Robert J. Fleming Samuel P. Graci
and
Joel E. Thompson NORTHERN MICHIGAN UNIVERSITY

THE DAWNING OF THE AGE OF QUANTITATIVE/EMPIRICAL METHODS IN ACCOUNTING RESEARCH: EVIDENCE FROM THE LEADING AUTHORS OF THE ACCOUNTING REVIEW, 19661985

Abstract: This study documents changes that took place in The Accounting Review during 19661985 compared with earlier 20year periods, 19261945 and 19461965. The comparisons are based on examining the articles published in The Accounting Review and written by its leading authors (i.e., those authors who published the most articles). The article considers topics, research methods, financial accounting subtopics, citation analyses (including influential journals, articles, books, and authors), length, author background, and other items. This study shows that The Accounting Review evolved into a journal with demanding acceptance standards whose leading authors were highly educated accounting academics who, to a large degree, brought methods and tools from other disciplines to bear upon accounting issues.

INTRODUCTION

Accounting research changed noticeably in the 1960s. Various factors played a role in this change, including criticisms of business education in the 1959 reports by Gordon and Howell and by Pierson [Dyckman and Zeff, 1984]; the adoption by the American Assembly of Collegiate Schools of Business in 1967 of the doctorate as the terminal degree for accounting faculty [Bricker and Previts, 1990]; change in research and writing standards as required by business faculties in promotion and tenure decisions [Langenderfer, 1987]; and a general belief that scientific methods could help solve social and business problems [Whitley, 1988]. Chatfield [1975] noted that The Accounting Review was in transition during the late 1950s and early 1960s, with nonaccountants making contributions using methods from other disciplines. In addition, Chatfield [1975, p. 6] noted that in the 1960s, there was a trend toward empirical studies.

While it is generally acknowledged that there was a change in the type of research published in The Accounting Review, the extent of that change has yet to be documented. Were previous methods abandoned altogether? Among the newer research methods, which were the most popular? Did interest in financial accounting topics continue to decline? Were new, neverconsidered topics addressed? Which journals/authors/articles/ books influenced the contributors to The Accounting Review? What was the educational/professional background of the contributors, and how did their work differ from that of their predecessors?

These questions are here addressed for the leading authors (i.e., those authors who published the most articles; see Table 1 for the list of names) of The Accounting Review during the 19661985 period.1 The results are directly comparable with The Accounting Review’s leading authors during the 19261945 and 19461965 periods as reported by Fleming et al. [1990, 1991].2 Hence, this study extends our previous work by analyzing the output of the leading authors during 19661985, the next 20year period.

Specifically, this research analyzes, relative to the earlier studies, the following attributes of the articles published by the leading authors: (1) topic, (2) research method, (3) crossclassification of topic and research method, (4) financial accounting

1Heck and Bremser [1986] compiled the list of the leading authors. For their list based on “all articles,” Heck and Bremser counted main articles, notes, and articles appearing in the Education Research (previously, Teacher’s Clinic) and Financial Reporting Sections. Not included were comments, replies, and articles appearing in featured columns (e.g., Accounting Exchange).

2Heck and Bremser [1986] identified the leading authors for each of the 20year periods 19261945, 19461965, and 19661985. There has been a strong interest in the accounting literature in prolific authors as evidenced by Heck and Bremser’s article as well as others [Williams, 1985; Jacobs et al., 1986; Richardson and Williams, 1990].
3The emphasis in this study is on analyzing the next 20year period. One of the major findings, as reflected in the title to the paper, is the shift to quantitative/empirical research methods.

subtopics, and (5) citations of articles and books including journals and authors. Other details are also provided, such as article length, background information on the leading authors, and other changes in The Accounting Review. Through examining and classifying the individual articles of the leading authors, conducting a “single” citation analysis (similar to that of Brown and Gardner [1985a, 1985b] and others), reporting other details, and comparing the 19661985 results to those of the earlier periods, a perspective emerges on the evolving nature of accounting research in the work of the leading authors.

The results of these analyses are presented in the following sections of this paper. The first section discusses the topics and research methods. The subsequent section reports the results of the citation analysis. This is followed by sections on article length, biographical background of the leading authors, and other changes. A synthesis of the nature of the changes in the work of the leading authors of The Accounting Review is contained in the final section.

TOPICS AND RESEARCH METHODS

As in the previous periods examined by Fleming et al. [1990, 1991], classification schemes adapted from Sundem [1987] were used to classify the articles by topic and research method. The topic classification scheme is shown in Exhibit 1, while the research method categorization is displayed in Exhibit 2.4 Each of the three authors of this paper independently classified each article written by the leading authors with respect to topic (as well as to financial accounting subtopic) and research method. The objective was to determine the primary emphasis of topic and research method in the article. The three authors then reached group decisions as to the appropriate classifications. In almost all cases, a unanimous consensus was

4The topic classification scheme is identical to that reported in Fleming et al. [1991, p. 31]. The research method categorization is slightly embellished to be more definitive. Specifically, the description of “economic modeling” was expanded to include explicitly papers dealing with information economics and economic choice theory, which have strong elements of analytical and statistical modeling. In addition, the description of “history” was expanded to include papers tracing the development of a practice or concept which relied on secondary sources, not just those papers employing archival methods. This description better reflects the classification decisions made in this and the earlier periods. Also, book reviews of accounting classics are included in the history category. Similarly, the description of “deductive” was expanded to reflect better the nature of the articles classified as such.

achieved. Occasionally categories were decided by a split 2to1 vote.

EXHIBIT 1 Topic Classification Scheme
FINANCIAL (Fin): External reporting issues even though they may impact internal reporting also. Inventory valuation papers are classified as financial rather than managerial.
MANAGERIAL (Man): Internal reporting issues.
EDUCATION (Edu): Studies on pedagogy and curriculum matters.
RESEARCH METHODS (Res): Focused completely on such methods without direct application to an accounting issue.
AUDITING (Aud): Related to tasks performed by auditors.
PROFESSIONAL (Pro): Professional practice of accounting firms.
TAX: Federal income tax issues.
INFORMATION SYSTEMS (Inf): Broad range of papers from office automation, to evaluation methods for accounting software, to the effects of different data storage systems on decision making, etc.
NONPROFIT/GOVERNMENTAL (Non): Requiring the special circumstances of such organizations to be a major influence on the research.
INTERNATIONAL (Int): Assessed uniquely international aspects of an issue such as differences in accounting practices, generally involving more than one country.
OTHER (Oth): Not related to one of the above.
Source: adapted from Sundem [1987, pp. 194195]

EXHIBIT 2 Research Method Classification Scheme
DEDUCTIVE (Ded): The deductive studies that do not fit in other categories, including opinion pieces. These nonempirical studies are primarily verbal/ descriptivetype articles where a logical conclusion follows from a set of assumptions or premises (other than modeling studies). In addition, this category was interpreted to include inductive and legal research methods as well.
ANALYTICAL MODELING (Ana): Studies using models with no specific underlying economic theory but using mathematical techniques.
GENERAL EMPIRICAL (Gen): A catchall that includes primarily descriptive empirical work.
ECONOMIC MODELING (Eco): Studies which bring economic analysis to bear on a topic; they may be mathematical or verbal models. Papers based on information economics or economic choice theory are included here.
STATISTICAL MODELING (Sta): Studies which use models where the main focus is on statistical models.
CAPITAL MARKET (Cap): Studies using security prices to measure reaction or association.
BEHAVIORAL (Beh): Studies conducted to measure the reaction of students or professional subjects.
HISTORICAL (His): Papers that use archival methods to study an issue of current interest. Also included are papers that trace the development of a practice or concept using secondary sources and book reviews of accounting classics.

EXHIBIT 2 (continued)
SIMULATION (Sim): Studies where the analysis is so complex that computer
simulation is necessary. SURVEY (Sur): Studies reporting information gathered by questionnaire about
practices or attitudes.
Source: adapted from Sundem [1987, p. 198]

Topics: Table 1 shows that the leading authors published 154 articles in ten topical areas.5 The international area was the only one in which they did not publish a paper. Financial accounting was the most popular topic (48% of the articles), followed by managerial accounting (25%), education (8%), research methods (6%), and auditing (5%). The leading authors published only a few papers in the professional, tax, information systems, and nonprofit/governmental areas.

The leading authors of the 19661985 period published essentially the same percentage of their articles in the financial accounting area as did the leading authors of the 19461965 period. Although this percentage is down from the 19261945 period, it nevertheless shows the continued strong interest in financial accounting among the leading authors. Interestingly, all but Manes of the 19 leading authors published a financial accounting article in The Accounting Review during the 19661985 period. Similarly, only two (William Campfield and Robert Van Voorhis) of the 22 leading authors during 19461965 and only one (Lloyd Morey) of the 19 leading authors during 19261945 did not publish a financial accounting article in The Accounting Review. Hence, for 60 years there has been widespread interest in financial accounting among the individual leading authors.

5As noted, the leading authors and their respective number of publications were originally identified by Heck and Bremser [1986]. With only two exceptions, using the criteria identified by Heck and Bremser [1986, pp. 735736], articles equal in number to that reported by them were located. The two exceptions were for Ijiri and Swieringa where one less than the number of articles reported by Heck and Bremser could be found. Also, it should be noted that five of the articles were coauthored by the leading authors listed in Table 1. Hence, there were only 149 actual articles examined. The coauthored articles were classified as financial (three articles), managerial (one article), and auditing (one article). Adjusting the overall classification numbers for this double counting does not appreciably affect the results.

Managerial accounting was the second most popular topic during the 19661985 period. Its 25% share of the articles was nearly double that of the 13% of the 19461965 period which, in turn, was more than double the 6% of the 19261945 period. It was the only topic whose proportion of articles increased in both 19661985 and 19461965.

The percentage of education articles during 19661985 declined to 8% from 21% during 19461965. This decline was likely due, at least in part, to the changing role of education articles in The Accounting Review. In 1971, the “Education Research and Academic Notes” section of The Accounting Review replaced the “Teacher’s Clinic” where Flesher [1991, p. 153] reported articles on education research, teaching methods, and “think” pieces that had been published. In 1975, the name of the section was changed to “Education Research,” with the academic notes portion [Tracy, 1971, p. 156], consisting of classroom innovations, practical pointers, observations, materials’ development, clever examples, or other experiences, dropped. The “Education Research” section was restricted to “… research findings regarding theory and technique applied in accounting courses …” and “… developments . . . that are of primary and fundamental importance” [DeCoster, 1975, p. 160]. Hence, the section became more restrictive in the type of articles which could be published. In addition, Flesher [1991, p. 154] noted that Issues in Accounting Education started publication in 1983 on an experimental basis, and that the “Education Research” section of The Accounting Review was discontinued in 1986 when Issues began publication on a regular, semiannual basis.

Although still relatively small in number, articles on research methods and auditing increased among the leading authors during the 19661985 period. The interest in research methods reflected the use of quantitative/empirical research methods of the era. The increase in auditing articles was apparently part of a general trend among all authors in The Accounting Review. Sundem [1987], in his analysis of all published articles in The Accounting Review during his term as editor (19821986), reported that auditing papers represented 16 percent of the articles.

Few articles were published by the leading authors in the other areas during the 19661985 period or, in most cases, in these areas during the prior periods. Moreover, the exceptions were largely due to the efforts of specific individual authors. For example, 5% of the articles during the 19461965 period were in the international area. Of the ten international articles published, nine were authored by Mary Murphy. Similarly, 8% of the articles during 19461965 were in the professional area, but nine of the 15 articles in this area were authored by Campfield. In addition, while 8% of the 19261945 articles were in the nonprofit/governmental area, six of the 13 were authored by Morey. Thus, there was not a widespread interest among the leading authors in professional, tax, information systems, nonprofit/governmental, or international areas during the first 60 years of The Accounting Review.
6Sundem’s results, based on all authors, were not directly comparable to those of this study which are based only on the work of leading authors. In addition, his results were just for 19821986 rather than the 20year period examined in this study. For those readers with an interest, the other major topical areas while Sundem [1987, p. 202] was editor were financial accounting (45% of articles published), managerial accounting (17%), tax (6%), and professional (5%). The remaining areas constituted 2% or less of the articles.

Research Methods: Table 2 illustrates a dramatic change in the research methods used by the leading authors. The use of the deductive method fell from over 80% during the 40year period 19261965 to only 29% during 19661985.7 Moreover, each of the other methods was used more during 19661985 than 19461965.8
TABLE 2 Research Methodology by Author

7The classification “deductive” is a misnomer to some extent. Modeling studies, for example, could also be considered “deductive.” Opinion pieces are not “deductive.” However, rather than developing an alternative classification scheme, the one developed by Sundem [1987] was adopted since he developed it specifically for The Accounting Review and since developing a definitive classification scheme would be impossible and inevitably arbitrary to some degree. Hence, the articles written by the leading authors were classified into that category which was considered best from those specified by Sundem (excluding “other” which was not defined by Sundem). Consequently, the “deductive” category came to represent other studies which were primarily verbal/ descriptive, including legal research methods and even a few inductivetype studies. Perhaps a better classification title is “other deductive/descriptive.”
8 Again, the double counting of five articles because of coauthorship among the leading authors (see footnote 5) did not appreciably affect the results. Two of these articles were classified as economic modeling, and the others as analytical modeling, statistical modeling, and capital markets respectively.

Perhaps the most surprising result is that the deductive approach was used to the extent found despite the changing research method environment described previously. The change away from the deductive method to quantitative/empirical methods did not occur all at once. This gradual change is made even more evident by examining the 19661975 and 19761985 subperiods. Of the 44 articles which primarily used the deductive method, 32 of them (73%) were published during 19661975, while only 12 of them (27%) appeared in the later decade. Clearly, 19661985 was a transitional period in research methods.

Among the methods other than the deductive method, analytical modeling was the most popular as shown in Table 2. This was followed by general empirical, the other two modeling methods (economic and statistical), and by the empirical areas of capital market and behavioral. Hence, modeling methods tended to lead the change away from the deductive method.10
Although they were generally employed more in the 19661985 period than in the earlier periods (except for the historical method), the research methods of historical, simulation, and survey were used relatively infrequently by the leading authors. It is interesting to note that the historical method was used relatively more commonly during the 19261945 period than in subsequent periods. This was largely due to the work of A.C. Littleton and Stanley Howard. Of the 19 articles classified as using the historical method during 19261945, 11 of them were authored by Littleton with an additional five by Howard. Hence, as in the case of topics, the efforts of specific authors accounted for the passing popularity of certain methods or topics during the earlier periods.

At the individual author level, all but two (Demski and Petri) of the leading authors during 19661985 used the deduc

9 Although not directly comparable to the results of this study, it is interesting to note that during Sundems’s term as editor (19821986), only 4% of the published articles in The Accounting Review used the deductive method primarily [Sundem, 1987, p. 202]. Sundem [1987, p. 202] also reported the following frequencies for the other research methods in published articles: general empirical (31%), capital market (14%), behavioral (22%), analytical modeling (6%), economic modeling (7%), statistical modeling (4%), simulation (4%), historical (2%), survey (1%), and other (4%).

10Sundem [1987, p.196] combined some methods into groups which he referred to as modeling (including analytical, economic, statistical, and simulation) and empirical (including general empirical, behavioral, capital market, and survey). Combining methods into groups in a like manner for this study resulted in modeling with 40% of the articles and empirical with 29%.

tive method. However, only three (Revsine, Weygandt, and Swieringa) employed the deductive method a majority of the time. The leading authors tended to utilize a variety of methods. Petri used the fewest (two), Revsine the most (seven).

Topics by Research Methods: Table 3 shows a crossclassification of topics by research methods. During the previous 20year periods, the deductive method was clearly the dominant method, not only overall but for each topic except for the “other” category during 19261945, when it was the primary method in only 29% of the articles. Its next two lowest percentage rates of use were 60% for the professional area during 19261945 and 76% for the managerial area during 19461965. In the financial area, it was the primary method in 86% of the articles for 19261945 and 87% for 19461965.

By contrast, the deductive method during 19661985 was the primary method a majority of the time for only the education, professional, and information systems topics.11 Although it was the most popular method in the financial area, the deductive method’s use declined to 32% of the time, with 15 of the 24 financialdeductive articles published during the first half
11These results are not affected by the five papers which were coauthored by the leading authors (see footnotes 5 and 8). The five papers were crossclassified as financialstatistical modeling, financialeconomic modeling, financialcapital markets, managerialanalytical modeling, and auditingeconomic modeling.

Overall, Table 3 shows that a variety of research methods was used in each area. The only methods that were used a majority of the time in a given area (other than the deductive method) were analytical modeling in the tax area and general empirical in the nonprofit/governmental area. However, the modeling methods of analytical, economic, statistical, and simulation tended to dominate managerial articles, accounting for 31 of the 39 articles (79 percent).

Financial Accounting Subtopics: Table 4 shows a classification by subtopic and author of the articles in financial accounting, the most popular area. The articles were classified using the chapter titles taken from an intermediate and an advanced accounting textbook.12 Each of the 74 financial accounting articles could be classified in this manner. The seven articles classified as “other” each related to a single subtopic (i.e., pensions, inventories, leases, current and contingent liabilities, statement of changes, foreign currency, and intangible assets).

As shown in Table 4, the four most popular subtopics were changing prices, environment and concepts, plant and equipment, and income taxes.13 Except for income taxes, there has been considerable interest in each of these topics throughout the first 60 years of The Accounting Review. Changing prices was the fourth most popular topic among the leading authors during 19261945, and it was tied as the fifth most popular during 19461965. Environment and concepts was the most popular topic during each of the periods 19261945 and 1946
12For definitiveness, as done for the earlier periods, the chapter titles from Kieso and Weygandt’s Intermediate Accounting [1989] and Baker et al.’s Advanced Accounting [1989] were used as the basis for the classification scheme. Some of the related chapters were combined into a single subtopic (e.g., the two inventory chapters were considered one subtopic, the chapter on depreciation was combined with plant and equipment; etc.). Also, the chapter titles were shortened in some cases (e.g., the first two chapters, entitled “The Environment of Financial Accounting and the Development of Accounting Standards” and “Conceptual Framework Underlying Financial Accounting,” were combined into a single subtopic called “environment and concepts”).

13Because of coauthorship among the leading authors, there were actually only six articles in the plant and equipment subtopic; hence, plant and equipment and income taxes were tied as the third most popular area. The other two subtopics affected by the double counting of articles due to coauthorship were accounting changes (actually three articles) and income statement (actually three articles).

That changing prices was the most popular during 19661985 was not surprising given the high level of inflation during the 1970s and early 1980s, as well as the experiments by the Securities and Exchange Commission (SEC) and the Financial Accounting Standards Board (FASB) with accounting for changing prices. Similarly, the continued interest among the leading authors in standard setting and basic concepts reflected the changes in the standardsetting process at the time, such as the formation of the FASB and the development of its conceptual framework. Less obviously, plant and equipment, especially depreciation, continued to be an area of strong interest. This interest was indicative of an ongoing concern with allocations [Thomas, 1969, 1974].

The income tax subtopic area of financial reporting became a new area of concern for the leading authors during 19661985. There were only three tax articles written by the leading authors during 19461965 and none during 19261945. This interest reflected issues such as the controversial accounting for deferred taxes at the time prescribed by Accounting Principles Board Opinion No. 11 in 1967 [Johnson, 1996, p. 191].

After adjusting for double counting (see footnote 13), it is interesting to note that these four areas accounted for 54 percent (38 of 71) of the financial accounting subtopics. Hence, there was some concentration of interest among the leading authors. The leading authors of the earlier periods also tended to concentrate on specific topics; the comparable percentage for 19461965 was 45 percent (39 of 86) and 61 percent (70 of 115) for 19261945. Hence, while a variety of subtopics was addressed, the leading authors focused on contemporary financial accounting issues. For example, ten of the 15 articles classified as changing prices made direct reference to one or more of the professional pronouncements regarding changing prices issued by the Accounting Principles Board, the FASB, or the SEC.14 As another example, Demski [1973, 1974] referred to the formation of the FASB in his articles dealing with the general problem of standard setting.

A CITATION ANALYSIS

As done for the 19261945 and 19461965 periods, this study employs a “single” citation analysis to help determine the impact on the leading authors of a wide spectrum of accounting literature. Similar analyses have been performed in other contexts to determine the influence of journals [Dyckman and Zeff, 1984; Brown and Gardner, 1985a], to identify significant works [Brown and Gardner, 1985a; Gamble and O’Doherty, 1985], and for other uses (e.g., investigating the role of historical articles in research [Bricker, 1988a, 1988b]).
This study employs citation analysis as follows. To be

14It is interesting to note that of the 15 articles classified as changing prices, eight of them used the deductive method. Hence, the deductive method still played a significant role in this area. However, of the financial accounting subtopics containing at least two papers, the only other area where the deductive method was used in a majority of the papers was in the longterm liabilities area (in two of the three papers). In terms of absolute numbers, the next highest use of the deductive method after the changing prices area was in the environment and concepts area where it was used in three papers (out of 11).

counted as a journal article, the author, title, and journal minimally had to be identified in a leading author’s article. For books to be counted, at a minimum, the author and the title had to be given. Papers in proceedings were counted as journal articles with the proceedings counted as a journal. Chapters in books and unpublished Ph.D. dissertations were counted as books. A work cited more than once in the same leading author’s article counted as one in the tabulations (it did not matter if a work was cited more than once in the same article). Selfcitations, including coauthored works of the leading author, were eliminated in determining which journals, articles, books, or authors had the greatest influence on the leading authors. In counts for influential authors, full credit was given to any coauthors of cited works.

As an overview, Table 5 shows the number of citations of articles and books made by each leading author individually and collectively as a group.15 The overall averages showed that there was a dramatic change in the use of references in the period 19661985 compared with the previous periods. The average number of citations for both books and articles (i.e., total citations divided by total articles) of 13.2 far exceeded the averages of 3.4 for 19461965 and of 3.2 for 19261945. Similarly, the average for article citations grew to 8.5 in 19661985, compared with only 1.4 in 19461965 and 1.2 in 19261945. The average for book citations more than doubled to 4.7 in the 19661985 period, compared with an average of 2.0 in each of the earlier periods.

Dramatic differences were apparent at the individual author level as well (see Table 5). During 19661985, all but one leading author averaged at least four citations per article and 14 of the 19 averaged at least ten.16 In contrast, only seven of the 22 leading authors during 19461965 averaged at least four citations, while no leading author averaged at least ten. During 19261945, there were only three leading authors out of 19 who averaged at least four citations per article and only two who averaged at least ten.

The practice of utilizing several references in articles became widespread by the leading authors in 19661985. As noted by Fleming et al. [1990, 1991], the leading authors of 19261945 had a practical orientation which decreased somewhat during the 19461965 period. This change in orientation toward an academic one no doubt accounted for some of the change in the use of references. However, given that 20 of the 22 leading authors during 19461965 had earned a Ph.D., it is difficult to attribute the dramatic increase in references during 19661985 to the fact that all of the leading authors had a Ph.D. (discussed subsequently). Rather, at least part of the change is due more plausibly to the expanded use of research tools and articles and books from other disciplines, as is documented below.

16The one exception to an average of at least four citations per article is Bierman. Interestingly, he was the only leading author during 19661985 who was also a leading author during 19461965. Bierman published articles in The Accounting Review during the second half of the 19461965 period (19561965) and during most of the first half of the 19661985 period (19661974). While there was somewhat of a shift in his research methods from primarily the deductive method (used in eight of his ten articles) during the earlier period to other methods during the later period (other methods used in five of his ten articles), his style of writing did not change appreciably over the two periods. He tended to analyze problems using straightforward frameworks such as present value analysis and basic valuation models. Hence, his rate of 2.6 citations per article in the later period was not much higher than that of the 19461965 period when he averaged 2.1 citations per article.

Another interesting feature of Table 6 is that only 11 of the 54 journals (21 percent) are accounting journals. Similarly, only five of the mostcited 15 journals (33 percent) have an accounting emphasis. The remaining ten journals come from the fields of general business, finance, economics, management, psychology, and statistics. In contrast, nine of the 13 mostcited journals (69 percent) during 19461965 had an accounting orientation. Hence, 19661985 can be characterized as a period in which the leading authors were significantly influenced by other disciplines.

The results portrayed in Table 6 also differ from those of the 19261945 period. While only three of the mostcited 15 journals (20 percent) during that early period were accounting journals (not many accounting journals existed at that time), five of the most frequently cited were legal journals, such as the Columbia Law Review and the Harvard Law Review. The legal journal citations essentially disappeared during the 19461965 period. As previously noted, business and other social science disciplines became influential during 19661985.
With respect to individual journals, The Accounting Review was the mostcited journal, finishing well ahead of the Journal of Accounting Research. However, part of this difference is attributable to the fact that the Journal of Accounting Research

17Citations to accounting journals accounted for 59.7% of the total number of citations for journals listed by name in Table 6. During 19461965, accounting journals accounted for 87.8% of the citations for journals cited three or more times. Hence, based on number of citations of journals, the increase in influence of other disciplines on the leading authors during 19661985 compared with 19461965 is also evident.

did not start publication until 1963 and, therefore, had fewer articles which could have been referenced compared with The Accounting Review. Similarly, the Journal of Accounting & Economics ranked only 15th on this list, having started publication in 1979.

Somewhat surprisingly, the Journal of Accountancy tied as the third mostcited journal, along with the Journal of Business. However, its diminishing influence among the leading authors was evident over the first 60 years of The Accounting Review. During 19261945, the Journal of Accountancy was cited 35 times compared with 32 for The Accounting Review, a ratio of 35/32 = 1.09 (partly biased toward the Journal of Accountancy since The Accounting Review started in 1926). During 19461965, the comparable ratio was 38/107 (.36), while the ratio for 19661985 was 59/336 (.17). This downward trend is consistent with the changing orientation of the leading authors from practical to academic. Moreover, the only other journals with a practical orientation in the top 15 journals listed in Table 6 are Management Accounting and, to some extent, the Harvard Business Review. Hence, the leading authors of 19661985 were influenced to a large degree by academic journals from other disciplines.

Authors: Table 7 shows the mostcited authors of journal articles. Beaver, a leading author himself, is the mostcited author. His citation record clearly is an outlier and exceptional. The leading authors cited 14 of his articles, outdistancing all other authors. His mostcited article, “Predictive Ability as a Criterion for the Evaluation of Accounting Data” [Beaver et al., 1968], was cited six times.

The second mostcited author was Ball, tied with Dopuch. The leading authors cited seven of his articles, with the mostcited article, “An Empirical Evaluation of Accounting Income Numbers” [Ball and Brown, 1968], cited nine times. No other article was cited nine or more times by the leading authors.

Dopuch, longtime editor of the Journal of Accounting Research, was tied with Ball as the second mostcited author. The leading authors cited 11 of his works, the second most number of works cited after Beaver. Two of Dopuch’s articles were cited four times each by the leading authors.

Table 7 also shows that five of the 19 listed authors are included in this study as leading authors, two (including Bierman) were leading authors from 19461965, two are finance professors (Fama and Roll), and one a psychology professor (Slovic). Three are members of the Accounting Hall of Fame. Most remarkably, however, is that of the 19 individuals listed, 15 of them (except for Ijiri, Slovic, Dyckman, and Feltham) either earned their doctorates or were on the faculty at the University of Chicago.

Table 8 lists the nine articles cited five or more times. Feltham’s “The Value of Information,” published in The Accounting Review, was cited eight times (second to Ball and Brown), while Fama’s “Efficient Capital Markets: A Review of Empirical Work,” published in the Journal of Finance, was cited seven times. Hence, unlike the earlier periods when three was
18This shows that the influence of particular universities on the accounting literature may be even greater than suggested by Lee [1997] in his examination of the impact of 20 elite universities on the editorial boards of six journals.

Overall, five of the top nine cited articles were published in The Accounting Review, three in the Journal of Accounting Research, and one in the Journal of Finance. In addition, based on a review of these articles, five were related to financial accounting issues, three to managerial accounting issues, and one (Beaver et al.) to research methods. All of the articles were published between July 1964 and May 1970, early enough to be cited frequently by the leading authors of 19661985. Two of the articles (those by Ball and Brown and Sorter) also appeared on a list of mostcited articles based on an analysis of The Accounting Review between 19761982 [Brown and Gardner, 1985a, p. 101].

Table 9 shows the mostcited authors of books. Ijiri heads the list. Six of his books were cited by the leading authors with his The Foundations of Accounting Measurement cited six times. Bell and Edwards tied for the second most citations. All 14 of their citations were for their coauthored classic, The Theory and Measurement of Business Income. This book was cited most frequently by the leading authors. Paton had eight different books cited, more than any other author.

20As will be seen, there is a stronger relationship between book citations and Accounting Hall of Fame membership than between combined book and article citations (Table 11) or article citations (Table 7) and Accounting Hall of Fame membership. During the previous periods, the combined book and article citations had the strongest relationship with the Accounting Hall of Fame. As noted by Fleming et al. [1991], this implies that books (and not just articles as is sometimes done) should also be taken into account in citation studies which try to determine influential authors, schools, doctoral programs, etc. Of course, other criteria besides contributions to the accounting literature are involved in selecting inductees to the Accounting Hall of Fame. These include professional and public service [Burns, 1975].

other statistically oriented works are included in the list (Econometric Methods by Johnston and The Foundations of Statistics by Savage). Otherwise, seven of the 12 books relate to financial accounting, of which two were written by economists (The Theory and Measurement of Business Income by Edwards and Bell and Value and Capital by Hicks). Two are cost accounting textbooks (Horngren and Shillinglaw).

aa leading author during 19661985 ba leading author during 19461965 ca leading author during 19261945 *member of the Accounting Hall of Fame

Table 11 shows the mostcited authors of both articles and books combined. Not surprisingly, since article citations were considerably more frequent than book citations during the 19661985 period, this list is dominated by authors of articles. In fact, all but one (Drake) of the individuals named in Table 7 as the mostcited authors of articles are also included in Table 11. By contrast, only five individuals named in Table 9 as the mostcited authors of books are also included in Table 11. Five of the authors listed in Table 11 were leading authors during 19661985, two during 19461985, and one for 19261945. Five of the authors have been inducted into the Accounting Hall of Fame.

Remarkably, Moonitz, Paton, and Sorter also appeared on the comparable list for 19461965. In addition, Paton appeared on the list for 19261945. Paton was one of the most influential individuals on the leading authors of The Accounting Review for 60 years.

ARTICLE LENGTH

The average length of the articles written by the leading authors during 19661985 was 10.9 pages, compared to an average of 6.5 pages during 19461965 and 8.6 pages during 19261945. At the individual author level during 19661985, seven of the leading authors averaged less than ten pages, while the other 12 averaged more than ten. Kaplan had the highest average with 15.8 pages; Copeland had the lowest with 6.4 pages.

By contrast, between 19461965, all of the leading authors averaged less than ten pages, while 11 of the 19 leading authors during 19261945 averaged less than ten pages. Hence, the work of the leading authors during 19661985 tended to be longer than that of their predecessors. This is consistent with using more citations, including a literature review in some cases, and employing more quantitative/empirical methods which usually require explanation.

SOME BIOGRAPHICAL INFORMATION

Table 12 reveals the doctoralgranting institution and primary affiliation for each leading author. The doctoralgranting institution was obtained from Hasselback’s Accounting Faculty Directory [1988] which contained the necessary information for each individual. The primary affiliations, in chronological order, are those reported with the articles published in The Accounting Review. As mentioned, all of the leading authors earned a Ph.D., generally from a BigTen or private university. Except for Livingstone’s time with Coopers and Lybrand, they

22Interestingly, during 19661985 there was a weaker relationship compared with the earlier periods between being cited (books and articles) by the leading authors and being inducted into the Accounting Hall of Fame. As one of the referees suggests, the cited authors from the earlier periods may have made more contributions in areas other than the accounting literature.

As noted by Fleming et al. [1990, 1991], eight of the 19 leading authors of 19261945 had not earned a Ph.D. Five of

23The Big Ten is an athletic conference comprised generally of large universities in the midwestern U. S. The Big Ten includes: Ohio State University, Michigan State University, University of Michigan, University of Wisconsin, University of Minnesota, University of Iowa, University of Illinois, Northwestern University, Indiana University, Purdue University, and (a recent addition of an 11th school) Pennsylvania State University. Interestingly, the University of Chicago was a member of the Big Ten when it had a football team. Also, all but two leading authors (Manes and Petri) earned their Ph.D.s at one of the elite universities identified by Lee [1997].

them had nonacademic positions (and four others held positions outside of academia) when published in The Accounting Review. During 19461965, all but two had earned Ph.D.’s and four of the 22 had nonacademic positions. By the period 19661985, the leading authors of The Accounting Review had evolved into an almost exclusively Ph.D./academiconly group. Perhaps, only such individuals have the time, training, and/or inclination to be familiar with the literature (as reflected in the growth of citations) and the frequently employed quantitative/empirical research methods (as documented earlier).

OTHER CHANGES

The percentage of editors who were also leading authors decreased in the succeeding 20year periods. During 19261945, all three editors (Paton, Kohler, and Littleton) were also leading authors. During 19461965, only Littleton and Mautz of the six editors were also leading authors. By the 19661985 period, only DeCoster of the seven editors was a leading author.

The occurrence of coauthorship also changed over time. Of the 172 articles written by the leading authors during 19261945, only four articles (2.3 percent) were coauthored. In the 19461965 period, 16 of the 186 articles (7.5 percent) written by the leading authors were coauthored. However, a dramatic change occurred during 19661985 as a majority of the articles by the leading authors were coauthored (80 of the 149 articles, 53.7 percent).

More generally, and not just in the work of the leading authors, additional changes occurred in The Accounting Review, such as the practice in 1975 of listing references at the back of articles rather than in footnotes, perhaps reflecting their increased importance and number. In 1977, abstracts began to be included with the published articles, a change Flesher [1991, p. 169] noted to be a consequence of a study by Abdelkhalik [1976]. Abdelkhalik found that practitioners responding to a survey, apparently concerned about the readability of The Accounting Review, desired to have abstracts published with the articles.

25Another intriguing suggestion by Abdelkhalik [1976, p. 616] to increase the readability of The Accounting Review was not implemented. This suggestion was to include the description of the research technique in an appendix rather than in the body of the article.

Another notable change began in 1966 when an editorial board was added to the article review process. This change was made during Trumbull’s term as editor; Trumbull had been making the editorial decisions himself [Flesher, 1991, p. 167].26 Interestingly, the acceptance rate declined from about 40 percent in 1967, to 24 percent in 1969, to 20 percent in 1972, and to 13 percent from 19751977 [Flesher, 1991, pp. 167168]. During this period, the readability of The Accounting Review came under question although a study by Caplan and Griffin did not find “… any widespread dissatisfaction with the publication” [quoted by Flesher, 1991, p. 168]. Nevertheless, the very undertaking of this study was indicative of the changes occurring in The Accounting Review.

THE CHANGING NATURE OF THE WORK OF THE LEADING AUTHORS OF THE ACCOUNTING REVIEW: A SYNTHESIS

This study examines characteristics of the work of the leading authors of The Accounting Review during 19661985. The results are directly comparable to those for the 19261945 and 19461965 periods reported by Fleming et al. [1990, 1991]. In short, the work of the leading authors of The Accounting Review has changed dramatically over the years. While financial accounting topics continued to be popular among the leading authors to about the same extent as in 19461965, there was growing interest in managerial and, to a lesser extent, auditing issues. Articles concerning research methods and, though small in number, information systems appeared for the first time during 19661985. Education articles, which hit their pinnacle during 19461965, experienced a decline in interest among the leading authors and were phased out of The Accounting Review in the 1980s.

The big change was in research methods. Modeling and empirical methods became prominent during 19661985, with analytical modeling and general empirical methods leading the way. Although used to a surprising extent, deductivetype methods declined in popularity, especially in the second half of the 19661985 period. Among the more popular topics, only in the education area, which was in decline among the leading au

26Actually, this was not the first time that The Accounting Review had an editorial board; Littleton was technically a chair of a threeperson editorial board during his term as editor (19441947).

thors, were deductive methods used a majority of the time. To a large extent, the world of accounting research for the leading authors had changed. One major exception to the changing world, however, was the continued interest among the leading authors in the financial accounting issues of their time.

Other changes accompanied the innovation in research methodology. The leading authors referenced about four times as many sources than in the earlier periods. While The Accounting Review, the Journal of Accounting Research, and the Journal of Accountancy were the mostcited journals, a host of journals from other disciplines including business, finance, economics, management, and psychology were also referenced. The use of journal articles greatly expanded, with the most influential authors cited by the leading authors being former doctoral students or faculty at the University of Chicago, including Beaver, the mostcited article author, and Ball and Brown, authors of the mostcited article, “An Empirical Evaluation of Accounting Income Numbers” [1968].

Books were also cited more frequently during 19661985 than in the previous periods. Like journals, the books often came from cognate disciplines, including economics and statistics. Edwards and Bell’s classic, The Theory and Measurement of Business Income [1961], was the mostcited book, while Ijiri was the mostcited book author.

As a group, the leading authors had evolved into an almost exclusively Ph.D./academiconly set of contributors. All of the leading authors had earned a Ph.D., were accounting professors, and were affiliated with a university while publishing in The Accounting Review, excepting Livingstone’s time with Coopers and Lybrand.27 During 19461965, four of the leading authors had nonacademic positions, while during 19261945, five of the leading authors had nonacademic positions (four others held both academic and nonacademic positions). The leading authors of 19661985 were academics, not practitioners.

The leading authors tended to write longer articles than had their predecessors. Other changes included the higher incidence of coauthorship and the institution of an editorial process involving a review board. At the same time, the acceptance rate for publication in The Accounting Review plummeted from

27This is contrary to Chatfield’s [1975, p. 6] claim, at least with respect to the leading authors, that more articles by “nonaccountants” were published in The Accounting Review using “… ideas or methods from their own discipline.”

40 percent to 13 percent in a tenyear period. Questions began to be raised regarding the readability of the journal. To summarize, The Accounting Review during 19661985 had become a journal with demanding acceptance standards whose leading authors were highly educated, accounting academics who, to a large degree, brought methods and tools from other disciplines to bear upon accounting issues.

REFERENCES

Abdelkhalik, A. (1976), “An Analysis of the Attitudes of a Sample of the AAA Members Toward The Accounting Review,” Accounting Review, Vol. 51, No. 3: 604616.
Baker, R., Lembke, V., and King, T. (1989), Advanced Financial Accounting (New York: McGrawHill).
Ball, R. and Brown, P. (1968), “An Empirical Evaluation of Accounting Income Numbers,” Journal of Accounting Research, Vol. 6, No. 2: 159178.
Beaver, W., Kennelly, J., and Voss, W. (1968), “Predictive Ability as a Criterion for the Evaluation of Accounting Data,” Accounting Review, Vol. 43, No. 4: 675683.
Bricker, R. (1988a), “Knowledge Preservation in Accounting: A Citational Study,” Abacus, Vol. 24, No. 2: 120131.
Bricker, R. (1988b), “Influences from Early Accounting Literature on Contemporary Research,” Accounting Historians Journal, Vol. 15, No. 2: 8398.
Bricker, R. and Previts, G. (1990), “The Sociology of Accountancy: A Study of Academic and Practice Community Schisms,” Accounting Horizons, Vol. 4, No. 1: 114.
Brown, L. and Gardner, J. (1985a), “Using Citation Analysis to Assess the Impact of Journals and Articles on Contemporary Accounting Research (CAR),” Journal of Accounting Research, Vol. 23, No. 1: 84109.
Brown, L. and Gardner, J. (1985b), “Applying Citation Analysis to Evaluate the Research Contributions of Accounting and Faculty Doctoral Programs,” Accounting Review, Vol. 60, No. 2: 262277.
Burns, T. (1975), “The Accounting Hall of Fame,” Accounting Historians Journal (originally published as The Accounting Historian), Vol. 2: 3739.
Chatfield, M. (1975), “The Accounting Review’s First Fifty Years,” Accounting Review, Vol. 50, No. 1: 16.
DeCoster, D. (1975), “Editor’s Note,” Accounting Review, Vol. 50, No. 1: 160.
Demski, J. (1973), “The General Impossibility of Normative Accounting Standards,” Accounting Review, Vol. 48, No. 4: 718723.
Demski, J. (1974), “Choice Among Financial Reporting Alternatives,” Accounting Review, Vol. 49, No. 2: 221232.
Dyckman, T. and Zeff, S. (1984), “Two Decades of the Journal of Accounting Research,” Journal of Accounting Research, Vol. 22, No. 1: 225297.
Edwards, E. and Bell, P. (1961), The Theory and Measurement of Business Income (Berkeley and Los Angeles, CA: University of California Press).
Fleming, R., Graci, S., and Thompson, J. (1990), “Characteristics of the Work of the Leading Authors of The Accounting Review, 19261945,” Accounting Historians Journal, Vol. 17, No. 1: 525.
Fleming, R., Graci, S., and Thompson, J. (1991), “Tracing the Evolution of Research in The Accounting Review through Its Leading Authors: The 19461965 Period,” Accounting Historians Journal, Vol. 18, No. 2: 2653.
Flesher, D. (1991), The ThirdQuarter Century of the American Accounting Association 19661991 (Sarasota, FL: American Accounting Association).
Gamble, G. and O’Doherty, B. (1985), “Citation Indexing and Its Use in Accounting: An Awareness Survey and Departmental Ranking,” Issues in Accounting Education, Vol. 3: 2840.
Hasselback, J. (1988), Accounting Faculty Directory (Englewood Cliffs, NJ: PrenticeHall).
Heck, J. and Bremser, W. (1986), “Six Decades of The Accounting Review: A Summary of Author and Institutional Contributors,” Accounting Review, Vol. 61, No. 4: 735744.
Jacobs, F., Hartgraves, A., and Beard, L. (1986), “Publication Productivity of Doctoral Alumni: A TimeAdjusted Model,” Accounting Review, Vol. 61, No. 1: 179187.
Johnson, R. (1996), “Deferred Income Tax Accounting,” in Chatfield, M. and Vangermeersch, R. (eds.), The History of Accounting (New York: Garland Publishing, Inc.): 190191.
Kieso, D. and Weygandt, J. (1989), Intermediate Accounting, 6th ed. (New York: John Wiley & Sons, Inc).
Langenderfer, H. (1987), “Accounting Education’s History—A 100Year Search for Identity,” Journal of Accountancy, Vol. 163, No. 5: 302331.
Lee, T. (1997), “The Editorial Gatekeepers of the Accounting Academy,” Accounting, Auditing and Accountability Journal, Vol. 10, No. 1: 1130.
Richardson, J. and Williams, J. (1990), “Canadian Academic Accountants’ Productivity: A Survey of 10 Refereed Publications, 19761989,” Contemporary Accounting Research, Vol. 6, No. 1: 278294.
Sundem, G. (1987), “Overview of Four Years of Submissions to The Accounting Review,” Accounting Review, Vol. 62, No. 1: 191202.
Thomas, A. (1969), The Allocation Problem in Financial Accounting Theory, Studies in Accounting Research #3 (Sarasota, FL: American Accounting Association).
Thomas, A. (1974), The Allocation Problem: Part Two, Studies in Accounting Research #9 (Sarasota, FL: American Accounting Association).
Tracy, J. (1971), “Editor’s Note,” Accounting Review, Vol. 46, No. 1: 156.
Whitley, R. (1988), “The Possibility and Utility of Positive Accounting Theory,” Accounting, Organizations and Society, Vol. 13, No. 6: 631645.
Williams, P. (1985), “A Descriptive Analysis of Authorship in The Accounting Review,” Accounting Review, Vol. 60, No. 2: 300313.