Miklos A. Vasarhelyi Da Hsien Bao
and
Joel Berk GRADUATE SCHOOL OF BUSINESS,
COLUMBIA UNIVERSITY
TRENDS IN THE EVOLUTION OF SCHOLARLY ACCOUNTING THOUGHT: A QUANTITATIVE EXAMINATION
The authors are grateful for the assistance of Georgiana Hsu of Columbia University. The comments and suggestions of Professors L. Brown, R. Colson, J. Gardner,N. Preis, G. Previts, and D. Yang, Messrs. W. J. Shugard and H. S. Tsao of the AT&T Bell Laboratories as well as two anonymous referees are greatly appreciated.
Abstract: Contemporary Accounting Research (CAR) has expanded substantially in scope over the past two decades. This paper provides an overview of these trends using both quantitative techniques from statistics and exploratory data analysis (EDA). Articles in CAR are classified into taxonomies and the literature tracked over 22 years.
Analysis focuses on four taxonomies: foundation discipline, school of thought, research method and mode of reasoning. The paper first examines journals vis-a-vis article publication frequency and dominant taxonomies. Secondly, three assertions concerning the relative posture of the Journal of Accounting Research and the literature are examined. Next the context of the literature is examined through major taxonomies and a crosstabulation of research method vs school of thought. The last part of the analysis focuses on trends within the taxonomies in the 1963-1984 period.
1. INTRODUCTION
The past two decades have witnessed a rigorous process of paradigm development, interdisciplinary “borrowing”, hypothesis testing, and theory refinement in the literature of accounting. Both the volume and breath of this research have created difficulties in understanding its current trends, applying its results, and generating a coherent set of accounting theories that are grounded in its history.
Notwithstanding this difficulty, numerous surveys have provided extensive classification and evaluation of this body of research. However, the focus of these surveys has been typi-cally on an accounting area (e.g., auditing, budgeting) or a school of thought (e.g., human information processing, agency theory). Little effort has been made to examine a larger subset of the literature or to evaluate its results in the light of the entire literature.
1.1 Research Issue
The major goal of this paper is to systematically examine the historical evolution of certain key characteristics of recently published accounting articles. Such examination facili-tates a better understanding of the nature, scope and trends of modern accounting research. Specific attention is given to the nature of scholarly journals, the content of journal articles, and certain trends of the literature (within specific taxons).
1.2 Method
A common set of multiple taxonomies identifies the impor-tant characteristics of 2136 published articles included in the multiple taxonomy databank (MTDB). The large sample allows for the generalizability of our findings to the scholarly accounting literature as a whole. In addition, Exploratory Data Analysis (EDA) [Tukey, 1977] techniques in conjunction with traditional confirmatory statistics and graphics [Chambers et al., 1983; Becker and Chambers, 1984] provide specific insights into the development of the literature.
2. PREVIOUS CLASSIFICATION AND EVALUATION EFFORTS
Most previous accounting research surveys focused on an accounting area, a school of thought, or a research methodol-ogy. Different taxonomies were developed by each author(s) to facilitate their evaluation. Budgeting and auditing are two accounting areas that have received classification and evalua-tion attention. Ijiri, Kinard, and Putney [1968] surveyed the budgeting literature, classifying articles along two taxonomies: areas of application and techniques. Felix & Kinney [1982] surveyed the audit literature focusing their review on the opinion formulation process.
Schools of thought that have been classified and evaluated include behavioral accounting research, human information processing research, and security price research. Hofstedt [1975, 1976] examined behavioral accounting research and classified articles along two taxonomies: accounting versus nonaccounting, and research versus practice. Gonedes and Dopuch [1974] focused on security price research and classified the articles in terms of research methodology. Ashton [1982[ and Libby and Lewis [1977, 1982] reviewed the human information processing literature, dividing the field into a set of paradigms and examining the literature by evaluating articles according to their membership in these paradigms
Research methodologies have been also surveyed. Ball [1971] and Hakansson [1973] surveyed empirical research. Ball [1971] attempted to develop a comprehensive index of accounting topics very similar to the index of an accounting textbook. Hakansson [1973] surveyed empirical research along general accounting issues. In addition, the 1982 supplement of the JAR examined the state of the art of current research methodologies.
Surveys from other points-of-view can also be found in the literature. Several articles in the The Accounting Historians Journal have examined the historical evolution of specific ac-counting topics.1 The Journal of Accounting Literature is dedi-cated to the survey of accounting research studies. Articles published in the JAR are typically oriented towards the evolu-tion of the literature in a field of endeavor within the account-ing literature.
Two recent studies [Brown & Gardner; 1985, 1985a] adopted a different approach. They examined the impact of articles and journals as well as the research contributions of faculty and doctoral programs through citation analysis.
Dyckman & Zeff [DZ] [1984], adopted yet another approach. They focused on a comparison between the Journal of Accounting Research and the broader accounting literature. Their classification scheme is displayed in Table 1.
1For example Rayburn [1986] examined the authoritative literature on Interperiod Tax Allocation.
2For example Baiman [1982] examined agency research in managerial accounting, Kelly [1983] focused on positive theory research and Waller & Jiambalvo [1983] scrutinized normative models in the HIP literature.
TABLE 1. Dyckman & Zeff’s Taxonomy
1. Recent Interdisciplinary Borrowing
• Nonmath
• Math.
2. Mathematical Modeling (other than the above)
3. Conceptual Development
4. Empirical
5. Normative Policy Prescription
6. History
7. Education
8. Other
The objective of the DZ article was to “gauge the contribu-tion of Journal of Accounting Research during its first 20 years, 1963-1982.” (p. 225). It examined the research environment prior to the JAR, the position of JAR in its first decade and various measures of its impact (through circulation, ratings, citations, award winning articles, citations in FASB Discussion Memorandums, and university interest). Among their findings were conclusions that:
•” . . . JAR and its Supplements have hastened the integration into the accounting literature of ideas and methods from other disciplines,” •”JAR . . . has played a primier role in establishing a tradition of empirical research in accounting. . ” •” . . . there is a virtual disappearance of historical
research from JAR.”
Overall, the accounting literature provided extensive taxonomization efforts within particular research areas but little efforts in generalizing results to the entire accounting research domain. This study is intended to fill the void. Fur-thermore, this study adds to the literature by attempting to provide quantitative analysis and results that can be replicated in the evaluation of issues that often are only analyzed in qualitative terms.
3. THE SAMPLE The sample consists of the main articles3 published in the
3In addition to main articles, a few selected Accounting Review notes and Capsules from the Journal of Accounting Research were included in the sample. This subsample inclusion criterion was primarily judgmental.
1963-19844 period, in six5 refereed accounting journals. It in-cludes 2,136 articles (as described later in Table 1). The methodology of taxonomization adopted in the MTDB is dis-cussed in Vasarhelyi, Bao & Berk [1985] and Brown and Vas-arhelyi [1985]. A brief discussion of the categories used in the MTDB follows.
4. TAXONOMIES
The taxonomies of the MTDB were developed to describe three research dimensions (paradigms, research tools, and date reference set) of each article. Paradigms are the basic building blocks of any science [Kuhn, 1962] and are examined through two taxonomies: foundation discipline and school of thought. The research tools dimension (used to develop or test the paradigm) is examined along two taxonomies: research method and mode of reasoning. The data or reference set dimension (used in working with the paradigm) has three taxonomies: accounting area, treatment, and information. Each taxonomy consists of several taxons. Appendix A lists all the research dimensions, taxonomies, and taxons. This paper, however, focuses only on the paradigms and research tool dimensions of research. The analyses based on the data or reference set dimension are too lengthy to fit into this study.
The advantage of having a perfect set of taxonomies [Johnson, 1972], in an information theoretic sense, is its suc-cinctness of description. Unfortunately, a perfect set of taxonomies implies orthogonality among the taxonomies, mutual exclusiveness among these taxons, comprehensive coverage by the classes, and perfect information content (adequate description) by the set of classifications. These features are not easily achievable. For example, Table 6 examines research method versus school of thought using a chi-square technique and shows these taxonomies as not independent. Taxonomies and taxons, therefore, are operationally defined but classifications are judgmental in nature.
41963 is the year of establishment of Journal of Accounting Research.
5The Accounting Review (TAR), Journal of Accounting Research (JAR), Accounting, Organizations and Society (AOS), Journal of Accounting, Auditing and Finance (JAA), Journal of Accounting and Economics (JAE), and Auditing: A Journal of Theory and Practice (AUD).
5. ANALYSIS
The ensuing analysis will first concentrate on journals (and subsequently the above mentioned three DZ findings), then on the content of the literature, and finally on the trends within the database.
5.1 Nature of the Journals
Table 2 presents the number of articles published by the six journals in the 1963-1984 period. The Z value [Lehmann, 1975, pp. 290-297]6 at the bottom of the table is a summary statistics relating the 22 year trend in the data. The table, however, aggregates numbers on a three-year-period basis except for the four-year period of 1963-1966.
TABLE 2. Journal by Year Frequency
Year AOS TAR AUD JAA JAE JAR TOTAL
63/66 295 69 364
67/69 183 71 254
70/72 147 83 230
73/75 128 68 196
76/78 67 134 14 68 283
79/81 72 87 5 57 27 109 357
82/84 71 93 42 67 29 150 452
TOTAL 210 1067 47 138 56 618 2136
Z-value +0.75 -4.40 +1.73 +2.60 +1.85 +4.03
Significant
trend decr. incre.
Table 2 shows that, in terms of number of articles published, TAR dominated other journals until 1979 when JAR became dominant. TAR shows a significant decreasing trend while JAR shows a significant increasing trend. Significance is considered at the 0.01 level.
The significant decrease in quantity by TAR since the 1979-1981 period followed editor change. Stephen Zeff became the editor of TAR in 1979 and decided to segregate TAR’s articles into main articles and notes. The notes section contained articles that “hitherto were published as main articles” [Zeff, 1979, p. 132], and most of the notes are not included in the MTDB. The significant increase in quantity by the JAR
6Lehman [1973] devised a nonparametric statistical method to test the increasing or decreasing trend in data. In this study, a yearly trend is tested. This method is a revised version of the Wilcoxon test. The sign of the Z values indicates the direction of the trend. The level of significance is determined through a normal probability distribution table.
beginning in the 1979-1981 period may be explained by the change in the JAR’s editorial board. Besides the editor, JAR had fourteen editorial members before 1979, and twenty-six members thereafter. The board expansion might have accelerated the review process and therefore stimulated the interest of potential authors.
5.1.1 Publication Taxons by Journals
Table 3 examines the journals’ predominant taxons.7 It displays the major taxons for four taxonomies and the percen-tage of occurrence of the dominant taxons. Cramer’s V values which measure the degree of association of journals and taxonomies are also reported.
Although the journals are different in characteristics, they can be classified into three groups by examining the percen-tages of the major taxons: TAR and JAA, JAR and JAE, and AOS and AUD. TAR and JAA have the same major foundation disciplines, schools of thought, and research methods.
However, the secondary modes of reasoning are different. TAR is more analytical while JAA is more descriptive. This probably can be explained by the fact that about half of JAA’s board of advisors and contributors were practitioners who specifically solicited articles from practitioners whose emphasis was not the technical aspects of accounting research, until a change in editorship in 1986.
JAR and JAE have the same major foundation disciplines, schools of thought, and research methods. However, JAR is more analytical while JAE focuses more on regression analysis. JAE is also more economics/finance oriented. Its editorial policy and editors guide it to a narrow and specific line of research. The journal’s title emphasizes the links of economics and accounting.
AOS and AUD have the same major foundation disciplines. However, AOS is more behaviorally and qualitatively oriented while AUD is more quantitative. In addition, AUD focuses on auditing while AOS includes all areas of accounting.
7A similar analysis, concentrating on comparing AOS to other journals can be found in Brown, Gardner & Vasarhelyi [1987].
8Phi’s and contingency coefficients are also calculated, however, only the most conservative Cramer’s V values are reported.
52
The Accounting Historians Journal, Spring, 1988
TABLE 3. Journal versus Taxonomy
Taxonomy Cramer’s V Journal Major Taxons Percentage
Foundation discipline 0.34 AOS: accounting 31%
psychology 25%
TAR: accounting 45%
economics/ finance 33%
AUD: accounting 85%
psychology 4%
JAA: accounting 69%
economics/finance 15%
JAE: economics/finance 66%
accounting 23%
JAR: accounting 38%
economics/finance 18%
School of thought 0.27 AOS: behavioral 49%
TAR: accounting theory 31%
AUD: statistical modeling 47%
JAA: accounting theory 42%
JAE: statistical modeling 86%
JAR: statistical modeling 50%
Research method 0.21 AOS: analytical – internal logic 41%
opinion – survey 20%
TAR: analytical – internal logic 64%
archival – primary 15%
AUD: analytical – internal logic 36%
empirical – laboratory 23%
JAA: analytical – internal logic 58%
archival – primary 26%
JAE: archival – primary 57%
analytical internal logic 25%
JAR: archival – primary 35%
analytical – internal logic 32%
Mode of Reasoning 0.20 AOS: qualitative 51%
quantitative-descriptive
statistics 12%
TAR: qualitative 46%
quantitative-analytical 31%
AUD: quantitative-analytical 30%
qualitative 23%
JAA: qualitative 49%
quantitative-descriptive
statistics 20%
JAE: quantitative-regression 36%
quantitative-analytical 18%
JAR: quantitative-analytical 35%
quantitative-descriptive
statistics 14%
The following three subsections deal with the three issues raised by DZ relating JAR to the accounting literature observed in this study: interdisciplinary integration, empirical research, and historic research.
5.1.2 JAR and Interdisciplinary Integration
Figure 1 displays the comparison of foundation disciplines between all the journals and JAR. The vertical axis displays the “contribution ratio” reflecting the proportion of articles having accounting as a foundation discipline. Therefore the lower the ratio the more articles having a non-accounting discipline as their foundation.
Figure 1. Contribution from other Disciplines: JAR vs All
70 75 80 85
C
o n t r i b
R
a t i o
Year 1= All journals 2=JAR
The ratio of all articles (plotted with a 1) is lower than that of JAR (with a 2) until 1976 when JAR turns further towards the integration of other disciplines. The lines drawn in the chart use a 2/3 factor for smoothing the point fit.9 This graph does not show that the JAR has hastened integration to a great extent. It was below the average in its integration index during the first 13 years. Each of these journals published numerous articles with a non-accounting foundation discipline. In addi-tion, the majority of AOS and JAE articles had a foundation discipline other than accounting.
5.1.3 JAR and Empirical Research
DZ also argue that the JAR has played a premier role in fostering empirical research. In Figure 2 we label research as empirical when its “research method” subcategory is empirical (case, field & laboratory), archival (primary & secondary) and
Figure 2. Empirical Research: JAR vs Others
Raw Scores
Normalized
200
150
0.4
100
0.2
50
0.0
60 65 70 75 80 85
60 65 70 75 80 85
N – non-JAR, non-empirical E – non-JAR, empirical
n – JAR, non-empirical e- non-JAR, empirical
opinion (survey).10 Consequently, non-empirical “research methods” entail the analytical subcategories of internal logic (“apriori” and analytic) and simulation studies.
Figure 2 compares the publication of empirical (case, field, laboratory, archival, and opinion) vs. nonempirical (analytical) research in JAR versus other journals. The charts show overall frequencies and percentages of articles by period.
The JAR behaves in a similar pattern to the rest of the literature prior to the 1970-1972 period in the overall frequency chart. Further examination of this issue, using a percentage plot, indicates that JAR started with a higher percentage of empirical papers than other journals before 1970 but had a lower percentage thereafter. Since that time the other 5 journals have had a higher average percentage of empirical articles than the JAR. For the 1963-1974 period the data entail only JAR and TAR therefore the chart depicts merely a comparison of these two journals. It shows that until 1974 the JAR had a higher percentage of empirical research. It is because prior to 1974 JAR published annually an issue of Empirical Research in Accounting (the title of its annual research supplement). Since 1974, the JAR decided to expand its annual supplement to include “other types of research” [Dopuch, 1974, p. ii]. Another puzzling observation, in Figure 2, is the sharp decrease in empirical research published in the JAR during the 1976-1978 period complemented by an analogous increase in the non-JAR population. This effect is difficult to explain considering the continuity of the JAR’s editorial policy and the reversion back to “normal levels” in the next period.
5.1.4 Historical Research
Figure 3 displays the number of articles dealing with accounting history topics. The picture shows a small but steady percentage of accounting history research in the literature. The numbers reported in Figure 3 for accounting history are con-servative since the field developed its own journal, The Ac-counting Historians Journal, during this time period, and the journal is not represented in the database.
10Archival primary research relates to the use of empirical data from databases (e.g. COMPUSTAT) and/or financial reports. Data in this case are not generated and recorded by the researcher as in laboratory studies. Archival secondary studies relate primarily to literature studies where the source is articles that discuss a particular topic. This taxonomy was adopted from Buckley [1976] and is discussed in detail by Vasarhelyi, Bao and Berk. [1985, p. 10].
History Publications per Period
500
0
63-66 67-69
70-72
73-75 76-78 Period
79-81
82-84
Figure 3 can be contrasted with the 17 occurrences of the history articles in the JAR shown in Table 4.
TABLE 4. Accounting History Articles in the JAR and non-JAR
Year 63-66 67-69 70-72 73-75 76-78 79-81 82-84
JAR 5 6 5 1 0 0 0
non-JAR 9 5 5 2 4 10 7
The comparison of non-JAR and JAR confirms DZ’s assertions vis-a-vis the JAR, but not vis-a-vis the entire sample. The increase in history articles in the non-JAR population since 1979 resulted from Zeff encouraging this type of research when he became editor of TAR [Zeff, 1983, p. 134].
The content of accounting research in the MTDB can be examined through the composition of taxons within the four paradigms and research tool taxonomies. Table 5 presents the major taxons (taxons with the higher frequency of occurrence) in each taxonomy.
TABLE 5. Major Taxons
Taxonomy Taxons Percentage
Foundation discipline accounting economics psychology
mathematics/decision/game theory 44% 18% 12% 6%
School of thought statistical modeling accounting theory behavioral-other 34% 25% 11%
Research method analytical-internal logic archival-primary empirical-laboratory 51% 22% 10%
Mode of reasoning qualitative quantitative-analytical quantitative-descriptive 36% 28% 11%
Table 5 shows foundation discipline — accounting, school of thought — statistical modeling, research method — analytical-internal logic, and qualitative mode of reasoning dominating their respective taxonomies. The major imports are from economics and psychology. This partially explains the heavy adoption of archival-primary, empirical-laboratory research methods, statistical modeling, and behavioral schools of thought taxons as shown at the right column and bottom row of Table 6.
Futher insight can be obtained by examining multivariate effects among these categorical variables. Table 6 tabulates research method versus school of thought. Cells contain fre-quencies with bold numbers emphasizing high frequency oc-currences.
TABLE 6. Research Method vs School of Thought
BEH. BEH. STAT. ACCNTG. ACCNTG. INSTITU- OTHER TOTAL
-HIP -OTHER MODE THEORY HIST. TIONAL
ANALYTICAL
-INT.LOGIC 20 39 284 422 35 111 67 978
-SIMUL. 2 1 47 8 0 2 4 64
ARCHIVAL
-PRIMARY 4 5 324 45 4 28 33 443
-SECOND. 4 10 26 13 20 7 9 89
EMPIRICAL
-CASE 1 6 7 13 0 3 4 34
-FIELD 1 21 10 2 0 0 3 37
-LABORAT. 87 88 21 11 0 2 3 212
OPINION
-SURVEY 5 56 8 14 0 17 11 111
TOTAL 124 226 727 528 59 170 134 1968
* 166 values missing chi-square = 643 degrees of freedom = 42
Research is clustered in the internal-logic accounting theory, primary archival-statistical modeling, and internal logic-accounting theory combinations.
The high chi-square value suggest that the two dimensions are not independent and indicate that the taxons are not fully orthogonal. They also reflect the real effects of preferences and biases by researchers. Further research may be needed to examine the effect of editor or editorial policy change over time upon the clustering shown in Table 6
li
5.3 Trends within the Database
Table 7 examines the significant current trends of particular taxons in the literature. Two types of trends are examined. The first is the trend in absolute number of articles, and the second is the relative trend of percentage of publications.
11For example Zeff [1983] expressed concern about the effect that the application of modern empirical and analytical research methods may have over the development of thought along classical approaches, in particular accounting history.
TABLE *7. Significant Trends of Taxons
Trend
Taxonomy Taxon Absolute Number Percentage
Foundation Discipline Psychology I
Economics/Finance I
Accounting I I
School of Thought Behavioral-HIPS I I
Behavioral-Other I
Statistical Modeling-EMH I I
Statistical Modeling-Time Series I
Accounting Theory D D
Accounting History D
Institutional D
Research Method Analytical-Internal Logic D
Archival-Primary I I
Archival-Secondary I
Empirical-Laboratory I I
Opinion-Survey I
Mode of Reasoning Quantitative-Regression I I
Quantitative-ANOVA I I
Quantitative-Factor Analysis I I
Quantitative-Nonparametric I
Quantitative-Analytical I
Qualitative D
*I = Significant increasing-trend at the level of 0.01 D = Significant decreasing-trend at the level of 0.01 The trends are determined by the Z-values [Lehmann, 1975, pp.290-297].
Most of the significant trends shown are increasing occurrence of taxons. However, one taxon (accounting theory) has a decreasing trend in absolute number and in percentage. One taxon (accounting history) has decreasing trend in absolute number while three taxons (institutional, analytical-internal logic, qualitative) have a decreasing percentage trend.
Table 7 also confirms that the absolute number of studies with a psychology foundation discipline is significantly in-creasing while the percentage of studies with an economics/ finance foundation discipline is significantly increasing. Archival-primary and empirical-laboratory studies are in-creasing significantly in absolute number and in percentage. Accounting history studies in the database are decreasing in numbers but have neither a significantly increasing nor a decreasing trend in percentage.
Quantitative-regression, quantitative-ANOVA, and quantitative-factor analysis studies are significantly increasing in absolute number and in percentage. Analytical-internal logic, qualitative studies are decreasing significantly in percen-tage.
The analyses presented in Table 7 show the significant trends for the twenty-two-year period. They, however, do not show the configurations of the trends. A different, but substantially more detailed analysis can be performed using a graphics. For illustration purposes, the percentage trends of accounting theory and human information processing are shown in Figure 4.
Figure 4. Percentage Trends
e
e
85
75
80
65
0.2
A 1 1
0.1
A r t i
c 1
0.0
60
e s
Years l=Behavioral-HIP, 2=Accounting Theory
Figure 4 shows a steep decrease in the percentage of ac-counting theory studies over the years, particularly the 1965-1975 decade. Behavioral-HIPs studies, show a significant increase since 1970, then a new area of research.
The graphic analysis has also been applied to other taxons although the graphs are not presented. The graph for founda-tion discipline taxonomy shows that accounting is a dominant foundation. The curve is U-shaped with its dominance in the
1963-1969 and the 1976-1984 periods, and is supplanted by economics/finance and psychology based studies in the middle period.
The graph for the school of thought taxonomy shows a steeply decreasing trend in accounting theory research prior to 1972, and a much flatter decreasing trend thereafter. Both the behavioral and the statistical modeling taxons show a steadily increasing trend, although the latter dominates the former, during the twenty-two-year-period. Both accounting history and institutional research show a flat pattern.
The graph for research method taxonomy shows that both archival and the empirical research taxons have a steadily increasing trend. The analytical research taxon shows a con-tinuously decreasing trend, dominates other taxons until 1981 when it is supplanted by archival research. Opinion research shows a flat pattern.
The graph for mode of reasoning taxonomy shows that quantitative research has a steadily decreasing trend prior to 1975, and a flat pattern thereafter. It dominates other taxons until 1972. The quantitative taxons, in general, have a steadily increasing trend during the twenty-two-year period.
6. Conclusions
This paper examined Contemporary Accounting Research through the classification of articles in this literature along four taxonomies. Both exploratory graphic techniques and confirmatory non-parametric statistics focused the examination on a set of issues to depict the recent development of the accounting literature. In addition, data were presented in such a manner to allow futher examination of other issues by the readers.
Journal analysis led to the pairing of journals in their nature. TAR and JAA were matched, as well as JAE and JAR, and AOS and AUD.
DZ’s assertions about the JAR, quoted earlier, are examined. There has been increased integration into accounting of ideas and methods from other disciplines by both the JAR and other journals. There is a clear increase of empirical research in the sample and the JAR led this pattern through the 1963-1969 period. There is substantial decrease in historical research in JAR but not in the entire literature.
12The advent of the Accounting Historians Journal, and non-inclusion in the MTDB sample leads to the indications that the percentage of history articles in the literature must have substantially increased.
It was found that accounting imports its theories primarily from economics and psychology, particularly since 1976. Modeling studies cover about one-third of the literature. Despite its decreasing emphasis, a priori studies still comprise a substantial part of the literature leading to a large number of qualitative studies.
The analyses of time patterns show many significant in-creasing trends in frequency and some noteworthy decreasing trends. The most significantly and steadily increasing trends since 1963 are the behavioral, statistical modeling, archival, empirical, and quantitative studies. There was a significant decrease in accounting theory, analytical-internal-logic and qualitative studies since 1963. All of the above indicate a decreased emphasis on a priori studies.
The analysis and discussion in this paper barely scratched the richness of the data in the MTDB. A series of research questions such as the transition and evolution of basic paradigms, and the prediction of trends still require extensive studies.
REFERENCES
Ashton, Robert H., Human Information Processing in Accounting, Studies in Accounting Research #17, American Accounting Association, Sarasota, Florida, 1982.
Baiman, S, “Agency Research in Managerial Accounting: A Survey,” Journal of Accounting Literature, Vol. 1, Spring 1982 pp. 154-212.
Ball, Ray, “Index of Empirical Research in Accounting,” Journal of Accounting Research, Vol. 9 No. 1, Spring 1971.
Becker, Richard A. & Chambers, J. M. “S”: An Interactive Environment for Data Analysis and Graphics, ” Wadsworth Inc., Belmont, Calif., 1984.
Brown, L. D. and Vasarhelyi, M. A., Accounting Research Directory: The Database of Accounting Literature, Marcus Wiener Publishing Inc. New York, 1985.
Brown, L. D. & Gardner, J. C., “Applying Citation Analysis to Evaluate the Research Contributions of Accounting Faculty and Doctoral Programs,” The Accounting Review, Vol. LX, No. 2, April 1985, pp. 262-277.
Brown, L. D. & Gardner, J. C., “Using Citation Analysis to Assess the Impact of Journal and Articles on Contemporary Accounting Research (CAR),” Journal of Accounting Research, Spring 1985a, pp. 84-109.
Brown, L. D., Gardner, J. C. & Vasarhelyi, M. A., “An Analysis of the Research Contributions of Accounting Organizations and Society, 1976-1984,” Ac-counting, Organizations and Society, Vol. 12, Number 2, 1987, pp. 193-204.
Buckley, John W., Research Methodology and Business Decisions, National Association of Accountants, New York, 1976.
Chambers, John M., Cleveland, W. S., Kleiner, B., & Tukey, P. A., Graphical Methods for Data Analysis, Wadsworth Inc., Belmont, Calif. 1983.
Chatfield, Michael, “The Accounting Review’s First Fifty Years,” The Account-ing Review, Vol 1, No. 1, 1975, pp. 1-6.
Dopuch, N., “Editor’s Preface,” Journal of Accounting Research: Research Sup-plement – Studies in Financial Accounting Objectives, 1974, pp. ii.
Dyckman, T. R. and Zeff, S. A., “Two Decades of JAR”, Journal of Accounting Research, Vol. 22, No. 1, Spring 1984, pp. 225-297.
Felix Jr., W. L. and Kinney Jr., W. R., “Research in the Auditor’s Opinion Formulation Process: State-of-the-Art” The Accounting Review, April 1982 Vol. 57 pp. 245-27.
Gonedes, J. N. And Dopuch, N., “Capital Market Equilibrium, Information Production, and Selection of Accounting Techniques: Theoretical Framework and Review of Empirical Work,” Supplement to Vol. 12: Journal of Accounting Research: 1974. Studies on Financial Accounting Objectives, 1974, pp. 48-129.
Hakansson, Nils, “Empirical Research in Accounting 1969-1970: An Appraisal,” Accounting Research 1960-1970: A Critical Evaluation, Champaign, I11., University of Illinois, 1973, pp. 137-73.
Hofstedt, T. R., “A State of the Art Analysis of Behavioral Accounting Research,” Journal of Contemporary Business, Autumn, 1975, pp. 27-49.
Hofstedt, T. R., “Behavioral Accounting Research, Pathologies, Paradigms and Prescriptions,” Accounting, Organizations and Society, Vol 1, No. 1, 1976, pp. 43-58.
Ijiri, Y. Kinard, J. C., Putney, F. B., “An Integrated Evaluation System for Budget Forecasting and Operating Performance with a Classified Budgeting Bibliography,” Journal of Accounting Research, Spring, 1968, pp. 1-28.
Johnson, O., “On Taxonomy and Accounting Research,” Accounting Review, January 1972, pp. 64-74.
Kelly, L. “Positive Theory Research: A Review,” Journal of Accounting Literature Vol. 2, Spring 1983, pp. 111-150.
Kuhn, Thomas S., The Structure of Scientific Revolutions, University of Chicago Press, 1962.
Lehmann, E. L., Nonparametrics, Holden-Day, 1975.
Libby, R. and Lewis, B. L., “Human Information Processing Research in Accounting: The State of the Art,” Accounting Organizations and Society, 1977, Vol. 2, No. 3, pp. 245-268.
Libby, R. and Lewis, B. L., “Human Information Processing Research in Accounting: The State of the Art in 1982,” Accounting Organizations and Society, 1982, Vol. 7, No. 3, pp. 231-285.
Paton, William A., “Tendencies in Accounting Literature,” Paper and Proceedings of the Ninth Annual Meeting, Vol. IX, No. 1, American Institute of Accountants, 1925.
Rayburn, F. R., “Chronological Review of the Authoritative Literature on Interperiod Tax Allocation: 1940-1985,” The Accounting Historians Journal, Fall 1986, pp. 89-108.
Tukey, J. W., Exploratory Data Analysis (EDA), Addison-Wesley Publishing Co., Reading Mass., 1977.
Vasarhelyi, M. A., Boa, D. H. & Berk, J., “The Multiple Taxonomies of Ac-counting Research,” working paper, Columbia University, 1985.
Waller, W. and Jiambalvo, J. “The Use of Normative Models in Human Information Processing Research in Accounting”, Journal of Accounting Literature, Vol. 3, Spring, 1983, pp. 201-223.
Zeff, S., “Editorial,” The Accounting Review, January 1983, pp. 129-134.
Zeff, S., “Editorail,” The Accounting Review, January 1979, p. 132.
APPENDIX A: Elements of Research
Dimension Taxonomy Taxon Abbreviation
Paradigms • Foundation Discipline Psychology (P)
Sociology (S)
Economics/ Finance (E)
Engineering/Communication (eN)
Mathematics/Decision/Game Theory/Statistics (M)
Law (L)
Accounting (A)
Management (mnGt)
* School of Thought Behavioral (B)
-Human Information Processing (IIIP)
-Other
Statistical Modeling (S)
-Efficient Market Hypothesis
-Time Series, Econometrics
-Inform. Economics / Agency Theory
-Mathematical Programming
-Other
Accounting Theory (T)
Accounting History (II)
Institutional ( I )
Research Tools *Research Method Analytical -Internal Logic (Anl)
-Simulation—
Archival -Primary (aRc)
-Secondary
Empirical – (Emp)
-Case
-Field
-Laboratory
Opinion Survey (Opi)
* Mode of Reasoning Quantitative -Descriptive Statistic (D)
-Regression (R)
-ANOVA (A)
-Factor Anal., MDS, Probit, etc. (F)
-Non-parametric Statistics (N)
-Correlation (C)
-Analytical (aNl)
Qualitative (Q)
Data or Reference Set Information12
Financial Statements (F)
Internal Information (1)
External Information (E)
Market Based Information (M)
Accounting Area Tax (T)
Financial (F)
Managerial (M)
Audit (A)
Information Systems CD
Treatment13
] 2 This taxonomy was examined in an abbreviated form as described above 13 See Brown A Vasarhelyi (1985), pp.418-420. • This taxonomy was focussed in this paper.
