A holistic construal

Both theoretical and methodological concerns guide organisational research. Indeed, a theory is regarded as incomplete unless it is testable; similarly, empirical research is valued most when guided by sound conceptual thinking. Both are essential when one considers the contribution that a paradigm as a whole or a single study within a discipline makes to knowledge.

Organisational research, however, lacks a formal mechanism for integrating the specification of theory with its test. Rather, the practice has been to formulate a theory in abstract terms and then to test its predictions using only concrete observations. Although there are well-developed criteria to guide the formulation of theory, and statistical and observational standards can be found to direct empirical analysis, the linkages between theoretical concepts and their measurement are often left unspecified or else are stipulated in loose, unverifiable ways.

This failure to represent explicitly the degree of correspondence between measurements and concepts undermines the test of the theory. A hypothesis may be rejected because of inadequate theory, a lack of correspondence between measurements and the concepts that the measurements are intended to represent, and/or excessive random error in measurements. Recognizing this problem, researchers have used a variety of methods to assess construct validity and to test hypotheses[50].

Common approaches include heuristics, such as the Campbell and Fiske (1959) multitrait-multimethod matrix, and statistical models, such as regression and exploratory factor analyses. These methods have serious shortcomings, however, for they make naive assumptions as to the meaning of concepts, provide limited information as to measurement and method error, and examine only primitive aspects of construct validity.

More important, these procedures are usually applied independently of statistical procedures used to test causal hypotheses.

In the typical study, construct validation procedures are used to justify the creation of composite scales, which serve as the single indicators of the theoretical concepts of interest. These scales are then used as input to procedures for testing hypotheses, such as ordinary least-squares (OLS) regression or examination of correlations or cross-tabulations. When this practice is followed, an underlying assumption in the model being tested is that all concepts are measured perfectly by their respective indicators without error; however, measurement or methodological errors are common in most organisational studies, so this assumption is not often justifiable. Indeed, invoking this assumption ensures inconsistency among the different stages of analysis; whereas construct validation procedures typically establish the presence of significant amounts of measurement and/or method error, contemporary hypothesis-testing procedures assume it away entirely.

The failure of organisational studies to control for random and systematic measurement error in testing substantive hypotheses ensures that such hypotheses are tested on data only loosely tied to their theoretical structures. When measurements and concepts do not correspond perfectly, the use of traditional procedures can result in spurious confirmation of inadequate theories, tentative rejection of adequate theories, and/or distorted estimates of the magnitude and relevance of actual relationships. What is needed is a methodological paradigm that permits a direct assessment of the degree of correspondence between measurements and concepts and at the same time takes this relationship into account in the test of substantive hypotheses.

a new methodological paradigm for organisational research is the holistic construal.

The holistic construal has its roots in the physical and social sciences and may be considered a synthesis and extension of the ideas of many scholars, some of whom we note below. The procedures share some features of the positivistic, realist, and instrumental approaches, yet are flexible enough that one can choose or reject certain aspects of any of the approaches, depending on the goals of the researcher and the particular context at hand. The holistic construal is neither rigidly deductive (or formalistic) nor purely exploratory.

Rather, it subsumes a process by which theories and hypotheses are tentatively formulated deductively and then are tested on data, and later are reformulated and retested until a meaningful outcome emerges. The deductive formulation of theory and hypotheses is guided by past research and criteria of internal consistency - e.g., logic, formal rules of theory construction, etc. -yet is regarded in a speculative sense.

An evolutionary process, in which theory and hypotheses are compared to observation and experience, is necessary to refine and move theories closer to the actual phenomena or events they are intended to represent. But at any given time, theories are still imperfect models of actual phenomena or events and must be continually subjected to analysis and criticism.

The holistic construal attempts to integrate the work of philosophers of science, who have developed an extensive conceptual scheme for representing theory but no means to operationalise their scheme, and statisticians, who have developed procedures for estimating parameters and testing hypotheses but have failed to unite their methods with formal criteria for theory construction. The holistic construal is intended to encompass aspects of both the theory-construction and theory-testing phases of organisation science.

Questions and answers – examples

1. Why post-modernist organisation theories appeared in 1970s ?

The period of oil crisis in world economy, with creating of OPEC and start of control on oil prices by petroleum suppliers, leaded up to the principal changing of structure of resources costs for all organisations. At that moment, the markets were quite saturated, so the changing of organisation and of human capital were almost exclusive way to find the appropriate managing system. That is why corporations were deeply interested in the organisation theory.

Date: 2016-03-03; view: 932