The Rise of Qualitative Research in Psychology
DAVID L. RENNIE KIMBERLYD. WATSON ALTHEA M. MONTEIRO York University
Abstract A study is presented on the rise of qualitative research in psychology over the 20th century. The incidence of qualitative research as indicated by several search terms (i.e., "qualitative research," "grounded theory," "discourse analy*," "empirical phenomenological," and "phenomenological psychology") was traced through the PsycINFO and Dissertation Abstracts International databases. It was found that, with the exception of the search terms having to do with phenomenology, records containing these search terms were basically non-existent until the 1980s, when there was a sharp rise that intensified in the 1990s. The PsycINFO records were sorted according to (1) whether they came from psychology or other social and health science disciplines; (2) region of origin; (3) the types of document to which they referred; and (4) whether they focused on the methodology or the application of qualitative research. A number of interesting differences emerged from this comparative analysis. Implications of the findings for the supposition that a paradigm shift may be underway are discussed.
The term "qualitative research" refers to a variety of approaches to enquiry in the health and social sciences that address the meaning of verbal text in verbal rather than numerical terms. More fundamentally, qualitative research is more subjective than quantitative research; more exploratory than confirmatory; more descriptive than explanatory; more interpretive than positivist (see Denzin & Lincoln, 1994). Thus, in many ways qualitative research cuts across the grain of accustomed research practice. Accordingly, there is resistance to accepting it in many quarters. Nevertheless, a path has been cleared for it in part by the postmodern critique of the modern quest for objective knowledge. More fundamentally, perhaps, Canadian Psychology/Psychologic canadienne, 43:3
the appeal of qualitative research is so great for some investigators that they are engaging in it despite the resistance to it (McMullen, 2002; Stoppard, 2002). There seems to have been an uptake of qualitative research in psychology in recent years. It also seems that no one has examined the extent of this uptake. A number of questions are of interest: When did the turn to qualitative research in psychology begin and how has its growth developed? What kinds of publications constitute the resulting literature? Are there regional differences in the use of qualitative research? How much emphasis has been placed on qualitative research methodology as opposed to its application? It appeared to us that the answers to questions like these would have a bearing on the most interesting question of all: What is to be made of the arrival of qualitative research on the scene in terms of the predominance of the natural science approach to enquiry? Is its presence now large enough that it can be considered a major movement - the emergence of an alternative "paradigm," as Kuhn (1970; cf. McMullen, 2002; O'Neill, 2002) might put it? Or alternatively, is it the case that, under close scrutiny, it becomes evident that qualitative research has yet to make a significant impact on the psychology research infrastructure? In an attempt to address these questions, we decided to study the psychological literature produced in the 20th century. This decision presented a number of difficulties having to do with changes in the use of language over that period of time. This problem could have been addressed through the study of whole documents. To do that would have necessitated sampling in order to handle the volume of them, however. The alternative was to apply search terms to an electronic database. This approach had the appeal of being more comprehensive. Even here, though, such a study would be limited by the scope of the particular database and by the particular search terms used. Thus, neither approach is ideal but, on balance, given that we wished to do an exploratory study, we decided that the latter strategy is the better of the two and so turned to PsycINFO, produced by the American Psychological Association (APA). After applying a number of search terms (see below) to PsycINFO, we eventually chose five terms having to do with qualitative research. One was the
180 Rennie, Watson, and Monteiro blanket term, "'qualitative research." The other four terms pertained to particular methods and approaches. The first was "grounded theory," referring to the method developed by Glaser and Strauss (1967). The second was phenomenological psychology represented by the terms "phenomenological psychology" and "empirical phenomenology" (see, e.g., Fischer, 1998; Giorgi, 1970). Finally, "discourse analy*" was used to cover a number of approaches related to the analysis of discourse (see, e.g., Ibanez & Iniquez, 1997; Potter & Wetherell, 1987). Moreover, once using PsycINFO, our attention was drawn to its coverage of publications coming out of fields other than psychology. Hence, we traced the appearance of the terms in records coming out of these other fields to provide a rough comparison between them and psychology. Finally, upon the discovery that the database incompletely addressed dissertation abstracts, we applied the search terms to a second database, Dissertation Abstracts International, to get a better sense of the uptake of qualitative research by graduate students. We did not isolate psychology dissertations in this large search, however. As will be seen, the study indicated that the increase in publications entailing several of these search terms has been dramatic in the last two decades. It also revealed interesting variations in the use of different approaches to qualitative research, among the disciplines and regions that we studied. Method 1. PsycINFO In what follows, the term "record" refers to depiction in PsycINFO and Dissertation Abstracts International (DAI) of publications and dissertations. This depiction typically consists of the item's title, author(s), affiliation of the first author (but see below), source, date of publication/dissertation, abstract and major descriptors or key words. In the case of PsycINFO, records containing the search terms were scrutinized to determine if they in fact were pertinent to the search terms as we understood them. The records that survived this weeding are referred to as "hits." The searches were conducted with the "word anywhere" option, which scanned across all fields in the database. The initial search in PsycINFO included 16 word phrases - "qualitative analysis," "qualitative research," "grounded theory," "grounded analysis," "phenomenology," "empirical phenomenology," "empirical phenomenological," "phenomenological psychology," "discourse analy*," "discursive psycholog*," "content analysis," "case study," "hermeneutic," "interpretive," "interpret and research," and "interpret* and analysis." (The use of the asterisk
picked up various suffixes of the stem term, thus "analy*" caught "analytic," "analytical," and "analysis.") This approach had limitations. There were terms we did not use, such as "ethnograph*," "action research," and "heuristic research." Also, there may be disciplines such as anthropology that customarily use qualitative research methods of various sorts without putting any label on them because, for that discipline, the practice is taken for granted. Further, one of the terms we considered - case study - yielded 12,190 records where that term was used in a variety of ways, only some of which seemed to relate to qualitative research. It was our judgment that case studies, although around for a long time, have been considered in mainstream psychology as a source of theory development requiring scientific investigation; they have not been granted research status in their own right (see Kvale, 2001). Instead, it has been empirical phenomenology and grounded theory methodologists, especially, who have led the way in advocating that these methods are legitimate, alternative methods of research in psychology. In our interpretation, this assertion has been made on the ground that these methods allow for generalization, albeit of limited scope. Thus, it has only been recently that users of case studies have made a concerted bid to have this mode of enquiry counted as a legitimate form of research by integrating them into the qualitative research family, as in narrative analysis (e.g., McLeod & Balamoutsou, 1996; Polkinghorne, 1995). It also happens that much of discourse analysis involves case studies. The focus of this approach is often critical rather than empirical, however. Still, there are many who are engaged in this type of discourse analysis who believe that it is a form of qualitative research (see Reicher, 2000) and so, in using "discourse analy*" as a search term, we decided to go with this tide. On another front, an effort is being made by declared qualitative research methodologists to derive generalizations from case studies, as in ideal type analysis (e.g., Frommer & Langenbach, 2001; Stuhr & Wachholz, 2001). Overall, then, it was our judgment that the case study is being integrated into qualitative research more than being considered formative of it. It was thus our opinion that the term "case study" is not as important an indicator of qualitative research as are the search terms we finally decided on, which were "qualitative research," "grounded theory," "empirical phenomenological," "phenomenological psychology," and "discourse analy*." The searches were conducted by decade, from 1900 to 1999. We also obtained the total number of records in PsycINFO per decade, so that we could com-
Rise of Qualitative Research in Psychology 181 pare the growth of records containing the search terms to the growth of records as a whole. A total of 3,262 records were produced by the search terms for the period from 1900-1999 (see Table 3). We printed these records and studied the abstracts therein, in teams of two and sometimes three, in deciding whether or not to declare the record a hit. As will be seen, a residual of 2,467 hits resulted from this weeding (see Table 1). Judgments of what constituted a hit. In conducting the search, we had to develop a working definition of each term. For "qualitative research," we included any research that entailed an interpretive representation of meanings emerging from analysis of text and was expressed primarily in nonnumerical language. Thus, we excluded content analysis involving a priori categories and/or the cursory use of qualitative data (i.e., verbal text) in quantitative studies. The use of "grounded theory" was governed by evidence that the term applied to this method as a form of qualitative research, as opposed to any other use of the term, as in, say, a theory that is grounded without any reference to the method. The features of the method that we looked for were categorization as a form of representation of meaning of text; constant comparative analysis; theoretical memoing; and theoretical sampling. With regard to "empirical phenomenological," we drew on Fischer's (1998) depiction of this method as involving the conceptualization of structures of experience and conduct described in psychological language. The other two terms were comparatively more problematic, especially "discourse analy*." We found that in PsycINFO this stem applied both to qualitative and quantitative analysis of discourse. Moreover, a number of records having to do with psychoanalysis were called "discourse analysis" (e.g., when the analysand's discourse was referred to in the record) when the record simply referred to discourse rather than depicting what appeared to be an actual analysis of it. We judged that this use of the term extended it beyond the range of the broadest definition of "research," whether qualitative or quantitative. Thus, we found that we had to refer to the literature on discourse analysis repeatedly in our development of a working definition (the edited book by Ibanez & Iniguez, 1997, proved to be especially helpful). According to this definition, qualitative discourse analysis is the nonquantitative analysis of units of text larger than a sentence where the emphasis is on the discourse itself. It is not on the discourse as a representation of meaning interpreted to have been experienced by the author of the text, as is the focus in, say,
empirical phenomenology and the grounded theory method. Instead, it is directed more to the analysis of political, cultural, and social meanings evident in the text irrespective of the inner experience of its author. Nevertheless, we included as a hit for "discourse analy*" methods that strictly speaking were other than discourse analysis but which were labeled as discourse analysis by the PsycINFO compilers - methods such as narrative analysis, conversation analysis, ethnomethodology, and pragmatics. (On the other side of the coin, of course, because we did not search the last four methods as such, we would have missed those for which "discourse analy*" was not mentioned in the records.) We also had difficulty with the term "phenomenological psychology," finding many records that had more to do with the philosophy of mind than with psychology. It is possible that this development derived from Husserl's (1977/1925) lectures on phenomenological psychology. This work was Husserl's attempt to descend from philosophy to psychology and, going with it, from rationalism to empiricism, but this was a half-hearted attempt that had more to do with philosophy than with psychology despite his intention to do otherwise (see Scanlon's Introduction to the above book). Initially, we were inclined to exclude the former type of records but eventually changed our minds, on the ground that any application of the method of phenomenological analysis, whether rational or empirical, should count. By this we mean the attempt to bracket (i.e., suspend) predisposing conceptualizations of the phenomenon under study in the interest of describing its structure. Other records had to do with existential-phenomenological psychotherapy theory, which we did exclude. Procedurally, for each search term, either Monteiro or Watson did an initial examination of the records pertaining to a given search term to determine its fit. Rennie then checked these judgments. In the case of "discourse analy*," the weeding and checking were done twice. When weeding through the records, we noticed that there was some overlap in search terms (e.g., a given record would contain references to both "grounded theory" and "qualitative research"). To account for overlaps, an additional PsycINFO literature search was conducted in which each two-way combination of the five search terms was entered. The returns of these searches were recorded. We decided against going through the hits again to choose between competing terms when they overlapped, however, on the ground that doing so would not appreciably alter the overall picture (more on this in the Results section). Classifying the hits. In a subsequent textual analysis, we
182 Rennie, Watson, and Monteiro examined all hits and noted information on each one. We developed the following categories. Discipline referred to either Psychology or Other. Document Type meant whether the hit was a Book, Chapter, Journal, or Dissertation. Region applied to North America (i.e., Canada and the USA); the United Kingdom and Ireland; Scandinavia (i.e., Finland, Sweden, Norway and Denmark); Australia and New Zealand; Germanic countries (i.e., Germany, Austria, and the Netherlands); South Africa; or Other. Focus referred to either Methodology or Application of a method. In the case of chapters and books and edited books, the categorization of Discipline was based on the first author's affiliation. This decision was not ideal because, of course, it is possible for nonpsychologists to be affiliated in a psychological setting and for psychologists to be affiliated in nonpsychological settings. To complicate matters further, author affiliations are not listed consistently in PsycINFO. When author affiliations were not listed for books and chapters, we conducted a search of the author's name in PsycINFO, to see if the affiliation was listed in other publications. If this was not fruitful, then the second author's or editor's affiliation was sought. Journal articles were too numerous to allow for such a return to the database each time the information was not present, so in this case the categorization of Discipline was based on the nature of the journal rather than author affiliation. Again, this tactic was not ideal because nonmembers of a discipline sometimes publish in journals outside it. Most journals were easily categorized but for some interdisciplinary journals a judgment was required and we recognize that some of our classifications of journals could be disputed. All of this meant that our estimates of affiliations are only approximate. Finally, for the purpose of analysis, the fields other than psychology were collapsed into "Other." Dissertation records in PsycINFO did not include authors' affiliations so we relied on the discipline code. Specifically, we learned that all dissertations in PsycINFO were classified as "psychology." As a result, we conducted a supplemental literature search in the Dissertation Abstracts International (DAf) database, the details of which are described below. Author affiliation was also used to determine the hit's geographical Region. When it was impossible to classify Region in this way, it was assigned to an "Other" category. The decision about a hit's Document Type was usually straightforward because it was based on whether the publication was a book, chapter, journal article, or dissertation. If all of an edited book was relevant to the search term, then the hit was counted as a Book.
If only certain chapters from an edited book were relevant, then each of the relevant chapters was counted as a Chapter. 2. Dissertation Abstracts International (DAI) Because PsycINFO picks up dissertations coming out of psychology, only, we decided to conduct a supplementary literature search in DAI to address qualitative research activity evident in dissertations emerging from other disciplines as well. Again, a search was conducted for each of the five search terms, by decade from 1900 to 1999. We noted the number of records for each search term but we did not print records, weed through them, or collect data on them, given that they totalled to 7,146. We also obtained the total number of records for each decade in DAI, to get an indication of the rise of search terms in this regard compared to the rise in dissertation records taken as a whole. Finally, we did a search of the overlap among search terms, limited to combinations of two terms. Results OCCURRENCE OF THE SEARCH TERMS: 1900-1999 PsycINFO. The data for the five search terms - "qualitative research," "grounded theory," "discourse analy*," "phenomenological psychology," and "empirical phenomenological" - in the PsycINFO database during four time intervals since 1900 (i.e., 1900-1969, 1970-1979, 1980-1989, and 1990-1999) are shown in Table 1. It can be seen that the pattern of hits for the two search terms having to do with phenomenology is different than that for each of the other three terms. First, it is only "phenomenological psychology" that appeared in the literature with any frequency of note prior to 1970. This is likely explained by the history of this term, which goes back to 1925 when Husserl paid attention to it (see above). Second, it is evident that the term "empirical phenomenological" started to appear in the 1970s comparatively more than did "qualitative research," "grounded theory," and "discourse analy*." By that time, Giorgi's (1970) pioneering effort to apply phenomenology empirically to psychological phenomena was beginning to take effect, in advance of other approaches to qualitative research, certainly in psychology. Finally, it is apparent that the pattern of hits for the two search terms having to do with phenomenology is similar for the '70s, '80s, and '90s. As Table 1 indicates, then, apart from "phenomenological psychology" and "empirical phenomenology," a sharp rise in the appearance of our search terms began in the '80s, and increased further in the '90s. It can be seen that this rise across these two decades far outstrips the increase in records as a whole.
Rise of Qualitative Research in Psychology 183 TABLE 1 Growth of Search Term Hits in PsycINFO Database SEARCH TERMS Decade
Total Records
Qualitative Research
Grounded Theory
1900-1969
347,545
Psych Other
1 4
Psych Other
1970-1979
270,445
Psych Other
3 0
1980-1989
442,551
Psych Other
1990-1999
543,743
Psych Other
64 72 324 460
Phenomenology PP EP Totals
Discourse Analysis Psych Other
24 2
2 0
27 8
Psych Other
0 2 4 3
Psych Other
88 3
12 0
108 7
21 27
Psych Other
28 26
Psych Other
41 0
229 127
178 216
Psych Other
408 265
Psych Other
75 2 69 1
39 1
1,018 943
Psych Psych 200 Psych 440 Psych 256 392 Other 536 Other 244 Other 296 Other 8 PP = Phenomenological Psychology; EP = Empirical Phenomenological; Psych = Psychology; Other = Non Psychology.
94 1
1,382 1,085
Totals
Psych Other
Psych Other
0 0 1 1
Psych Other Psych Other
1,604,284
TABLE 2 Growth of Search Term Records in Dissertation Abstracts International Database SEARCH TERMS Decade
Total Records
Qualitative Research
Grounded Theory
Discourse Analysis
1900-1969 1970-1979 1980-1989 1990-1999 Totals
351,780 356,801 394,470 566,258 1,669,309
0 1 556
0 13 582
2,726 3,283
2,135 2,730
0 10 253 692 955
Phenomenology PP
EP
3 9 17 19 48
0 8 73 49 130
Totals 3 41
1,481 5,621 7,146
PP = Phenomenological Psychology; EP = Empirical Phenomenological.
Specifically, the number of total records for all publications increased by a factor of 2.3 between the 1970s and the 1990s. In contrast, the number of hits for "qualitative research" went from 3 to 784; "grounded theory" from 2 to 394; and "discourse analy*" from 7 to 673. Dissertation Abstracts International. As indicated above, PsycINFO addresses dissertations coming out of the discipline of psychology only, whereas the database addresses books, chapters, and journals from many other disciplines as well. It was also indicated above that those dissertations entailing our search terms are included in Table 1. Having gone to DAI, then, the returns from our search for the terms in this database are shown in Table 2. It must be remembered that, in this analysis, we made no attempt to isolate "other" dissertations from "psychology" dissertations. Nor did we look for false positives in the records. Thus, these figures must be inflationary compared to those from PsycINFO. Even so, the data in this table relate well to those in Table 1 in showing that a sharp rise in the search terms other than the ones to do with phenomenological psychology began in the '80s.
Overlap of search terms. As indicated, the data in the above two tables represent the returns from a procedure in which each of the five terms was searched separately. This procedure thus did not control for overlap between terms in the same record in a database. A subsequent analysis was done to examine the extent of the overlap in the records prior to any weeding for false positives. The results of this analysis are shown in Table 3. In terms of PsycINFO, it can be seen that in Part 1 of the table there were two main overlaps. The first was between the term "qualitative research" and the other search terms, as would be expected. There were 118 instances in which the terms depicting the four approaches to qualitative research overlapped with "qualitative research." The second entailed 13 instances of overlap between "phenomenological psychology" and "empirical phenomenological." Given that the total number of records for "qualitative research" was 928, and for "phenomenological psychology" was 264, the overlaps amount to 12.2% in the first case and 5% in the second. Thus, the incidence of hits needs to be revised downwards to that extent. Given that our interest is exploratory, howev-
184 Rennie, Watson, and Monteiro TABLES Overlap of Search Terms 1900-1999 1. PsydNFO Database Grounded Theory (527) Grounded Theory (527) Qualitative Research (1,160) Discourse Analy* (1,101) Phenom Psych (366) Empirical Phenomen (108)
Qualitative Research (1,160)
Discourse Analy* (1,101)
Phenomen Psych (366)
Empirical Phenomen (108)
76 — — — —
5 32 — — —
1 8 0 — —
0 0 13 —
Qualitative Research (3,283)
Discourse Analy* (955)
Phenomen Psych (48)
Empirical Phenom (130)
256 — — — —
8 11 — — —
0 1
0 3 0 1 —
— — — — —
2
2. Dissertation Abstracts International Database
Grounded Theory (2,730) Grounded Theory (2,730) — Qualitative Research (3,283) — Discourse Analy* (955) — Phenom Psych (48) — Empirical Phenom (130) —
2 — —
Note: The numbers in brackets are the total number of records.
TABLE 4 Hits in Terms of Source, Region, and Type of Publication in Psychology and Other Disciplines (PsydNFO) Source
Disc
Region
Book
Chap
Jrnl
Diss
NA
UK+ Irlnd
Sc
32 90
30 30
308 414
22 2
249 383
68 66
19 15
Au + Nz
Ger
S.Afr
Other
Mthd
App
11
14 12
10 1
21 32
213 263
179 273
0 1
0 1
3 6
29 30
171 214
19 15
11 0
53 32
113 46
327 250
0
12 1
6 0
34 1
114 5
142 3
0 0
0 0
2 0
2
3 0
91 1
Qual Research
Other
Grounded Theory
Psy
1
7
16 13
99 193
84 31
161 210
19 6
8 13
9
Other
Discourse Analysis
Psy
50 23
70 20
296 244
24 9
175 173
144 40
15 14
23
Other
Phenomen Psych
Psy
17
2
6 0
226 3
7
Other
173 6
18 0
11 0
Empirical Phenom
Psy
5 0
0 0
17 1
72
Other
87 1
1 0
2 0
Psy
3 0
Focus
27 7
22 2
0
Disc = Discipline (Psychology or Other); Chap = Chapter; Jrnl = Journal; Diss = Dissertation; NA = North America (US and Canada); UK + Irlnd = United Kingdom and Ireland; SC = Scandinavia (including Sweden, Norway, Finland, and Denmark); AU + NZ = Australia and New Zealand; Ger = Germany, Austria, and the Netherlands; S.Afr = South Africa; Other = Other countries including Belgium, Brazil, Chile, China, Czechoslovakia, France, Greece, Hong Kong, Hungary, India, Iran, Israel, Italy, Japan, Korea, Mexico, Namibia, Poland, Portugal, Singapore, Spain, Switzerland, Taiwan, Tanzania, and Unknown Countries; Mthd = Methodology; App = Application; Qual = Qualitative; Psy = Psychology; Other Disciplines = Disciplines other than psychology; Phenom Psych = Phenomenological Psychology; Empirical Phenom = Empirical Phenomenology.
er, Table 1 still gives a reasonably accurate portrayal of the rise in incidence of the search terms. Regarding DAI, Part 2 of Table 3 shows that the greatest overlap was between "grounded theory" and "qualitative research," which amounted to 9.4% of the records for the former term (see Table 2). Apart from these two terms, the overlap was small. Indeed, there are only 11 overlaps between "discourse analy*" and
the 3,283 records for "qualitative research." We find this discrepancy between grounded theory and discourse analysis to be interesting and wonder if it reflects less affiliation with qualitative research among graduate students doing discourse analysis, the above remarks about the claimed inclusion of discourse analysis in qualitative research notwithstanding. In any case, taken as a whole the analysis of overlaps in
Rise of Qualitative Research in Psychology 185 TABLE 5 Psychology Journals Publishing five or More Articles Involving the Search Terms: JOURNAL Addiction (Health) American Journal of Community Psychology British Journal of Guidance and Counselling British Journal of Social Psychology Canadian Psychology Counseling Psychologist Counselling Psychology Quarterly Disability, Handicap and Society (Rehabilitation) Discourse Processes (Language and Communication) Feminism and Psychology Human Communication Research Humanistic Psychologist Journal of Applied Behavioral Science Journal of Communication Disorders Journal of Community and Applied Social Psychology Journal of Counseling Psychology Journal of Family Therapy (Counselling) Journal of Health Psychology Journal of Language and Social Psychology Journal of Marital and Family therapy (Counselling) Journal of Phenomenological Psychology Journal of Pragmatics (Language and Communication) Journal of the Association for Persons with Severe Handicaps (Health) Narrative Inquiry (Language and Communication) Philosophical Psychology Psychoanalytic Dialogues Psychologist Psychology and Health South African Journal of Psychology Sport Psychologist Theory & Psychology Topics in Language Disorders
QR
GT
DA
PP
EP
2
2
181
8
8 5 5 11 10
7 4 4
3 1 4 3
4
7 2 8
3 15 11 6
1 4 1
7 1
1
5
3 6
5 9 1 3
1
7 10
6 6 54 5 6 6
7 7 1
2
6 3
2
1
3
2
2 11 7 3
Note: Focus of some journals indicated in parentheses. QR = "qualitative research"; GT = "grounded theory"; DA = "discourse analy* "; PP = "phenomenological psychology"; EP = "empirical phenomenological ".
the DAI search does not appreciably alter the picture of growth given in Table 2, despite our having chosen terms that in some cases were expected to overlap. Type of document, region of origin, type of publication, and focus of publication: Psychology and other disciplines (PsycINFO only). Table 4 gives a breakdown of the hits of the five search terms regarding the type of document (book, chapter, journal article or dissertation); region yielding the document; and focus of the document (emphasis on methodology vs. application of method). In each case, a further breakdown is made between what, we decided, are "psychology" versus "other" publications. The values entered into this table are collapsed across the time intervals used in Table 1. It is important to mention that the values given in the table with respect to dissertations refer to those cited by PsycINFO, not to the returns from our
search of DAI. (We used the former database for this table despite its incompleteness relative to what the latter search gave because to delete the dissertation data would have required a re-analysis of all our PsycINFO data.) The comparisons that stand out for us are as follows. First, the term "qualitative research" was found more in the "other" than in the "psychology" publications (536 vs. 392). Second, in all regions except the United Kingdom and Ireland, the hits for "discourse analy*" were fairly evenly distributed between the discipline of psychology and the other fields (296 vs. 256). In the UK and Ireland, on the other hand, "discourse analy*" was much more pertinent to psychology than to the other fields (144 vs. 40). Third, the terms "phenomenological psychology" and "empirical phenomenological" overwhelming appeared in psychology publications and dissertations rather than in
186 Rennie, Watson, and Monteiro TABLE 6 Nonpsychology Journals Publishing 5 or More Articles Involving the Search Terms JOURNAL Advances in Nursing American Journal of Occupational Therapy (Health) Archives of Psychiatric Nursing Exceptional Children (Education) Family Practice (Health) Health Care For Women International Health Education Research IMAGE: Journal for Nursing Scholarship International Journal of Linguistics (Language and Communication) International Journal of Nursing Studies Issues in Mental Health Nursing (Nursing) Journal of Research in Science Teaching (Education) Journal of Social Issues Journal of Market Research Society Journal of Narrative and Life History (Language and Communication) Language in Society Modern Language Journal Music Therapy (Counselling) Nursing Research Qualitative Health Research Research in Nursing and Health (Nursing) Review of Educational Research Scandinavian Journal of Caring Sciences (Health) Scholarly Inquiry for Nursing Practice Social Science and Medicine (Health) Teaching and Teacher's Education Western Journal of Nursing Research
QR
GT
DA
2 7 1
6 5 9
2
5 6 1 10 4
4 10 1 13
PP
EP
1 1 1 5
2 3 1 4 21
3 4 3
2 1 3 1 33 14 6
5 2 1 78 8 5 3 1 13 1
12 29 9 8 5 5
4
1 5
8
Note: Focus of some journals indicated in parentheses. QR = "qualitative research"; GT = "grounded theory"; DA = "discourse analy* "; PP = "phenomenological psychology"; EP = "empirical phenomenological."
those coming from other fields (350 vs. 9). Also, the "phenomenological psychology" and "empirical phenomenology" publications and dissertations came mainly from North Americans (269 vs. 92). Finally, publications involving the "grounded theory" and "empirical phenomenology" search terms addressed the application of these methods more than their methodologies. Also, publications entailing "discourse analy*" that focused on methodology were more frequent in psychology than in other disciplines. JOURNALS PUBLISHING THE SEARCH TERMS We focused more narrowly on just one of the four document types -journals - and broke them into "psychology" vs. "other" journals. All told, as of the fall of 1999, PsycINFOhad records for 1,490journals in both of these categories combined. Of these, 496 (32.3%) published at least one article picked up by our search terms as we defined them. As indicated, we felt that it was important to distinguish between journals, books, and chapters that seemed to come from inside as opposed to outside psychology, even though the latter were included in PsycINFO. (For
examples of our breakdown of publications, compare Tables 5 and 6 below.) Because only those publications from outside psychology deemed relevant to psychology are recorded by this database, however, our "other" category represents the qualitative research publications in these other fields less than does our "psychology" category as a representation of publications in the discipline of psychology. Psychology. There were 315 journals that we judged to pertain to the field of psychology. There were 946 hits of the search terms among these journals (see Table 4). The distribution of hits was as follows: range = 1195; median = between 1 and 2; mode = 1; mean = 3.0. The high end of the range resulted from the hits for the phenomenological search terms appearing in the Journal of Phenomenological Psychology. Accordingly, the mode and median are more representative than the mean as indicators of the typical extent to which one or more of the five search terms appeared in each of the 315 journals. In order to isolate the journals that were most active in the publication of the articles involving the
Rise of Qualitative Research in Psychology 187 search terms, we set a cut-off of five hits or more for all search terms taken collectively, based on our sense of the literature. Table 5 shows the frequency of hits of each of the five search terms for the 32 journals meeting this criterion. The term "qualitative research" was a hit at least once for 22 of these journals; "discourse analy*" for 22; "grounded theory" for 12; "phenomenological psychology" for 2; and "empirical phenomenological" for 2. Among the journals involved, seven have to do with language and communication; seven with counselling and psychotherapy; six with health, rehabilitation or sport psychology; and two with community psychology. The remaining 11 are mixed. Lastly, of what may be described as flagship journals of national psychological associations, only Canadian Psychology and the South African Journal of Psychology were in this group of journals that published at least five articles involving the search terms. Other journals. According to our judgment, 181 of the journals listed by PsycINFO, and for which our search terms were relevant, had to do with fields outside psychology. As can be determined from Table 4, the total number of hits for these "other" journals was 855. Across these journals, the distribution of hits was as follows: range = 1-211; mode = 1; median = between 1 and 2; mean = 4.7. The high end of the range was caused by the hits for "qualitative research" in Qualitative Health Research. Thus, this particular journal created a large skew in the average number of hits, analogous to the skew created by the Journal of Phenomenological Psychology in the case of the "psychology" journals. Accordingly, the mode and the median are the best indicators of central tendency for these "other" journals as well, and, as is evident, these values are quite small. Parallel to the approach used for the "psychology" journals, for these "other" journals we isolated those that had five or more hits of the five search terms taken collectively. The results are given in Table 6. It can be seen that 27 of the 181 journals met this criterion. The term "qualitative research" was a hit at least once for 22 of them; "grounded theory" for 17; "discourse analy*" for 15; while "phenomenological psychology" and "empirical phenomenological" occurred in none. Thus, the pattern for these journals is different than that of the "psychology" journals in the sense that reference to grounded theory is now on par with discourse analysis, and there was no reference at all to the use of the phenomenological approach to qualitative research. Lastly, when these "other" journals were classified according to disciplinary fields, three predominated, namely Nursing = 9, other Health disciplines = 7, and Education = 4.
Discussion Strictly speaking, this study was about the rise of the appearance of five search terms judged by us to be pertinent to qualitative research more than it was about the rise of qualitative research per se. Our most inclusive term was "qualitative research," which we employed knowing that it would overlap to some extent with our other search terms. As for particular qualitative research methods, after experimenting with a number of search terms, we settled on terms having to do with grounded theory, discourse analysis, phenomenological psychology, and empirical phenomenology. We also included conversation analysis, narrative analysis, and pragmatics into discourse analysis whenever they appeared under the "discourse analy*" search term but did not search them in their own right. On the whole, doubtless many documents were missed in our search. Despite the many limitations of the study, however, it is likely that our findings reflect an actual marked rise of qualitative research. Before turning to PsycINFO, we had the impression that the uptake of qualitative research has been greater in some fields other than psychology. Thus, the coverage given by the database to these other fields offered a chance to check on that impression. And so we made the partition between psychology and other fields, knowing that it would underrepresent the qualitative research activity in those fields because of the only partial coverage of them given by this particular database. It was thus interesting to find that, despite this limitation, grounded theory evidently is used more in nursing than in psychology (compare Tables 5 and 6). Alternatively, the approach favoured by psychologists taken as a whole would appear to be discourse analysis. In this regard, it is useful to keep in mind that psychologists engaging in discourse analysis in the United Kingdom draw heavily on poststructural and postmodern thought, whereas those in the United States tend to be more modernist (Reicher, 2000; Roiser, 1997). More broadly, we are rather surprised by what seems to be the strength of qualitative research in other health fields in addition to nursing. On the other side of the coin, we notice few qualitative research indicators for sociology and social work. We found only one hit among our search terms in the American Journal of Sociology and five hits distributed among Social Work Research, Social Work Research and Abstracts, and Social Work with Groups. These small frequencies are in line with what we found for two prestigious APA journals that are traditionally positivistic (i.e., the Journal of Personality and Social Psychology and the Journal of Consulting and Clinical Psychology, which
188 Rennie, Watson, and Monteiro had one hit each). It is possible, of course, that PsycINFO has under-represented coverage of journals in these two disciplines compared to health care fields. Admittedly, precisely because DAI applies to all disciplines, it offered an opportunity to make an objective estimate of the qualitative research activity in psychology versus other fields, at least in terms of the work by graduate students. However, to do so would have necessitated poring over more than twice as many abstracts as we derived from PsycINFO and studied. Moreover, just as PsycINFO reported studies coming from outside the field of psychology, there are databases such as MedLINE and ERIC directed to other fields that include studies by psychologists. Ideally, they would have been used as well. We commend others to these tasks. We were not up to them. Overall, the rise in qualitative research as reflected by three of our search terms, at least, has been dramatic, especially in the last 10 years or so. It is crucial, however, to look at the growth in the right way. On the one hand, it is the case that, according to even only one database (i.e., PsycINFO) qualitative research papers made their way into 496 journals in and outside psychology, by the end of 1999. On the other hand, for most of these journals the modal number of publications was one. This record is hardly a sign of a major shift by these journals toward publishing articles involving our search terms. Instead, the growth of publication has been heavily swayed by the production of just a few journals that were created for qualitative research - the Journal of Phenomenology Psychology and Qualitative Health Research in particular. Tucked well behind that impact, but still by a large margin greater than is characteristic of the 496 journals taken as a whole, is the contribution of those journals that we tabulated in Tables 5 and 6. A few of these are journals that have been firmly located in the mainstream and are thus noteworthy. We are thinking especially of the Journal of Counseling Psychology and Canadian Psychology, with 14 and 20 hits, respectively. But their publication of qualitative research articles reflects the sentiments of the editors they have had over the past 10-15 years, at least in part. Thus, the former has been under the editorship of Clara Hill, whose own research has taken a qualitative turn (see Hill, Thompson, & Williams, 1997). Meanwhile, two of the editors of Canadian Psychology from the '80s on have been John Conway and Patrick O'Neill, both of whom have practiced qualitative research (e.g., McMullen & Conway, 1994; O'Neill, 1998). In making the above sobering remarks, we do not wish to go too far toward scepticism. Whatever is the situation regarding journals, the increases in hits of
our search terms pertained to books and chapters as much as to journal articles (except for the phenomenological search terms). There is no question that there is a ready market these days for such material, as reflected in the emphasis given it by the publisher, Sage. The publication of books and chapters is a good way to bypass journal editorial prejudices. As books and chapters mount, they signal a shift in the wind. Graduate students take notice and force their supervisors to take notice along with them (McMullen, 2002; Rennie, Watson, & Monteiro, 2000), and courses in qualitative research begin to be offered (Stoppard, 2002). Professional conferences start getting proposals for qualitative research papers and put them on their programs as a result. Eventually, journal editors start to take notice. Some of them set up special sections or issues, as a way of stepping in the water. Others go so far as to invite qualitative research manuscripts for regular issues. All of this has been happening but it is still too early to declare that a rupture in the dominance of positivism is now underway, at least according to our findings. The number of PsycINFO hits involving our search terms (including overlaps among them) in the 1990s, constituted only 0.45% of the total number of PsycINFO records for that decade. As we have emphasized throughout, our study is directed toward the comparative rise of the appearance of five search terms across decades rather than to the actual number of publications having to do with qualitative research. But, on that score, even if our estimation of qualitative research activity in the '90s underestimates the actual activity by a factor of 20, the actual activity would still only amount to 9% of the whole. In this light, whether the increase in qualitative research represents the beginning of a significant shift in the conduct of psychological research as suggested by O'Neill (2002) remains to be seen. This article is derived from a paper presented in a panel on "Qualitative Psychology: History, Theory and Practice" (Michael Krai, Chair) at the annual meeting of the Canadian Psychological Association, Ottawa, June 2000. The authors appreciate the support given this study by the Faculty of Arts, York University and comments by anonymous reviewers on an earlier draft. Correspondence may be directed to David Rennie, Department of Psychology, York University, Toronto, Ontario, Canada M3J IPS. (E-mail:
[email protected]).
Rise of Qualitative Research in Psychology 189
Resume Une etude est presentee sur la montee de la recherche qualitative en psychologic au cours du XXe siecle. L'incidence de la recherche qualitative a etc revelee par la recherche de plusieurs mots-cles (p.ex. « qualitative research », « grounded theory », « discourse analy* », « empirical phenomenological » et « phenomenological psychology ») dans les bases de donnees PsycINFO et Dissertation Abstracts International. On a trouve, qu'a 1'exception des termes de recherche portant sur la phenomenologie, les notices bibliographiques contenant ces termes de recherche n'existaient pas a toute fin pratique avant les annees quatre-vingts, ou on a connu une montee prononcee qui s'est intensifiee au cours des annee quatre-vingt-dix. Les notices PsycINFO ont ete triees en fonction des criteres suivants : 1) si elles provenaient de la psychologic ou d'autres disciplines en sciences sociales ou de la sante, 2) la region d'origine, 3) les types de documents auxquels il est fait reference et 4) si ces documents portaient sur la methodologie ou 1'application de la recherche qualitative. Un certain nombre de differences interessantes ont emerge de cette analyse comparative. Les consequences de ces decouvertes a 1'effet qu'un changement de paradigme pourrait etre en cours sont abordees.
References Denzin, N.K., & Lincoln, YS. (Eds.). (1994). Handbook of qualitative research. Thousand Oaks, CA/London: Sage. Fischer, C.T. (1998). Phenomenological, existential, and humanistic foundations for psychology as a human science. In M. Hersen & A. Bellack (Eds.), Comprehensive clinical psychology, Vol. 1: Foundations (C.W. Walker, Ed.). London: Elsevier Science. Frommer,J., & Langenbach, M. (2001). The psychoanalytic case study as a source of epistemic knowledge. In J. Frommer & D. Rennie (Eds.), Qualitative psychotherapy research: Methods and methodology (pp. 153-168). Lengerich, Germany: Pabst. Giorgi, A. (1970). Psychology as a human science: A phenomenologically based approach. Pittsburgh, PA: Duquesne University Press. Glaser, B. G., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine. Hill, C.E., Thompson, B.J., & Williams, E.N. (1997). A guide to conducting consensual qualitative research. The Counseling Psychologist, 25, 517-572. Husserl, E. (1977). Phenomenological psychology: Lectures, summer semester 1925. Translated and introduced by J.
Scanlon. The Hague: Nijhoff. (Original publication 1925). Ibanez, T, & Iniguez, L. (Eds.). (1997). Critical social psychology. London/Thousand Oaks, CA: Sage. Kuhn, T. (1970). The structure of scientific revolutions (2nd ed.). Chicago, IL: University of Chicago Press. Kvale, S. (2001). The psychoanalytic interview as qualitative research. InJ. Frommer & D. Rennie (Eds.), Qualitative psychotherapy research: Methods and methodology (pp. 9-31). Langerich, Germany: Pabst. McLeod, J., & Balamoutsou, S. (1996). Representing narrative process in psychotherapy: Qualitative analysis of a single case. Counselling Psychology Quarterly, 9, 61-76. McMullen, L. (2002). Learning the languages of research: Transcending illiteracy and indifference. Canadian Psychology, 43, 195-204. McMullen, L.M., & Conway, J.B (1994). Dominance and nurturance in the figurative expressions of psychotherapy clients. Psychotherapy Research, 4, 43-57. O'Neill, P. (1998). Negotiating consent in psychotherapy. New York/London: New York University Press. O'Neill, P. (2002). Tectonic change: The qualitative paradigm in psychology. Canadian Psychology, 43, 190-194.. Polkinghorne, D.E. (1995). Narrative configuration in qualitative analysis. Qualitative Studies in Education, 8, 523. Potter, J., &Wetherell, M. (1987). Discourse and social psychology: Beyond attitudes and behaviour. London/ Newbury Park, CA: Sage. Reicher, S. (2000). Against methodolatry: Some comments on Elliott, Fischer, and Rennie. British Journal of Clinical Psychology, 39, 1-6. Rennie, D.L., Watson, K.D., & Monteiro, A.M. (2000). Qualitative research in Canadian psychology. Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 1 (2): http://qualitative-research.net/fqs-e/200inhalt-e.htm Roiser, M. (1997). Postmodernism, postmodernity and social psychology. In T. Ibanez & L. Iniguez (Eds.), Critical social psychology (pp. 95-110). London/ Thousand Oaks, CA: Sage. Stoppard,J. (2002). Navigating the hazards of orthodoxy: Introducing a graduate course on qualitative methods into the psychology curriculum. Canadian Psychology, 43, 143-153. Stuhr, U., &Wachholz, S. (2001). In search of a psychoanalytic research strategy: The concept of ideal types. InJ. Frommer & D. Rennie (Eds.), Qualitative psychotherapy research: Methods and methodology (pp. 153168). Lengerich, Germany: Pabst.
