Quality & Quantity 37: 393–409, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
Effect Sizes in Qualitative Research: A
ANTHONY J. ONWUEGBUZIE
Abstract. The American Psychological Association Task Force recommended that researchers al-
ways report and interpret effect sizes for quantitative data. However, no such recommendation was
made for qualitative data. Thus, the ﬁrst objective of the present paper is to provide a rationale for re-
porting and interpreting effect sizes in qualitative research. Arguments are presented that effect sizes
enhance the process of verstehen/hermeneutics advocated by interpretive researchers. The second
objective of this paper is to provide a typology of effect sizes in qualitative research. Examples
are given illustrating various applications of effect sizes. For instance, when conducting typological
analyses, qualitative analysts only identify emergent themes; yet, these themes can be quantitized
to ascertain the hierarchical structure of emergent themes. The ﬁnal objective is to illustrate how
inferential statistics can be utilized in qualitative data analyses. This can be accomplished by treating
words arising from individuals, or observations emerging from a particular setting, as sample units of
data that represent the total number of words/observations existing from that sample member/context.
Heuristic examples are provided to demonstrate how inferential statistics can be used to provide more
complex levels of verstehen than is presently undertaken in qualitative research.
Key words: effect sizes, qualitative research, quantitize, meta-theme, inter-respondent matrix, intra-
As stated in the May 2000 edition of the Educational Researcher, the theme of the
American Educational Research Association (AERA) 2001 annual meeting was
“What we know and how we know it” (AERA, 2000: 27). Moreover, AERA called
for “penetrating and weighty discussions around issues of research methodologies,
rigor, standards – within every research paradigm” (AERA, 2000: 27). An im-
portant step in determining “what we know and how we know it” in the ﬁeld of
education is to interpret ﬁndings within an appropriate educational context. One
method for contextualizing empirical data recommended by researchers is via the
use of effect sizes (Cohen, 1988).
Unfortunately, one of the most common errors in quantitative analyses in the
USA and elsewhere involves the incorrect interpretation of statistical signiﬁcance
and the related failure to report and to interpret effect sizes (i.e., variance-accounted
Correspondence should be addressed to Anthony J. Onwuegbuzie, Department of Human Devel-
opment and Psychoeducational Studies, School of Education, Howard University, 2441 Fourth Street,
NW, Washington, DC 20059, or E-Mail: (email@example.com). Phone number 202-806-7345
(work); 202-249-0416 (home); 202-249-0417 (Fax).