnson (2000) 
In her letter (Johnson, 2000) Dr. Johnson expressed agreement with our position, described in the Max and Onghena (1999)  tutorial, that researchers in the areas of speech, language, and hearing should consider more carefully the various assumptions underlying the use of statistical tests. In the cited tutorial, we pointed out that the authors of several published research articles had insufficiently addressed or completely neglected some of those assumptions. Therefore, we discussed two frequently occurring problems (violation of the sphericity assumption and incorrect definition of experimental units), and we offered several suggestions intended to promote more accurate use of statistical tests in speech, language, and hearing research. Of course, our work did not, and could not, explicitly address every possible research design. Rather, we quoted Kirk's (1995, p. 1)  general description of an experimental unit as “that entity that is assigned to an experimental condition independently of other such entities,” and we stated that “for the vast majority of research studies in the fields of speech, language, and hearing, it follows that each experimental unit corresponds to one individual subject who is assigned to an experimental group and/or who is tested under a number of different experimental conditions” (Max & Onghena, 1999, p. 265).
First Page Preview
First page PDF preview
First page PDF preview ×
View Large
Order a Subscription
Pay Per View
Entire Journal of Speech, Language, and Hearing Research content & archive
24-hour access
This Article
24-hour access