Make your own free website on

Critical Research
向上 ] CAPM Review ] [ Critical Research ]


Come Together


All knowledge can be regarded as a social artifact. It therefore is a product of the constituting labour of people as they seek to produce and reproduce their existence and welfare (Habermas, 1978). Therefore, the growing interest in methodology underlying the nature of accounting within organisations and society has generated a range of intellectual stemmed from social and political thought. This article mainly concerns with the positive research and its adoption in accounting approach. This article first uses Laughlin’s (1995) methods for classifying methodology in accounting approach. Secondly, this article describes the nature of three major approach, Positivism, Interpretative, and Critical theory. The third part is the perspectives of positive research in accounting approach and discussion the criticisms of positive research. In the conclusion, this article states the advantage of adopting positive research for investigating empirical accounting phenomena.

Classification of accounting research

Before moving to positive research in accounting, we first examine some fundamental assumptions upon which the accounting approach is developed. The diversity of approach has caused increased tension into literature between various researchers with different methodological assumptions in interpreting the nature of reality, the role of theory and the distinctive characters of empirical experimentation. It is important to know that these various approaches may be subject to aggressive and essentially uncomprehending criticism due to underlying deputes at philosophical level, although these approaches are accepted in terms of the methods employed. Therefore, the research may be rejected with a methodological scepticism and consequently be labelled as nonsensical and weak.

Methodology and Method

Methodology, or the study of method, can be referred as the branch of logic that deal with the general principles of the formation of knowledge. It deals with the conceptual, theoretical and research aspects of knowledge about the world in which we inhabit and is concerned with alternative techniques of discovering such knowledge. In other words, methodology involves both the combination of the nature and role of the discovering process and theoretical formality in identifying the character of the discovering methods (Laughlin, 1995, pp.66). Methods, moreover, are the systematically detailed means through which the knowledge is discovered. Methods reflect the nature of these alternative methodologies within which the methods are nested. Briefly, methodology indicates the research methods deemed appropriate for the collecting of sound evidence. Research method should therefore be permeated with methodological issues.

Burrell and Morgan’s Framework

Burrell and Morgan’s framework is considered as a useful starting point to identify the crucial characteristics for understanding broad streams of social science approaches (Laughlin, 1995). This framework successfully abstracts the key characteristics from the various schools of social science approach and allocates these characteristics into a range of continuums encapsulating. Burrell and Morgan argue that the different approaches of social theory can be analysed in terms of two key dimensions, the nature of social science and the nature of society (Burrell and Morgan, 1979, pp.21). These two key dimensions can be seen continuums with a range of bipolar characteristics at either extreme.

They argue that the assumptions about the nature of social science can be considered as in terms of a subjective – objective continuum, and the assumptions about the nature of social science is thought of a regulation – radical change continuum (Burrell and Morgan, 1979). According to Burrell and Morgan, nature of social science dimension actually consists of five key variables, ontology (the nature of existence and being), epistemology (the nature and theory of knowledge), human nature (the role of investigator), methodology (means to investigate the world), and societal order (perception of society, subjectivism and objectivism). Burrell and Morgan then measure these five variables with the continuum ranging from subjectivist to objectivist. The details are illustrated in the figure 1.








¬    ONTOLOGY     ®





















Figure 1 Assumptions about the nature of social science from Burrell and Morgan (1979)


Moreover, the assumptions about the nature of social science are considered in terms of sociology of regulation and sociology of radical change. The sociology of regulation implies the need to persevere a certain unity and stability in the society and assumes incremental improvement. On the other hand, sociology of radical change supposes the current societal configurations are not optimal and is undermining rather than improving human development, and therefore demands change for the society. Therefore, the two dimensions represent mutually exclusive schools of thought in social science (subjectivism and objectivism) and the nature of society (concerned with regulation and maintenance of order and radical change). From the two by two matrix, Burrell and Morgan classify the social theory into four categories, radical structurist, radical humanist, functionalist, and interpretative. The details are provided in the figure 2. Figure 3 gives a more detailed insight of the nature of these paradigms associated with the various variables in the nature of the social science continuum.











































































Figure 3 Some insights into the nature of the four prardigms

from Burrell and Morgan (1979)


Chua’s Framework

Chua’s framework (1986), regarded as inspired by Burrell and Morgan’s framework (Laughlin, 1995), classifies the accounting literature according to three main sets of assumptions: the domain of knowledge (the belief of knowledge), empirical phenomena (belief about physical and social reality), the relationship of the two (relationship between theory and practice). The underlying assumptions in Chua’s framework are stemmed from Burrell and Morgan’s five key variables to the nature of social science: ontology, epistemology, human nature, methodology, and societal order. Chua therefore divides the accounting approaches into three main categories: mainstream accounting, interpretative alternative, and critical alternative.

Laughlin’s Framework

However, the bipolar dualism introduced in Burrell and Morgan’s framework is thought too simplistic and insulated from the key domain of choice (Laughlin, 1995). Moreover, Laughlin argues that the five variables (ontology, epistemology, human nature, methodology, and societal order) are thought to be implicitly recognised in the various approaches to empirical research. Laughlin further asserts that it is possible to present much of this detail in a rather simpler manner in terms of three broad bands: theory, methodology, and change. Choice in these three dimensions can be expressed through the simple measure ranging from high to low.

The theory dimension can be expressed as the combination of concern related to the level prior theorising and prior theories that may be properly led to the empirical investigation. This thus involves with the ontological assumptions about the nature of the world the observers are investigating, and with the epistemological assumptions about the interpretation toward the world’s materiality along with its generality of representation through previous theoretical attempts. In other words, the theory dimension indicates determining on a view about the nature of the world (ontology) and the way the knowledge is formed and how it relates to the present locus of investigation (Laughlin, 1995, pp.66). The theory dimension can therefore range from high to low levels of usage of prior theorising before actually pursuing any investigation.

The methodology dimension involves a set of procedure or rule that constitutes the nature of the methods in empirical investigation which also links the implicit implication about the role of the observer in the process. Therefore, the eventually conduction of investigation should be involved with either the theoretical models of how the observer should see or mainly based on the perceptual capability of this observer. This embraces the assumption of the level of theoretical formality in characterising the nature of the discovery methods (methodology) and the role of the observer in such discovery process (human nature). The methodology dimension consequently associates for high to low levels of theoretical closure on the methods of investigation.

The change dimension indicates the complex and uncertain and clearly links to Burrell and Morgan’s nature of society assumption. This refers to the attitudes of the observer concerning the worth or otherwise of maintaining or changing the currently situation investigated (societal order). The change dimension refers to high to low levels of critique associated with the status quo and the demand for change in the phenomena investigated. The high level of change indicates those who view that everything is bound to be inadequate and incomplete and therefore demand for change. On the contrary, the low level shows those who view little problem in the status quo. Those in the middle level incorporate their attitude to change more strategically. Their attitudes do not only open to remain some certain aspects of the current functioning but also expose to the challenge to the status quo.



theory choice: levels of prior theorisation


Methodological choice: level of theoretical nature of methods






Positivism (L)

Realism (L)

Instrumentalism (L)

Conventionalism (L)





German Critical Theory (M)

Symbolic interactionism (Kuhn) (L)


Marxism (H)

Structuration (L)


French Critical Theory (L)

Pragmatism (L)


Symbolic interactionism (Blunmer) (L)


Ethnomethodology (L)

change choice: level of emphasis given to critique of status quo and need for change (high/medium/low)


Figure 4 Characteristics of alternative schools

 of thought (Laughlin, 1995)


With this straightforward but profound framework, the main social science approaches can be located and characterised. Figure 4 shows the result of the classification. Each approach is described in the cell referring to its own tacit theoretical and methodological position along with the nature of change as marked as L, M, or H (Laughlin, 1995, pp70). Thus it is possible to identify the various traces of social science approaches to the following three groups: positivist (Comtean), interpretative researches (Kantian/Fichtean), and critical researches (Kantian/ Hegelian). However, it may be better to recognise the difference between each grouping approach by tracing the history of their respective developments.

Brief description of positivism, interpretative research, and critical theory


Positive research assumes, according to Comte, that a world which would demands the absolute descriptions of the empirical situation to be made distinct from any bias inherent in the observers and such descriptions should not be involved with any attitude concerning the need for change in the observable referent (Laughlin, 1995, pp.73). Positivism can be discussed by the following two dimensions. In the methodological level, positivism asserts that the only true knowledge is scientific knowledge, which describes and explains the coexistence and succession of both physical and social observable phenomena. In social and political level, the positivist requires the positive knowledge to allow a new scientifically grounded intervention in politics and social events that would transform society.

In its broadest sense, positivism does not agree with the metaphysics (Ryan, Scapens and Theobald, 1992, pp.10). It is a position that holds that the goal of knowledge is simply to describe the phenomena that exists in the world. The purpose of science is simply to stick to what observers can measure. Knowledge of anything beyond that, a positivist would hold, is impossible. In positivism, science was seen as the way to get at truth, to understand the world well enough so that people might predict and control it.

The positivism believed in empiricism -- the idea that observation and measurement was the core of the scientific endeavour. The empirical reality is referred as objective and external to the subject. Furthermore, the human beings, in positivism, are not considered as the makers of social reality but identified as passive objects for empirical research or experiment (Chua, 1986). The positivism suggests two type of model to postulate the testable theories, hypothetico-deductive model and inductive-statistical model. The former one deals with the results gathered from the test of the hypothesis based on certain assumptions. The latter one involves with the observations from which some theoretical statistics technique inference is derived.

In summary, the positive research assumes the significance of empirical testability. Despite this is insufficient to fulfil whether theories are “verified” or “falsified”, such approach maintains that the scientific explanation can be constituted by empirical test of certain hypotheses.

Interpretative Research

The interpretative research is basically derived from Germanic philosophical interests and emphasises on the role of language, interpretation, and understanding in social science (Chua, 1986). It is concerned with the necessity for the observers to grasp the meaning of the actors. This approach believes a world in which the existence can not be separated from the its perception of the observer. Furthermore, in this approach, the generalisations can not be assumed to exist and the understanding is referred as subjective, specific and subjectively derived. Therefore the social reality is emergent, subjective created, and objectified through human interaction. It pursues only to explain the action and to realise how to produce and reproduce the social order (Chua, 1986).

This approach involves with broad interpretation in nature to discover the meanings individual researcher has for empirical phenomena. It assumes that all actions have meaning and intention retrospectively endowed and grounded in social and historical practices (Chua, 1986). In other words, these meanings are the products of social interaction and can be changed and manipulated through an interpretative process used by individual while dealing with the event he encounters. Laughlin (1995) classifies that it derivatives[1] have the tendency to be low in terms of the theoretical nature of the methods on the methodological choice dimension.

Critical Theory Research

Critical theory research primarily assumes that every state of existence, whether individual or a society, possesses historically constituted potentialities that are unfulfilled. Furthermore, human beings are not restrained to exist in particular state and instead people are able to recognise, grasp, and extend the possibilities contained in every being. However, the human potentiality is restricted by the predominant systems of domination that alienate people from self-realisation. Such material blockage can be found in the material economic and political relations in terms of consciousness that consists of two levels: ideological construct and repression (Chua, 1986). In brief, the major concern of the critical theory research is that the historically oriented theory of society can lead to the potential transformation and change in the nature of society in order to secure a world of freedom and justice.

Critical theory research views individuals as acting within a matrix of inter-subjective meanings. It agrees that the social scientist needs to learn the language of their subjects. In this sense, critical theory research is similar to interpretative research. However, critical theory assets that the interpretation is insufficient in essence. It can not understand that the world is symbolically mediated and shaped by material conditions of domination. Thus, Habermas argues that the social action can only be appreciated in the framework in which the language, labour, and domination conjointly constitute (Held, 1908, pp.307-317). Another radical aspect of critical theory is the concern with social theory. The theory may be manipulated as the products of particular forms of domination and ideology. Critical theory rejects the value position advocated by the orthodox social scientists and argues that the value position “bolsters existing forms of injustice inherent in the current system of property rights and in the capitalist appropriation of economic surplus value” (Chua, 1986, pp.621). The essence of theory hence, in the sense of critical theory, has a critical imperative: the identification and removal of domination and ideological practices.

Positive research in accounting approach

Theoretical Aspect of Positive Research

In the theoretical aspect of positive research, the main concern can be the articulation of the relationship among the models theoretically constructed. Such articulation involves with the quality of the relationship between theory and relevant evidence observed by proper experimental design (Ryan, Scapens, and Theobald, 1992, pp.102 – 112). Therefore, in the theoretical aspect of positive research, a theoretical structure is first developed on the basis of a set of assumptions about reality. Secondly, a set of predictions is generated and may be rejected by empirical test.

The concern about internal consistency (validity) and generalisability would play an important role in establishing such framework for conducting positive research method. In the theoretical aspect, the validity concern comes from the question of whether there are any contradictions in the analysis or logic adopted. For example, a reasonably abstract model may be tractable and easy to operate, it is necessary to ensure the structure is not so simplistic that particular and significant parts of the overall problem structure are ignored. Therefore, it is important to recognise, when conducting abstraction of the target problem situation, that the removal of the particular part of the problem structure would not impair the articulation between the remaining assumptions and the implication of the reality concerned. On the other hand, the concern about generalisability arises from that how specific the theoretical consequence is to the structure employed. For example, Fama and French (1993, 1996) examine simultaneously the factors originally not explained by the CAPM (capital asset pricing model) and generalise the alternative multi-factor model to explain the expected return, instead of using the traditional CAPM in which only one factor is used in determining the expected return.

Empirical Aspect of Positive Research

In the empirical aspect of positive research, the main concern should be the empirical testing of the models theoretically constructed. In this sense, theory should offer a set of empirical implications and theoretical hypotheses that can be investigated by empirical testing. Therefore, the empirical research can analyse or reject alternative theoretical articulations and allow the more general interpretation towards the model theoretical constructed. The concern about the development of hypotheses emphasises on the degree of being specified, unambiguous, and testable (Ryan, Scapens, and Theobald, 1992, pp.102 – 112).

The validity concern, in theoretical aspect, comes from the testing of joint hypotheses: the theory under review and the empirical constructs used in generating the empirical implications for the testing purposes. In this sense, the clear rejection of the theory tested would be unavailable and the result of testing could be contingent. Therefore, when conducting the empirical research it is necessary to carefully identify the operation of specific concepts and alternatives and the articulation between them and the theory and outcomes. For example, when conducting the empirical test of CAPM it is the crucial problem to identify exactly the constitution of the market portfolio. On the other hand, the concern about generalisability arises from that how specific the empirical outcomes are to the particular sample frame. The major consideration in dealing with generalisability problem would be the selection of sample. It can be described into two aspects, the sample size and the bias inherent in the sample selected. There is a possibility that new results would be different that from the existing size of sample when the sample size is largely changed. The result from biased sample would flaw the empirical research and make it meaningless. Hence, in determining the insufficient sample size, the empirical research should consider the trade-off between potential reductions in sampling fluctuations and the increased risk of degrading the representativeness of the sample. Moreover, the sample selection criterion should be ensured to be adequate and subjective.

Accounting as Neutrality

One of the significance in positive research is the imperative of neutrality. Positive research in accounting asserts a means-end dichotomy. It indicates that the accountants should only act on the observation of the most “efficient and effective” method to meet the information need of the manager but not involve themselves with moral judgements about manager’s own needs or goals (Chua, 1986). Because that information is independent from any particular goals and the interest or bias underlying these goals, the accounting information therefore is regarded as the neutral information and as value-free results. Sterling (1979, pp.89) suggests that accountants should act like scientists and produce “ought-to-be” financial statements about the means that are appropriate to the given goal.

Due to the belief of the mean-end dichotomy, positivist accounting research merely considers about the current institutionalised framework of government, market, and organisational forms. Goals of the firm and society are not carried in the consideration. Furthermore, the concern about the property rights, economic exchange, and the allocation and generation of wealth is left to ignore. In this sense, the accountant merely stands in the neutral value position by not evaluating those end-states and consequently processes only the financial information on the means to achieve these states. However, critiques emerge around its essential favour on the status quo and argue such implicit support to the status quo intends to legitimise extant relations of exchange, production, and forms of suppression.

Agency Theory

The agency theory assumes the separation between the owner and the decision-maker and that both the owner and decision-maker are rational economic persons who intends to maximise his utility. Furthermore, the markets are said to be available for managerial skill and information. Such model is to explain the behaviour of individuals as economic agents and how the agents ought to behave. To enhance their capability of generalising the utilities, they needs to relevant and objective information, such as the accounting information. Therefore, the recognition that accounting information is a costly commodity becomes to be a series of contracts that are freely negotiated between rational economic persons. The accounting information is regarded like any other commodity which can be priced and sold (Jensen, 1983). Moreover, the positive accountants should not help the choice in a decision model, but should rather assist the decision-maker to realise the relevant interaction between the variables.

Agency theory has driven positive accounting research into the development of contractual relationship between managers and shareholders. The manager is considered as the agent of the shareholders. Such development can be referred to the managerial incentive contracts and role of financial reporting in supplying the information on which the incentives are based. Another implication of the agent theory in positive account research is the lobby behaviour in standard setting process. This is because that the self-interest assumption underlines of agency theory. In other words, the managers intend to secure through political lobbying the accounting standards that best meet managers’ own interest and indirectly the interests of their shareholders (Watts and Zimmerman, 1978). In this sense, the accounting standard setting is a far from technical and neutral but political process.

Criticisms of Positive Research in Accounting

The first perspective of criticism can emerge from the limitation of those assumptions underlying positive accounting research. Such limitation indicates the assumption underlying the research paradigms will undermine the capability to provide useful understandings of social reality. The positive assumptions are most based on the empirical evidence and belief of the means-end dichotomy. Such inclination causes positive research merely focus on current institutionalised framework of society and ignores the deep understandings of social reality. Laughlin (1995) points out that positive accounting research is based the high levels of prior theorisation which is often emphasised on neo-classical economics. Tinker (1984, 1988) also challenges towards the use of these positive theoretical models.

The second limitation involves with the inherent obstinance to the scientific method. In this sense, accounting is treated as a social constructed and subjective reality. The positive accounting research, while complying with the strict rigor of scientific methodology, has be thought to draw its emphasis on measurement for measurement’s sake. Thus, the scientific method, which mainly investigates objectively the occurring phenomena, is an insufficient paradigm that unable to develop more the perspectives of doing accounting (Baker and Bettner, 1997). They further argue that the scientific obstinance is inappropriate and generates distorted abstractions of social reality (pp.304). Thus, accounting approach has treated as a discipline enchanted with methodological rigor and has little attempt to explore more interpretative and critical perspectives (Power & Laughlin, 1992).

The third limitation can refer to the exhaustive use of quantitative method. Campbell, Daft and Hulin (1982) suggest that the amount of particular research paradigm does not necessarily enhance its contribution to knowledge. Baker and Bettner (1997) further claim that the knowledge can sometimes be contributed more from the innovative and even controversial pursuits. However, if the knowledge is for the practical purpose, it may need the contribution derived from more scientific and systematic methods.


Generally speaking, the interpretative research concerns with the understanding and interpretation about the meaning observed from the symbols and structures of the setting in which the society constructed. Critical approach applies a particular viewpoint in developing more deeply about the influence on the shift of social structure introduced by certain discourse. However, in a practical viewpoint, the value of a theory depends on its capability of explaining the way the world works and the effect of its prediction on the user’s welfare (Ross and Zimmerman, 1990). Despite the increasing amount of debates about the positive research in accounting, positive methodology provides the effective means to discover empirical regularities in accounting choice and explanation for these accounting regularities. The inherent scientific method permits the observer a feasible means to investigate the event occurred in the environment he exists. Moreover, there is no the best methodology for discovering knowledge. The interpretative and critical theory truly has the potential to become the alternatives for positive research in mainstream accounting approach. They are, however, less likely to eliminate the influence of the observers’ value. In this sense, it may not be sufficiently practical and subjective. Positive research, therefore, may produce a better quality in predicting and explaining accounting phenomena by not ignoring the incentives of the individuals who account.


[1] Laughlin (1995) classifies the derivatives of interpretative research as Pragmatism, Structuation, Symbolic interactionism, and Ethnomethodology.