Positivism and “ alternative ” accounting research

Accounting is a vibrant and dynamic area, which in the last decades has assumed prime importance in the economy, in society, and in the world in general. Research in the eld is particularly relevant given the potential e ect that accounting information can produce and disseminate in society (namely among decision and policy makers). However, the research in the area has been accused of being sterile and stagnant. In this Editorial, I intend to assess its development over time and re ect upon its current problems and challenges. Until the 1960s, the dominant theoretical and methodological paradigm in accounting research was normative, based on the assumption that individuals seek to maximize gains and process information in a perfect way, to which they have access at zero cost and that there is no uncertainty (Ryan, Scapens, & eobald, 2002; Scapens, 2006). e idea was that it was the job of accounting researchers to develop models and tools that could be useful for company managers and help them to make decisions. In other words, the researchers should prescribe excellent solutions and models in the accounting manuals, which if properly implemented by managers (i.e. following the rules and prescriptions indicated) would lead companies to achieve higher levels of performance. e search for the answer to the question “what should managers do?” nely illustrates the concerns of normative researchers (Scapens, 2006). e shortcomings of this paradigm and the irrelevance of the theoretical models developed during this period for companies led researchers, such as Robert Scapens in the United Kingdom, to alert the academic and scienti c community to the gap between theory (textbooks) and practice (companies) (Scapens, 1994, 2006; Bromwich & Scapens, 2016). From the 1970s onwards, positivist or empirical research came to be the dominant paradigm, seeking to answer the question “what should managers do?” by observing reality and developing universal laws (Ryan et al., 2002). e formulation of these laws should obey the rules of the scienti c method prescribed for Natural Sciences, so that in explaining observed phenomena it is possible to predict its occurrence in the future. Despite the conceptual advances that positivism enabled in accounting as a science, in essence, this type of research was based on the same assumptions as the normative research. is dependency of accounting on neoclassical economic theory led some researchers to argue that it had become a “subarea” of this economic trend, con ning it to its objectives and assumptions (Williams, 2014; Reiter and Williams, 2002; Reiter, 1998). For positivists, reality is a concrete and objective structure that is external to the researcher and open to being reduced to explanatory (independent) and dependent variables via laws that express their relationship. e construction of complex statistical models associated with the use of large volumes of data that allow the relationships between di erent variables to be tested, expressed in previously formulated hypotheses, is described as a methodological approach par excellence in this type of research. e North American schools and academic journals dominated by the neoclassical economic trend, namely Journal of Accounting Research (JAR), The Accounting Review (TAR), and Journal of Accounting and Economics (JAE), have been great promoters of positivism in accounting research, based on the argument of its superiority in terms of the rigor involved in the quantitative methods adopted and in the predictive potential of the theories formulated (Ryan et al., 2002; Williams, 2014; Reiter, 1998).


INTRODUCTION
Accounting is a vibrant and dynamic area, which in the last decades has assumed prime importance in the economy, in society, and in the world in general.Research in the eld is particularly relevant given the potential e ect that accounting information can produce and disseminate in society (namely among decision and policy makers).However, the research in the area has been accused of being sterile and stagnant.In this Editorial, I intend to assess its development over time and re ect upon its current problems and challenges.
Until the 1960s, the dominant theoretical and methodological paradigm in accounting research was normative, based on the assumption that individuals seek to maximize gains and process information in a perfect way, to which they have access at zero cost and that there is no uncertainty (Ryan, Scapens, & eobald, 2002;Scapens, 2006).e idea was that it was the job of accounting researchers to develop models and tools that could be useful for company managers and help them to make decisions.In other words, the researchers should prescribe excellent solutions and models in the accounting manuals, which if properly implemented by managers (i.e.following the rules and prescriptions indicated) would lead companies to achieve higher levels of performance.e search for the answer to the question "what should managers do?" nely illustrates the concerns of normative researchers (Scapens, 2006).
e shortcomings of this paradigm and the irrelevance of the theoretical models developed during this period for companies led researchers, such as Robert Scapens in the United Kingdom, to alert the academic and scienti c community to the gap between theory (textbooks) and practice (companies) (Scapens, 1994(Scapens, , 2006;;Bromwich & Scapens, 2016).
From the 1970s onwards, positivist or empirical research came to be the dominant paradigm, seeking to answer the question "what should managers do?" by observing reality and developing universal laws (Ryan et al., 2002).e formulation of these laws should obey the rules of the scienti c method prescribed for Natural Sciences, so that in explaining observed phenomena it is possible to predict its occurrence in the future.Despite the conceptual advances that positivism enabled in accounting as a science, in essence, this type of research was based on the same assumptions as the normative research.is dependency of accounting on neoclassical economic theory led some researchers to argue that it had become a "subarea" of this economic trend, con ning it to its objectives and assumptions (Williams, 2014;Reiter and Williams, 2002;Reiter, 1998).
For positivists, reality is a concrete and objective structure that is external to the researcher and open to being reduced to explanatory (independent) and dependent variables via laws that express their relationship.e construction of complex statistical models associated with the use of large volumes of data that allow the relationships between di erent variables to be tested, expressed in previously formulated hypotheses, is described as a methodological approach par excellence in this type of research.e North American schools and academic journals dominated by the neoclassical economic trend, namely Journal of Accounting Research (JAR), The Accounting Review (TAR), and Journal of Accounting and Economics (JAE), have been great promoters of positivism in accounting research, based on the argument of its superiority in terms of the rigor involved in the quantitative methods adopted and in the predictive potential of the theories formulated (Ryan et al., 2002;Williams, 2014;Reiter, 1998).

EVOLUTION OR STAGNATION?
However, this type of research has been heavily criticized by various researchers from di erent parts of the world, namely Europe, and in particular by those dedicated to studying phenomena in management accounting (Williams, 2014;Scapens, 2006;Parker, 2012;Hopper & Bui, 2016).For them, the domination of positivism in accounting research results in serious problems in terms of developing knowledge and innovation in the area, impeding consideration of questions that do not methodologically t into "positivist logic" and that cannot be expressed numerically (Hopwood, 2007;Hopper & Bui, 2016).For critics of positivism, the domination of this paradigm has led to the "ossi cation" and sterility of the research, which the scienti c outputs of North American researchers are, in their understanding, a re ection of.For these researchers, the simpli cation and reduction of reality via its quanti cation and the formulation of hypotheses has not contributed to a better understanding of it and to predicting it.
Positivist type research assumes that reality is much simpler than it really is, seeking to isolate relationships between observed variables in the context in which they exist and studying them independently and in isolation.
is conditions the contribution that positivist research makes to explaining reality.In spite of this, a very signi cant portion of the academic community insists on using it.Anthony Hopwood (2007), in his "Presidential research lecture", presented at the Annual Meeting of the American Accounting Association, observes the following: More and more, accounting research is seen as too cautious and conservative, too rigid and traditional, and insu ciently attuned to grapple with the new and to embrace novel insights and bodies of knowledge.Instead of being enthusiastic about the emerging gaps in our knowledge, it is as if the accounting academic community prefers to concentrate on clues that arise from the existing research traditions.(p.1370).
Another reason for criticizing positivist research involves the rigor regularly attributed to positivism, which is the result of adopting sophisticated mathematical tools and statistical models; this rigor is an argument that has been instrumentally used by positivists in order to hierarchize the quality of the research and to demonstrate the superiority of the work they carry out.However, the rigor of positivism has been described as a "myth" by its opponents (Williams, 2014).According to them, there are two aspects that warrant questioning the rigor of positivist research.First, the concepts ("constructs") on which accounting is based are socially constructed (i.e. they are interpretative) as a result of human interaction.For example, the concepts of "assets", "liabilities", and "pro t" derive from human conventions, which have been discussed and revised over time.e existence of implicit conventions for measuring assets, liabilities, and pro t require value judgments to be made and there is no consensus with regards to the way these constructs should be quanti ed and measured.Measuring assets, liabilities, and pro t, or any other accounting construct, is not exactly the same thing as measuring phenomena in natural sciences, where natural measures exists (kilometer, minute, liter, etc.) that are intrinsic to elements in the physical world and that enable the phenomena being studied to be unequivocally and consensually measured.In contrast, in accounting the measures are only attempts to numerically represent interpretative concepts.In spite of this, positivists in accounting base their research papers exclusively on numbers, which are not questioned and contextualized because it is believed that they re ect reality in a complete and absolutely true way.It is therefore not surprising that the outputs resulting from positivist research are o en accused of sterility and lacking in relevance to practical knowledge.As Gilles (2004) re ects, although the physical world appears to be qualitative because of the diversity and richness of the elements that constitute it, it obeys precise quantitative laws.In contrast, the reality that accounting inhabits, although it can appear to be objective, in fact covers questions of an essentially qualitative nature, which cannot be captured via purely mathematical models.As Parker (2012) observes, in practical terms this means that accounting o en only measures that which is of no importance.
e second argument that critics of positivism use to question the rigor of positivist research relates to the complexity of accounting as a social and human practice.
is argument derives from the fact that the numbers that express accounting information are o en not independent from the theory being analyzed.According to Williams (2014, p.880), this is the "clock problem".e researcher exempli es this problem using an experiment involving the effects of alcohol consumption on individuals' motor reaction capacities.e reaction time to alcohol consumption is measured using a high precision clock.
e times obtained from the measurements taken by the clock are the same, independent of whether they reject or con rm the theory that was used in the study.is means that the time variable and the clock that measures this variable are independent variables that are outside of the theory (and this should be the case).In other words, the way the clock works and how time is measured should not depend on the theory.For Williams (2014), "independence of the clock" is not always veri ed in the accounting research.For example, in studies on earnings management, positivist researchers unquestioningly use data that has potentially been manipulated by managers when they analyze the hypotheses regarding earnings management.
e "clock problem" is described as something that a ects a large number of studies in nancial accounting. is question is a lot more serious given that reality is complex and involves many relationships that positivists do not recognize in the numbers that form the bulky databases with which they work.It is only by considering and understanding the contexts in which these numbers are produced that it is possible for researchers to produce outputs that adhere to reality.As Williams (2014) mentions, "Rigorous accounting research may be rigorous, but it currently lacks the one thing that this concept of rigor requires -rigorous measurement" (p.882).
In summary, the characteristics that are intrinsic to the reality of the phenomena and activities involved in accounting warrant the concerns and doubts regarding the real rigor that quantitative studies enable: the rigor purported by positivist researchers may, in the end, be no more than a "mirage".

ALTERNATIVE APPROACH
ese positivist research problems as a whole have led various researchers in accounting to distance themselves from the positivist theories and adopt "alternative" theoretical and methodological perspectives (Parker, 2012;Hopper & Bui, 2016). is distancing from the mainstream paradigm, as I have already mentioned, has occurred mainly in management accounting, with researchers from the United Kingdom and regions such as Scandinavia assuming an important role in the movement since the end of the 1970s (Bromwich & Scapens, 2016).Basically, these researchers defend pluralism in accounting research by adopting so-called alternative theoretical and methodological approaches that make it possible to study it in a holistic way within the context it forms part of, without reducing it to a mere set of mathematical variables.
From the perspective of "alternative researchers", positivism does not enable any in-depth understanding and explanation of what is observed (i.e. it does not enable "why?" and "how?" questions to be answered), but essentially involves describing and identifying regularities in phenomena, which is very di erent from understanding.Hopwood (2007) was one of the researchers who integrated this movement and who contributed most to the spread of the alternative paradigm, having founded, in 1976, the rst academic journal aimed at publishing accounting research of a multidisciplinary nature (Accounting, Organizations and Society -AOS) (Walker, 2016).In the opinion of this academic, it is essential for accounting to be studied in the social, economic, political, institutional, and technological context it forms part of, in order to enable rich, detailed, and contextualized information to be obtained with regards to accounting practices.Only this way can reality be understood and a source of illumination rather than dogma be constituted.As Hopwood (2007) observes: For the understanding to be a source of illumination rather than dogma, it should have a dynamic of change.Seen from this perspective, accounting, as a practice, can be and indeed should be constantly examined, reexamined, interrogated, and criticized in the world of knowledge.Rather than being a discipline in its own right, accounting needs to draw on a variety of sources of illumination and understanding.It has been and should continue to be a place of interdisciplinary investigation.(pp.1370-1371).
e adoption of a qualitative methodological research approach, rooted in organizational and sociological theories, is defended by alternative researchers.In contrast to the scienti c method, which is based on observation, the naturalistic methods derived from social sciences are proposed for forming and testing hypotheses and con rming/rejecting prevailing theories (Parker, 2012).The adoption of naturalistic methods, by means of qualitative research, is of key importance in this type of study. is means that the alternative paradigm is essentially concerned with what constitutes di erence and is unique, as opposed to the positivist tradition of studying aggregates and large datasets.Ontologically, alternative researchers describe the essence of reality as something that is subjective, abstract, and complex, being a product of social and human practices; for these researchers, it is impossible to quantify and reduce reality, unlike what positivists believe (Ryan et al., 2002).
In order to understand reality, researchers have to interact with the evidence and eld and cannot study reality from afar, as positivists do.Conducting explanatory and exploratory case-based research has been widely recommended as a way of acquiring knowledge about the speci c and particular, and this method is widely accepted by the editors of various academic accounting journals in Europe (e.g.Management Accounting Research -MAR, European Accounting Review -EAR, Accounting and Business Research, among others), Australia (Accounting, Auditing & Accountability Journal -AAAJ, Journal of Accounting and Organizational Change -JAOC), and New Zealand (Qualitative Research on Accounting and Management).e case studies prescribed by the alternative paradigm are rooted in organizational and sociological theories, allowing them to distinguish themselves from merely descriptive reports on accounting practice (Bromwich & Scapens, 2016).It is the interaction between theory and practice reported through case studies that contributes to the knowledge.eories of a social and organizational nature, such as political economy, labor process, new institutional sociology, critical realism, and actor network, among others, and authors such as Habermas and Giddens, are commonly adopted by alternative researchers (Walker, 2016).
Research based on case studies is of key importance in this type of paradigm, as it allows for in-depth questions to be asked with regards to "why" and "how" certain practices occur in organizations.Questions related to the distribution of resources, power relations, class, gender, relationships between individuals, as well as relationships between organizations and their surrounding environments, are topics that are analyzed via the adoption of case studies (Parker, 2012).
Unlike the scienti c method, which uses large volumes of samples (observations) in order to produce statistical generalizations for a whole area, as I mentioned previously, the case study method analyzes the particular and speci c, with the aim of understanding it and constructing theories.Although they do not provide statistical generalizations, by replicating them (a typical procedure in experimental logic), case studies make it possible to con rm whether the theoretical lessons produced previously explain the practices observed.In this sense, it is argued that the case study method produces generalizations, including of a theoretical (or analytical) nature.
An analogy can even be established between the quantitative methods of positivism and the qualitative methods of alternative research, where the notions of reliability and validity belonging to positivist research (which show the capacity for empirical results to be the same if a study were repeated) are substituted, in alternative research, by the notion of credibility -which, in the context of qualitative research, means formulating the most appropriate research questions, gathering adequate evidence for the study being developed, and adopting research methods that allow for answers to the research questions to be obtained.ere are di erent tests that make it possible to evaluate whether a study is credible or not; the authenticity criterion (related to the existence of evidence that the researcher was in the eld and gathered information regarding the processes being studied), the plausibility criterion (which asses the capacity for the arguments to be presented logically to the person analyzing them), and also the relevance criterion (whether the research contributes to the theoretical and empirical knowledge), are the main criteria used to test credibility in this type of research.

CHALLENGES IN THE WAY
Over the last four decades, alternative accounting research based on qualitative methods has ourished and proven to be fruitful, as can be currently assessed via the large number of conferences, workshops, doctorate courses, funded research projects, and academic journals in the area (Bromwich & Scapens, 2016).If this type of research were alternative at the end of the previous century, it no longer appears to be so now, and probably constitutes the dominant paradigm in management accounting.Despite this dynamism, however, many challenges lie in the path of accounting research as a scienti c area.First, there are deep divisions with regards to what is understood in the area to be "good science"; these divisions have led to "scienti c radicalism" and the appearance of "tribes" -or "islands", as Bromwich and Scapens (2016) call them -centered exclusively on certain methodological and theoretical perspectives that neglect to analyze questions outside of what they have de ned to be their eld of study. is type of positioning makes innovation in the area di cult and leads to a perpetuation of the same theoretical and methodological perspectives, resulting in a situation, which many currently describe, of sterility in the production of knowledge.
Recently, proposals have been made by some researchers for the theoretical triangulation and adoption of mixed methods, in order to connect and build a bridge between the various researcher "tribes" (Modell, 2009;Lukka, 2010).However, these proposals have come up against various barriers, whether due to the di erent ontological and epistemological assumptions on which the positivist and alternative paradigms are based, or due to the di erent agendas and concerns of the researchers who defend these paradigms.From the perspective of alternative researchers, their positivist colleagues maintain a closed and hegemonic perspective with regards to what they believe to be "good science", making any attempt to connect di cult (Parker, 2012).e proposal for building a bridge between positivists and non-positivists has been the object of criticism; for some, this attempt to connect merely represents an e ort on the part of the alternative paradigm to be closer to the deductive logic belonging to positivism, without there being any signs on the other side of attempts to converge with the qualitative and naturalistic methodologies (Lukka, 2010).
A second challenge that lies in the path of accounting research relates to its lack of relevance for practice and for resolving the problems in organizations and in society in general (Bromwich & Scapens, 2016).Both types of research seem to present difficulties with regards to this question.In the last two decades, each one of these intellectual traditions has come to center almost exclusively on questions of a theoretical and methodological nature, making signi cant e orts to develop theoretical contributions, but systematically neglecting the production of relevant and useful knowledge that can be put into practice.e result of this situation is visible in the lack of dialogue currently observed between the academic community, policy-makers, and society, as well as the disinterest of academics with regards to systematizing the results of their research by resorting to simple language that is accessible to managers and decision-makers (Bromwich & Scapens, 2016).In recent years, academic journals such as Management Accounting Research and Qualitative Research on Accounting and Management have sought to encourage interventionist type research as a way of bringing the theory closer to practice, but the results, although encouraging, are still relatively limited (see the special edition on this published by QMAR in 2010 -volume 7, issue 1).
I cannot fail to mention that the challenges (and problems) that accounting research currently faces are directly related with the pressure for scienti c production from accreditation and funding agencies, as well as the way of processing the management of university careers.
The research that is produced in accounting serves di erent purposes, being an instrument par excellence for evaluating the productivity of universities in the eyes of national and international accreditation agencies and funding institutions, as well as being the basis for decisions regarding the progression of academics in universities.e high competitiveness observed nowadays between universities (with a tendency to intensify more and more), obliges them to ght for the best ranking positions, for the most prestigious accreditations, and for the most productive researchers, in scienti c terms.The phenomenon of globalization and the increase in international student mobility has contributed to raising the importance of rankings, accreditations, and university reputation.In this context, where the scienti c prestige of universities is critical, scienti c production (in scienti cally prestigious academic journals subject to peer review) assumes overriding relevance in the activities developed.It is thus not surprising that researchers choose conservative strategies that enable them to access the desired publications safely and quickly.e questions to study and methodologies adopted are not chosen using criteria involving innovation in the production of knowledge and the potential to contribute to practice, but are based fundamentally on criteria that ensure the quick academic progression and scienti c reputation of researchers and institutions (Ter Bogt & Van Helden, 2014).
Research on questions accepted by the current mainstream that uses the methodologies and research method prescribed by this trend is a way that researchers nd in order to progress more easily in academic terms through the university hierarchy and to deal with the scienti c pressures resulting from rankings and accreditations; the methods that positivist research defends have already been tested in the last decades and are "safe", unlike the innovative methods prescribed by alternative research.is type of strategy has been encouraged by various players in the sector, namely university presidents, college and department directors, and accreditation and funding agencies; using this type of criteria is a simple and e ective way to ensure legitimacy in the processes in which all of these players are involved.

FINAL REMARKS
Unfortunately, the consequences of these decisions have been very serious, contributing to intensifying preconceptions and the barriers between positivist and alternative researchers.Due to the rising importance and impact of accounting in the economy, in society, and in the world, it is urged that together we re ect upon how the research in the area, as a whole, could be an e ective instrument of social and economic progress.e existence of "tribes" and "islands" of researchers who refuse to converse and close themselves o to discussing questions of an exclusively theoretical and methodological nature contributes nothing to the desired advancement.It is only by accepting that the phenomena in which accounting is involved are complex, multidisciplinary, and require analysis from di erent angles, using di erent lenses, will it be possible to begin a proli c dialogue between the various researchers, in order to e ectively contribute to resolving problems in practice.Perhaps the nancial crisis that has a ected the world in recent years and has led to greater pressure and control on the part of research funding agencies with regards to the way public resources are used for improving society will be a factor that forces this dialogue.