Publishing in the organizational sciences: an extended literature review on the dimensions and elements of an hypothetico-deductive scientific research, and some guidelines on "how" and "when" they should be integrated.

Fillion, Gerard

ABSTRACT

In Publishing in the Organizational Sciences, professors Cummings, Frost, Taylor, Deetz, Nord, Staw, and Daft raised many important issues with respect to the actual publication system. Moreover, in two courses of my doctorate, I had the possibility to evaluate different qualitative and/or quantitative scientific research (most of them being hypothetico-deductive). This experience then led me to note several shortcomings pointed out in the literature. Inspired by these two contextual aspects, this paper aims to enhance hypothetico-deductive scientific research in organizational sciences. Its objective is twofold: (1) it carries out a synthesized extended literature review on the dimensions and elements that can be included into such the research; and (2) it suggests to authors some guidelines on "how" and "when" they should be integrated into them. I think this paper can bring responses to some issues raised by professors Cummings, Frost, Taylor, Deetz, Nord, Staw, and Daft. From a theoretical point of view, it provides authors with a single very concise theoretical document that brings together and examines all of the actual dimensions and elements to conduct and describe hypothetico-deductive scientific research, whether they are quantitative, qualitative, or both. Further, it can even be used by editors and reviewers as a guide to evaluate hypothetico-deductive scientific research submitted by authors for publication. From a practical point of view, this paper brings a tool which can contribute to quality control of the production and dissemination of scientific knowledge in organizational sciences.

INTRODUCTION

Publication is a highly important activity of scholarly life. It allows, to a large extent, the dissemination of scientific knowledge. Thus, while remaining of a great integrity and fairness, it is essential for editors and reviewers to screen out low-quality works and force high standards in knowledge production among authors to maintain a high level of quality of published works and to protect the status of the profession. Moreover, it is extremely important of always keep in mind that the production of scientific knowledge must not be a race at the number of publications and that it is far better to put more emphasis upon publication quality rather than quantity (although the two are not mutually exclusive). In other words, even if scholarly life has often become an issue of "publish-or-perish", we must remain highly concerned about the quality of the production of scientific knowledge.

In my view, all scientific research, no matter it is quantitative, qualitative, or both, must be conducted with a high level of rigor and structure. By rigor and structure, I mean utilization of methods and models already established in existing literature, logical consistency between various steps of the research, strength of argumentation, as well as appropriate identification and coherence of the different parts of the article describing the research. But some will say that it is not quite to show rigor and structure in a scientific research, we must keep also place to the researcher's imagination and creativity. So I must answer that they are right. By imagination and creativity, I mean the enhancement of existing methods and models, the use of new methods and models, the logical presentation of new ideas, and the way to organize the ideas in a well-structured paper. In short, while showing a high level of rigor and structure to conduct and describe a scientific research, it is quite possible to keep also place to the researcher's imagination and creativity. And this paper is exactly oriented in this perspective.

The paper focuses on hypothetico-deductive scientific research (quantitative, qualitative, or both). More specifically, this one aims to enhance hypothetico-deductive scientific research in organizational sciences. Its objective is twofold: (1) it carries out a synthesized extended literature review on the dimensions and elements that can be included in such the research; and (2) it suggests to authors some guidelines on "how" and "when" they should be integrated into them. The paper is organized as following. First, I discuss the two contextual aspects which led me to write such a paper. Second, I conceptualize the dimensions and elements of an hypothetico-deductive scientific research. Third, throughout four relevant scenarios, including published research examples, I propose to researchers in organizational sciences some guidelines on "how" and "when" these dimensions and elements should be integrated into hypothetico-deductive scientific research. Finally, I bring my personal view about the theoretical and practical contribution of the article for the production and dissemination of scientific knowledge in organizational sciences.

CONTEXTUALIZATION

It is important here to emphasize the fact that two specific contextual aspects led me to write a paper that aims to enhance hypothetico-deductive scientific research in organizational sciences. On the one hand, we must be well aware that existing literature raises numerous issues with respect to the actual publication system. On the other hand, in two courses of my doctorate, I had the possibility to fulfill this interesting, enriching, and challenging task, but not always easy I must acknowledge, to evaluate different qualitative and/or quantitative scientific research (most of them being hypothetico-deductive). This experience then led me to note several shortcomings pointed out in the literature. Let me now describe in more detail these two contextual aspects.

In Publishing in the Organizational Sciences, professors Cummings, Frost, Taylor, Deetz, Nord, Staw, and Daft raised many important issues with regard to the actual publication system. Clearly, I will not stress here all of these issues. Rather, I will evocate only some of them which are closely linked to the object of this paper. First, as Cummings and Frost (1995a) point out, "Deciding what is false or even faked knowledge and distinguishing what is good scholarship from that which is the work of charlatans is not a simple matter" (p. 9). According to Frost and Taylor (1995), "Major sources of irritation and disaffection for dissatisfied readers tend to be the lack of quality in journal content, sterile journal material, and irrelevance of articles. Authors identify high rejection rates for their manuscripts (for example, four out of five submissions) [...]. Journal editors sometimes express frustration with the poor quality of manuscripts and identify their own concerns as being high workloads [...]. Reviewers are often dissatisfied with high workloads and relatively low extrinsic rewards" (pp. 14-15).

As Deetz (1995) says, "Publication has often become more of a credentialing process certifying expertise and assuring stature and appropriate club membership than a pursuit of socially important understanding" (p. 46). According to Deetz (1995), we can create systems and structures that aid the open pursuit of common understanding in changing our fundamental conceptions. In this perspective, he suggests to disconnect career reward structures from knowledge production activities, to eliminate traditional refereeing processes, and to focus on problems and provide answers in journal space rather than provide isolated expressions of claims. In Deetz's (1995) view of knowledge production, we must be able to approach major problems with consensual procedures and to have important innovative discussions that reveal genuine value differences and advance the role of professional knowledge in a more ideal democracy. Nord (1995) points out that in the Harvard-type case he used to organize an MBA class session in which the publishing-system-as-organization was presented as a case discussion to students, it is emphasized that the journals play a major role in the allocation of rewards and tenure in a field. "In fact, publication of at least several articles in the leading journals was described as a 'rite of passage' to permanent membership" (Nord, 1995, p. 65). Nord (1995) argues that the work of any field is to build theories, attract support from external constituents, and produce high quality information. He attributes the low status of organizational sciences to their failure to perform well on these tasks. Nord (1995) also discusses the virtues of the publication system as a control system that ensures quality, protects consumers from fraud, forces high standards among producers, and protects the status of the profession. On the other hand, Nord (1995) emphasizes the fact that the centralized control structure of the publication system is a source of both a lack of innovation and a lack of replication of ideas. According to Nord (1995), scientific journals have become too mechanistic, too much like palaces. He suggests the need for alternative organizational structures, such as adhocracies and tenets, to balance the existing ones.

Moreover, as Cummings and Frost (1995b) indicate, "It is not possible to talk for very long about publishing in the organizational sciences without addressing issues of relevance and rigor of what we publish in our field" (p. 79). According to Staw (1995), "Certainly what is relevant may not be rigorous, what is rigorous may not be relevant, and it is extraordinarily difficult for research to be high on both of these dimensions" (p. 85). Staw defines relevance as depicting the importance of a finding or idea for the advancement of knowledge. On the other hand, he raises a major problem with regard to confusion between a contribution to the literature and an advancement of knowledge. In fact, Staw (1995) says that "[...] we routinely judge the significance of a research paper by its contribution to the pile of studies already conducted and archived in the journals rather than its contribution to our understanding of organizations" (p. 86). As for the rigor, Staw (1995) defines it in terms of strength of inference made possible by a given research study. He also argues that strength in argumentation is central to a rigorous work. Furthermore, Staw (1995) raises an inevitable issue in publication, that is, the trade-off between normal science, which strives to achieve a set of replicated findings and clarified theoretical relationships, and creativity. From his point of view, it seems that publications are almost always biased toward normal science. "Our own creative ideas are criticized as shallow, ungrounded, inconsistent with existing theory, or just plain wrong. Our methods are often viewed by reviewers as deficient, flawed, and inappropriate when they are, of course, cleverly adapted to the new theory or type of data. As authors, we try to innovate, but are soundly rebuffed" (Staw, 1995, p. 93).

Finally, from the analysis of 111 manuscript reviews, Daft (1995) stresses the 11 more frequent problems that motivated his recommendation to editors to reject the manuscripts submitted by authors: (1) no theory; (2) concepts and operationalization not in alignment; (3) insufficient definition (theory); (4) insufficient rationale (design); (5) macrostructure (organization and flow); (6) amateur style and tone; (7) inadequate research design; (8) not relevant to the field; (9) overengineering; (10) conclusions not in alignment; and (11) cutting up the data.

As for the second aspect I evocated above, it is a lived experience in the setting of two courses of my doctorate whose the objective was to evaluate both quantitative and qualitative scientific research, most of them being hypothetico-deductive. Given in graduate school it is not very common for students to critically review the work of others (Rousseau, 1995), I am therefore fortunate, if not privileged, to have experienced this. In fact, in the first course, I evaluated only two quantitative research. As evaluation tool, the professor had suggested Davis and Cosenza's (1993) framework. I have then used this framework, but it seems to me very relevant to add it other essential dimensions and elements to evaluate a scientific research from all of its aspects.

Having greatly appreciated these first evaluations, I chose a second course related to student's development of a state-of-the-art knowledge of the more recent scientific works in the field of information systems (IS) and the introduction to critical review of articles. Given this time the professor had not suggested a specific evaluation tool, I have then decided to develop my own framework to this effect in integrating into it all of the actual dimensions and elements I found in key articles and best reference books on research in organizational sciences. Each week of the semester, we had to make a well-founded and realistic constructive critical evaluation of an article. Thus, to act as a reviewer in these two courses of my doctorate allows me to be confronted with several shortcomings pointed out in the literature.

Indeed, in my manuscript reviews, I often noted a ill-defined research objective, no research problem and question to justify the research need, a deficient or totally absent relevant literature review, no theoretical research model to guide the research or future research, no definition of the constructs and variables used in or emerged of the research, no hypotheses or propositions to test the research model, the weakness of the methodology used to conduct the research (for example, an inadequate research design, no sample and/or data collection description, no choice and justification of the data collection and/or data analysis methods, as well as no constructs measurements), no constructs validation, a deficient data analysis, an inadequate verification of the hypotheses or propositions, no answer to the research question, no comparison of the research findings with existing theory, no theoretical or practical research contribution to the advancement of scientific knowledge in the field involved, no research limits, no proposition of new research ideas to other researchers, no conclusions or conclusions not in alignment, as well as a poor research paper form (for example, illogical organization of the ideas and ill-structured sentences). We can see here that some shortcomings I observed in my manuscript reviews are similar to those noted previously by Daft (1995). Thus, to be confronted with these numerous shortcomings in the papers I reviewed in these two doctoral courses has been, in fact, the trigger element of the proposition of the dimensions and elements of an hypothetico-deductive scientific research I inventoried in the extended literature review carried out to develop my evaluation framework.

CONCEPTUALIZATION OF THE DIMENSIONS AND ELEMENTS OF AN HYPOTHETICO-DEDUCTIVE SCIENTIFIC RESEARCH

To conceptualize the dimensions and elements of an hypothetico-deductive scientific research, I took into account three scientific research paradigms: quantitative, qualitative, and multimethod (a combination of different elements of the first two). Gauthier (1992) (1) defines a scientific research paradigm as:

 A set of implicit or explicit rules guiding the scientific research
 for some times in providing, from knowledge universally recognized,
 the ways to formulate the problems, to conduct the research, and to
 find the solutions (p. 568).

Thus, each dimension and element inventoried in my extensive literature review applies to the three scientific research paradigms discussed above. And several elements might be included in different dimensions according to the researcher's style, the research paradigm (quantitative, qualitative, or both), the type of research (field experiment, laboratory experiment, field study, case study, longitudinal study, etc.), the research approach (data-driven (2), theory-driven (3), model-driven (4), etc.), and so on. The dimensions and elements that can be included into an hypothetico-deductive scientific research are diagramed in Figure 1. As shown in Figure 1, the diagram is articulated around 10 dimensions (see the internal circle) and 30 elements (see the external circle). In fact, the diagram is organized so that each dimension of the shaded internal circle can integrate one or several different elements of the external circle. It is fundamental to see the diagram depicted in Figure 1 just as a basic configuration of an hypothetico-deductive scientific research. Thus, according to the different factors related to the researcher and the research mentioned above, a researcher can include some elements in a certain dimension while another can include different elements in the same dimension, hence the shaded within the internal circle as well as the dashed lines at the boundaries of each area dimension-element of the diagram. Of course, the set of dimensions and elements diagramed in Figure 1 remains flexible and open to the addition of other ideas, dimensions, and elements. To better visualize the impact of the dimensions and elements of the diagram on scientific research, in the following subsections, I briefly present the dimensions whereas I provide the reader with a more in-depth discussion about the elements that can be integrated into them as well as their relevance in the research.

[FIGURE 1 OMITTED]

Dimensions of an Hypothetico-Deductive Scientific Research

Abstract

The first dimension of the diagram depicted in Figure 1 is the abstract. It is an important dimension because it provides very quickly the reader with an overall view of the research. We must recognize that it is really impossible to read all of the scientific research papers in our field and that we must then make some choices of reading. So the abstract is a precious tool to help us in these choices. In addition, it is an essential dimension to submit a paper for publication in most of the scientific journals and reviews, whether they are in print or electronic format on the Internet.

Introduction

The second dimension has to do with the introduction. It is really useful to put the reader in context and to provide him/her with a good insight into the discussion that follows. In effect, all document, no matter it is a research paper, a book, a doctoral thesis, a student's work, a business report, or others, must have an introduction. A document without introduction is like a good meal without appetizer: it lacks something!

Literature Review

Literature review is a very important dimension of the research. Its primary objective is to synthesize the relevant existing literature allowing to raise the research problem and question that require an investigation on the part of the researcher. Thus, a good review of the relevant existing theory allows the researcher to support his/her investigation on strong foundations.

Theoretical Approach

This fourth dimension of the diagram aims to present the theoretical approach to solve the problematic situation identified previously. In short, it is the conclusion of the conceptual work carried out until now by the researcher. It is in this extremely important part of the research that the researcher theoretically shows how he/she will get the response to his/her research question.

Methodology

Methodology is the fifth dimension. It is the core of the research, in fact. Unlike the previous dimension that shows the theoretical solution to the research problem raised by the researcher, this one shows the practical or pragmatic solution. It is in this part that is integrated and articulated the set of decisions to solve the problematic situation in a coherent way. Clearly, these decisions can differ on some points in accordance with the different factors related to the researcher and the research discussed previously, that is, the researcher's style, the research paradigm, the type of research, the research approach, and so on. What really matters, at this point, is to take the more effective decisions possible according to all of these factors and, especially, the fact that the theoretical approach, the results, and the discussion about the results must all correspond to this research operationalization. In brief, there are much factors at which the researcher must cope with in the methodology so that to take the more effective decisions possible to carry out the research.

Case

The case represents the sixth dimension. It should be noted that this dimension of an hypothetico-deductive scientific research is only appropriate for one type of research: case study research (quantitative, qualitative, or both). When the researcher is conducting a case study research, it is very important that he/she provides the reader with a good description of the case(s) studied so that the latter can better grasp the scope of the research.

Results

The next dimension has to do with the presentation of the research results. It is in this part of the research that the researcher gets the answer to his/her initial research question. Clearly, the more this procedure will be carefully performed, the more the answer to the research question will be precise and reliable. Overall, this procedure involves to analyse the data collected previously with some statistical and/or qualitative software, and to provide an appropriate, open, and fair interpretation of the results achieved.

Discussion

The eighth dimension of the diagram represented in Figure 1 is the discussion. Once the researcher has interpreted the research results, it is now relevant to have a good discussion about their impact in the scientific field concerned. In effect, this one aims to provide the reader with a more in-depth examination of some important aspects of the findings as well as a better view of both their effect and their scope at different levels of the field involved.

Conclusion

As for the ninth dimension, this one concerns the research conclusion. In this last part of the research, one could, for example, present to the reader a synthesis of both the results and the salient points of the research, as well as an anticipated view on the future of the scientific field involved. We must not forget that many readers read only the abstract, the introduction, and the conclusion of an article to get a quick overall view of its content. A good conclusion then becomes important. In short, it must be at the same time very concise and very consistent.

Research Paper Form

Finally, the tenth and last dimension relates to the research paper form. It should be noted that this dimension is not a part in itself of the research, but is rather related to the different ways to organize and write a research paper. The researcher can then include into it only relevant elements to the organization and writing of the article. The primary objective of a research paper is to communicate to other members of the scientific community how the researcher has conducted his/her research as well as the results achieved. Ideally, it must be logically organized, concise, clear, and well-written. In other words, it must be interesting and relatively easy to read for the reader. In this respect, as shown in Figure 1, I suggest three basic elements to integrate into this dimension, which reach, in my view, all of the most important aspects of a well-presented and well-written research paper: logical organization of the ideas, construction of the sentence structures, and orthographical quality. Yet another element one could also legitimately integrate into the research paper form is to make an attempt to get the discourse as exciting and living as possible.

Elements of an Hypothetico-Deductive Scientific Research

Introduction to the Object of Study

The first element of the diagram depicted in Figure 1 is the introduction to the object of study. So when the reader is beginning to read a research paper, the latter expects to know immediately the object of study or the general problem investigated by the researcher in order to assess the degree of relevance with his/her own research interests. To present the object of study in a scientific research, it could be relevant, for example, to answer this question: What is the problem to which I want to find a solution? In other respects, the diagram in Figure 1 shows that this element can be integrated into the abstract and the introduction of a research paper. Clearly, it is the same element, but it can be further discussed in the introduction than in the abstract so that the reader can better understand the problematic situation that requires an investigation on the part of the researcher. For example, in the introduction, it could be relevant to add the answer to the question: Why is it really important to treat this problem?

Description of the Research Objective

The description of the research objective represents the second element. As its name indicates, this very important element aims to inform the reader on the objective pursued by the researcher in his/her research. This description could be made, for example, in answering the following question: What I want to do or to show in this research? In other respects, as for the introduction to the object of study above, this element can be integrated in the abstract and the introduction of a research paper. Of course, it is the same element, but it can be further discussed in the introduction than in the abstract so that the reader can really see the work to be done through the same lens than the researcher. So to add to the understanding of the research objective in the introduction, one could, for example, answer this question: How can this objective be reached? Or this one: What is the scope of the research activities? Once the reader is well aware of the object of study and the research objective, then he/she can take a well-enlightened decision about whether he/she must going on in his/her reading.

Presentation of a Key Results Summary

The third element refers to the presentation of a key results summary. Such a summary provides the reader with an outline of the situation following the investigation. In this way, the reader having the same research interests can rapidly see whether these findings are either in the same or the opposite direction than previous research, or are shedding a new light on the object of study investigated by the researcher. And, on the other side, the reader having not the same research interests can then get an insight into what is happening in the field related to the object of study investigated by the researcher.

Presentation of the Research Paper Content

The fourth element of the diagram drawn in Figure 1 relates to the presentation of the research paper content. In fact, the goal of this element is merely to offer to the reader an outline of what is discussed in the research paper. One way to proceed could be, for example, to present each section that make up the latter. In short, at this point, the reader must be well aware of that he/she will learn in the paper.

Review of the Relevant Existing Theory

The review of the relevant existing theory is an extremely important element of a scientific research. In sum, it is essential to inquire the previous treatment of the object of study within its body of research. As Mace (1988) points out, the previous works are particularly useful when it is time to formulate the problem and to choose the verification strategy. By providing an orientation for what to look for, theory helps to determine which variables are relevant and which are not relevant (Pedhazur & Schmelkin, 1991). In addition, through problems and hypotheses derived from it, theory determines largely the type of research design, the analytic approach, and the results interpretation (Pedhazur & Schmelkin, 1991).

Formulation of the Research Problem

The next element of the diagram represented in Figure 1 has to do with the formulation of the research problem. Ideally, the review of the relevant existing theory on the object of study must allow the researcher to progress toward the formulation of a research problem. In fact, all scientific knowledge is fondamentally based upon a questioning process. According to Mace (1988), inscientificresearch, the only way to justify a work is to locate a gap in the previous works treating the same object of study and to fill this gap. This one becomes then the research problem. The interested reader can refer to Kerlinger (1986, pp. 16-17); the latter discusses three criteria to consider for the formulation of a research problem. We must not forget that an adequate expression of the research problem is one of the most important parts of the research (Kerlinger, 1986). Furthermore, a good part of the success or failure of the research effort is dependant on serious allowed to this initial step (Mace, 1988).

Formulation of the Research Question

The seventh element is the formulation of the research question from the research problem developed previously. The goal of this question is to guide and orient the research, to limit the area that will be covered by the researcher (d'Amboise & Audet, 1996). Therefore, it is important to ask the good question. In sum, we must formulate a relevant question, stated in clear and precise terms, and at which we will answer in taking into account our knowledge on the object of study and, above all, the available information [for the data collection] (Mace, 1988).

Development of the Theoretical Research Model and Definition of the Constructs and Variables

The eighth element concerns the representation, in form of model, of the theoretical solution allowing to solve the problematic situation and to define all of its components. Ideally, as stated before, the review of the relevant existing theory on the problematic situation allows us to discover constructs and variables that can be reused to provide a solution. We may also add our personal ideas. Thus, the function of the theoretical research model is to schematize all of these constructs, variables, and personal ideas in showing their relations (dependant, independant, mediator, or moderator). In addition, we can define them with existing literature and to bring into focus their role in the resolution of the research problem. On the other hand, it is important here to emphasize the fact that this element should be presented only after the data analysis in qualitative research and case study research. Indeed, as Daft (1995) argues, "In qualitative research, concepts and models should be defined at the end of the manuscript. The point of going out to observe organizations is to construct theory based upon the investigator's observations and interviews. The research goal is to end up with a well-defined set of constructs and a model that can be used to guide future research" (p. 174). Similarly, in the case study research, "The difference is that the construct, its definition, and measurement often emerge from the analysis process itself rather than being specified a priori" (Eisenhardt, 1989, p. 542).

Formulation of the Hypotheses or Propositions

The next element relates to the formulation of the hypotheses or propositions. Sekaran (1992) defines an hypothesis in a scientific research as following:

 An educated guess about a problem's solution ... a logically
 conjectured relationship between two or more variables expressed in
 form of testable statements. These relationships are conjectured on
 the basis of the network of associations established in the
 theoretical framework formulated for the research study (quoted in
 d'Amboise & Audet, 1996, p. 27).

According to Mace (1988), hypothesis is at the same time the result of the conceptualization and the starting point of the verification. It is the core foundation of all scientific work. Pedhazur and Schmelkin (1991) assess that "The guiding force of hypotheses in determining what to observe, what variables to relate, how to relate them, is undeniable" (p. 196). Kerlinger (1986) goes still more far while he argues that we must remember that there would be no science in any complete sense without hypotheses. In short, as for the research problem, hypothesis is a fundamental element of the research process. The interested reader can refer to Kerlinger (1986, p. 17); the latter discusses two essential criteria for the formulation of an hypothesis. On the other hand, in qualitative research, it is generally question of proposition rather than hypothesis. The researcher then will express his/her thoughts in form of research propositions, such the propositions playing the role of research hypotheses (d'Amboise & Audet, 1996). It should be noted that, as for the previous element, the formulation of the hypotheses or propositions should be presented only after the data analysis in qualitative research and case study research. In such the research, it is most of the time through the data analysis process that the themes, the constructs, the variables, and the relations between the variables, allowing the researcher to formulate hypotheses or propositions, begin to emerge.

Choice and Justification of the Research Design

The tenth element of the diagram depicted in Figure 1 is the choice and justification of the research design. In my view, to choose and justify a research design, it is to plan and describe the best way of investigation possible to get the answer to the research question while being at the same time the more rational possible for the necessary human, material, and financial resources. So it is a major element which can make all the difference in the success or failure of the research effort. In other respects, there is no consensus in the literature about the scope of a research design. For example, Kerlinger (1986) points out that "It includes an outline of what the investigator will do from writing the hypotheses and their operational implications to the final analysis of data" (p. 279). And Pedhazur and Schmelkin (1991) argue that the research design is used differently by different authors or researchers. "Some use it 'narrowly', almost synonymously with the term 'analysis', whereas others use it 'broadly' to refer to all aspects of the research, including measurement, sampling, setting, data collection, analysis, and theoretical formulations" (Pedhazur & Schmelkin, 1991, p. 211). Like the supporters of the last part of the Pedhazur and Schmelkin's quote above, I prefer to use the research design "broadly" in integrating it after the formulation of hypotheses or propositions, as shown in the basic configuration of the dimensions and elements of an hypothetico-deductive scientific research proposed in Figure 1. The interested reader can refer to Campbell and Stanley (1966, pp. 1-71), Contandriopoulos, Champagne, Potvin, Denis, and Boyle (1990, pp. 33-53), Kerlinger (1986, pp. 279-343), as well as Pedhazur and Schmelkin (1991, pp. 211-233 and pp. 250-317); these authors suggest a broad range of research design for experimental, quasi-experimental, and nonexperimental research, and discuss their internal and external validity.

Description of the Sample

The description of the sample represents the eleventh element. First, all research question defines a set of objects at which the research results should be applicable. This set can be more or less restricted, or more or less well-defined by the asked question (Contandriopoulos et al., 1990). It is, in fact, the target population. But the target population is most of the time too large to be studied in its whole. We must then choose a representative sample of the target population so that the research results can be the more generalized possible to this population. According to Kerlinger (1986), a representative sample has approximatively the same characteristics that the population relevant to the research question. "But we can never be sure; there is no guarantee" (Kerlinger, 1986, p. 111). It is therefore particularly important in a scientific research to carefully describe the selected sample so that the reader can well visualize the scope of the results on the target population. One could, for example, briefly introduce the target population and describe after in more detail the part of this population chosen for study.

Description of the Experimental Procedure

The next element of the diagram concerns the description of the experimental procedure. As its name indicates, this element is exclusively related to experimental research. And, more specifically, although this one might be included into a field experiment, it is rather associated to the laboratory experiment (see the different types of research addressed at the beginning of this section). The primary goal of this element is to provide the reader with a detailed description of the whole experimental procedure applied to the subjects of the research study. As Kerlinger (1986) says, "Research reports of laboratory experiments usually specify in detail how the manipulations were done and the means taken to control the environmental conditions under which they were done. By specifying exactly the conditions of the experiment, we reduce the risk that subjects may respond equivocally and thus introduce random variance into the experimental situation" (p. 367). As we can see in the last part of the Kerlinger's quote above, the latter emphasizes the importance to specify exactly the conditions of the experiment so that all of the subjects can perform this one in the same way to avoid as much as possible variance biases. Why make an attempt to avoid as much as possible variance between subjects in an experiment? I believe that the two most important reasons are the following: (1) so that the research results be representative and well-balanced with the experiment; and (2) so that the experimental procedure might be applied to other subjects, groups, situations, etc. and thus offer the possibility to increase the degree of generalizability (the external validity) of the results over time. Accordingly, not only it is very important to carefully describe the experimental procedure in the research paper, but it is essential, first, to well establish the conditions of the experiment so that all of the subjects can perform it in the same way. It is therefore a major element when the research conducted by the researcher is an experimental one. Taking into account that the subjects are already presented in the description of the sample above, one way to proceed to describe the experimental procedure could be, for example, to briefly present the environment where the experiment is done and specify after in more detail all of the conditions that it entails (material, training, tasks, tests, follow-up, etc.).

Choice and Justification of the Data Collection Methods

The thirteenth element has to do with the choice and justification of the data collection methods. It is another major step of the research process. So Mace (1988) stresses the fact that the quantity of information, its nature, and its degree of accessibility are as much conditions to the success or failure of the verification effort. Obviously, "[...] each approach has unique strengths and weaknesses, making it more or less suitable for studying certain phenomena, for specific purposes, in given settings, with specific resources, respondents, and the like" (Pedhazur & Schmelkin, 1991, p. 133). Consequently, it is the researcher's responsibility to choose the good approach and to justify this choice in accordance with the research context. Also, it is generally very interesting to combine two or more methods (quantitative, qualitative, or both) to get more details and thus enrich the information gathered on the object of study. The interested reader can refer to Gauthier (1992, pp. 251-514); the latter examines in-depth numerous data collection methods.

Description of the Data Collection

Although it is important to describe and justify the data collection methods, it is also important to describe the data collection itself. It is the fourteenth element. At this point, the reader expects to know how the data collection procedure took place and its results in order to better grasp the sense and value of the information gathered. For example, in a quantitative research using a questionnaire as data collection method, it would be interesting for the reader to know the number of distributed questionnaires, the respondents' status, the distribution time and means, the response rate, and so on. On the other hand, in a qualitative research using an interview as data collection method, it would be interesting for the reader to know the number of interviews, the interviewees' status, the time, place, and duration of the interviews, the language used, the number of interviewers, the way to grasp information (recordings, field notes, comments, etc.), and so on. Finally, when the research conducted is both quantitative and qualitative (the multimethod paradigm), one only has to bring together all of the components taken into account (among those suggested above, for example) in the description of the data collection.

Presentation of the Constructs Measurements

The fifteenth element refers to the presentation of the constructs measurements. Not only this step is considered as essential in the research process, but it is in the own researcher's interest to make it as carefully as possible given it is closely linked to the research results and to conclusions that can be drawn of them. Pedhazur and Schmelkin (1991) identify two major benefits of measurement: (1) it is appreciated when it is contrasted with alternative approaches to the description of or the differentiation among a set of objects with respect to a given aspect; and (2) it offers the possibility to apply the powerful tools of mathematics to the study of phenomena. Stevens (1951) defines construct measurement in the following terms:

 "In its broadest sense, measurement is the assignment of numerals
 to objects or events according to rules" (quoted in Kerlinger 1986,
 p. 391).

According to this definition, the point of the exercise is then to assign a quantitative sense to the constructs of the theoretical research model to confirm or infirm hypotheses. It is, in sum, to operationalize the constructs or the variables that make up the constructs on the basis of indicators and to attribute to each of them a scale of measurement (nominal, ordinal, interval, or ratio). The process is similar in qualitative research and case study research, that is, through constant comparison between data from diverse sources, the researcher builds evidence which measure the construct in each case (see Daft, 1995, p. 174; and Eisenhardt, 1989, pp. 541-542). On the other hand, constructs measurements should be presented only after the data analysis in qualitative research and case study research. As said before, it is most of the time through the data analysis process that the themes, the constructs, the variables, the relations between the variables, and the measures begin to emerge.

Choice and Justification of the Data Analysis Methods

The choice and justification of the data analysis methods represents the sixteenth element. Data analysis is a step of the primary importance in the research process. Hence, it is essential to choose the good analysis methods and to justify their choice with regard to reach the research objective. According to Contandriopoulos et al. (1990), when the researcher uses analysis techniques known and accepted as valid by the whole scientific community, he/she only has to briefly describe them. Otherwise, if he/she uses techniques less known, or known but in a less usual context, the researcher must further describe them. In this case, he/she should pay attention to the description so that this one not to be too much annoying and not to use a too much technical language. Ideally, the researcher must show the appropriateness of the selected analyses to answer the research question. The interested reader can refer to Creswell (1998, pp. 139-165), Miles and Huberman (1994, pp. 90-244), d'Amboise and Audet (1996, pp. 60-70), Hair, Anderson, Tatham, and Black (1995, pp. 78-670), Kerlinger (1986, pp. 125-276 and pp. 527-617), as well as Pedhazur and Schmelkin (1991, pp. 342-740); these authors examine in-depth numerous data analysis methods for qualitative (the first two quotes) and quantitative research (the last four quotes).

Description of the Case(s) Studied

The next element of the diagram depicted in Figure 1 has to do with the description of the case(s) studied. "The case study is a research strategy which focuses on understanding the dynamics present within single settings" (Eisenhardt, 1989, p. 534). "In brief, the case study allows an investigation to retain the holistic and meaningful characteristics of real-life events--such as individual lifecycles, organizational and managerial processes, neighborhood change, international relations, and the maturation of industries" (Yin, 1994, p. 3).

In a case study research, a clear and precise description of the case(s) studied allows the reader not only to better visualize the scope of the research, but also to better understand its results. Nevertheless, remember, there are some cases where anonymity is preferable. Yin (1994) emphasizes the two most important. "The most common rationale is that, when the case study has been on a controversial topic, anonymity serves to protect the real case and its real participants. A second reason is that the issuance of the final case report may affect the subsequent actions of those that were studied" (p. 143).

Constructs Validation

The constructs validation is the eighteenth element. It is an essential step to establish the reliability of the constructs measurements used by the researcher. According to Kerlinger (1986), it is probably the most important form of validity from the scientific research point of view. "Construct validation is concerned with validity of inferences about unobserved variables (the constructs) on the basis of observed variables (their presumed indicators)" (Pedhazur & Schmelkin, 1991, p. 52). In other respects, as for the constructs measurements, constructs validation should be presented only after the data analysis in qualitative research and case study research. In qualitative research, "To uncover the constructs, we use an iterative procedure--a succession of question-and-answer cycles--that entails examining a given set of cases and then refining or modifying those cases on the basis of subsequent ones (Huberman & Miles, 1994, p. 431). Traditionally, the resulting inferences are deemed 'valid' in the relaxed sense that they are probable, reasonable, or likely to be true" (Robinson, 1951, Znaniecki, 1934; quoted in Huberman & Miles, 1994, p. 431). Similarly, in the case study research "[...] researchers use multiple sources of evidence to build construct measures, which define the construct and distinguish it from other constructs. In effect, the researcher is attempting to establish construct validity" (Eisenhardt, 1989, p. 542). The interested reader can refer to Eisenhardt (1989, pp. 541-544), Huberman and Miles (1994, pp. 430-440), Kerlinger (1986, pp. 420-432), as well as Pedhazur and Schmelkin (1991, pp. 52-80); these authors describe several methods of constructs validation. Also, there are numerous software to help in this task, i.e., SPSS, SAS, PLS, LISREL, EQS, BMDP, QSR NVivo (NUD*IST Vivo), and ATLAS.ti.

Presentation of the Data Analysis

The nineteenth element concerns the presentation of the data analysis. Data analysis is a major step in the research process. It can make all the difference in a reliable or not reliable answer to the research question. Kerlinger (1986) suggests the following definition for the data analysis:

 "Analysis means the categorizing, ordering, manipulating, and
 summarizing of data to obtain answers to research questions. The
 purpose of analysis is to reduce data to intelligible and
 interpretable form so that the relations of research problems can
 be studied and tested" (p. 125).

On the other hand, we must be well aware that the data analysis is certainly one of the more difficult steps of the research process at the operational level (Mace, 1988). It is therefore essential to make this step as carefully as possible. Quantitative analysis is generally performed using statistical tools. On the other side, qualitative analysis can take different forms, for example grouping theme, pattern-matching, building explication, and content analysis. As for the constructs validation discussed previously, there are also numerous software to help us in the data analysis. But the point on which I want to stress here is the presentation of these analyses itself. In my view, it is essential in a scientific research to show all of the relevant analyses allowing to answer the research question, but no more so that the research paper not becomes too long because inappropriate analyses. Ideally, to present these analyses, one can use figures, tables, matrix, schemes, causal networks, graphics, and so on so that the reader can get the results in a visual way. For more details concerning different ways to analyze the data and present the results, the interested reader can refer to the same authors suggested above for the choice and justification of the data analysis methods.

Results Interpretation

Although it is very important to present all of the results leading to the answer to the research question, it is also very important to well interpret these results. It is the twentieth element of the diagram drawn in Figure 1. Results interpretation is another major step of the research process given it allows to better understand the meaning of the results and to provide a more complete answer to the research question. Thus, in his/her exercise of interpretation, the researcher will question himself/herself on the meaning of the results in the specific context of his/her research (d'Amboise & Audet, 1996). Overall, the exercise entails to carefully examine and describe all of the results presented previously in a visual way. If necessary, the researcher can also explain unexpected or outstanding results as well as the factors that can produced them. In addition, in a quantitative research, the latter can specify whether each hypothesis or proposition is confirmed or infirmed. As said earlier, in qualitative research and case study research, it is in this part of the research that the researcher should develop the research model, define the constructs and variables, formulate the hypotheses or propositions, and present the constructs measurements and validation. Finally, the researcher can indicate whether the research results answer or not his/her initial question.

Comparison of the Results with Existing Theory

The twenty-first element refers to the comparison of the research results with existing theory. To advance knowledge in a scientific field, it is essential to show the level of agreement or disagreement of the research results with those observed until now by other researchers on the same object of study. Obviously, we must take care to well support our discussion with the findings of the other researchers involved. As Kerlinger (1986) points out, "One compares the results and the inferences drawn from the data to theory and to other research results. One seeks the meaning and implications of research results within [existing theory], and their congruence or lack of congruence with the results of other researchers. More important, one compares results with the demands and expectations of theory" (p. 126). Similarly, Eisenhardt (1989) argues that "An essential feature of theory building is the comparison of the emergent concepts, theory, or hypotheses [or propositions] with the extant literature. This involves asking what is this similar to, what does it contradict, and why" (p. 544). In fact, is it not the ultimate aim of the scientist to discover better and yet always better theories able to overcome more and more hard tests (Popper, 1979)?

Indication of the Research Contribution to the Construction of Theoretical or Practical Knowledge in the Field Involved

The goal of the next element is to inform the reader on the research contribution to the construction of theoretical or practical knowledge in the scientific field involved. This can be briefly made using only one or two sentences and allows other researchers to quickly visualize the advancement of knowledge in the field. One could, for example, answer a question as this one: What this research brings to the field at the theoretical or practical level?

Identification of the Research Limits

The twenty-third element aims to identify the research limits. For example, to what extent the results can be generalized, the conceptual weaknesses, and the difficulties encountered by the researcher throughout the research. We must remember that all of these informations are very important for other researchers given they allow them, among other things, to be more aware of the scope of the research results and to benefit from this experience and thus decrease the risks to repeat the same errors in subsequent research. In sum, what really matters, at this point, is that the reader can clearly see all of the major aspects that limit the research and the scope of its findings, as well as what can be improved for future research.

Proposition of New Research Ideas

The proposition of new research ideas represents the twenty-fourth element of the diagram. When the researcher examined existing theory on the object of study, in different situations encountered throughout the research, in the results and conclusions drawn of them, and so on, the latter had numerous opportunities to have and/or identify new research ideas. It is then important that he/she communicates to other researchers these "fresh" ideas that can make the object of other very interesting and relevant research.

Recall of the Research Objective

The twenty-fifth element concerns the recall of the research objective. The goal here is to recall to the reader what the researcher would want to do in his/her research so that the former can see whether this objective has been reached or not. Clearly, the point of the exercise is not to repeat exactly what has been said on the research objective in the introduction of the article, but rather to briefly recall to the reader what the researcher would want to do.

Presentation of a Results Synthesis

At the end of a research paper, it is very relevant and interesting for the reader to get an overview of the findings. It is the twenty-sixth element of the diagram drawn in Figure 1. At the beginning of this results synthesis, one may also add some methodological elements such as the research design used, the number of participants, and the data collection and analysis methods. What really matters, at this point, is to provide the reader with a short summary of what the researcher has done in his/her research as well as the results achieved. Ideally, the research findings should be presented in more detail in the conclusion than in the abstract.

Openness Toward the Future of the Field Involved

The twenty-seventh element has to do with the openness toward the future of the field. In other words, the researcher makes an attempt to answer the question: What is likely to happen in the field in the long run? Thus, this particular element brings into focus the researcher's ability to act as a visionary. In sum, the researcher brings his/her personal view on some future tendencies or what it remains to be done in the scientific field concerned. Obviously, this view must be as realistic as possible, in the sense that it can be infered, in large measure, from the researcher's knowledge and experience in the field, as well as his/her research findings. As a result, it will be certainly very useful for the reader to have such an insight. This can be made using only a few sentences and allows other researchers to get a reflective view on the future of the field on the part of a colleague that was just actively working into it for a long time.

Logical Organization of the Ideas

As for the twenty-eighth element, this one refers to the logical organization of the ideas in the research paper. The logical organization of the ideas has to do with the macrostructure of the article. It is, in fact, to make sure that the different parts or microstructures of the article fit together into a coherent whole. As Daft (1995) says, "The theory has to be congruent with the method, the method with the results, the results with the discussion section, and all sections with each other" (p. 179). In other words, the research paper should flow logically in a straight line of thought without disgression. In other respects, one may also verify whether the style and tone used are appropriate, that is, a style and a tone that are not "amateur". According to Daft (1995), appropriate style and tone mean that the researcher masters very well what he/she is doing, that he/she avoids exaggeration, and that he/she is not justifying his/her research in criticizing the work of other researchers. These various aspects are really essential, in my view, to make a research paper interesting. Further, as Daft (1995) points out, "A good paper is extremely disciplined" (p. 170). In short, a paper with ill-organized ideas as well as amateur style and tone is not likely to attract the reader's attention.

Construction of the Sentences Structures and Orthographical Quality

Finally, the twenty-ninth and the thirtieth elements of the diagram depicted in Figure 1 are related to the construction of the sentence structures and the orthographical quality of the article. Clearly, a sentence that is relatively short, well-structured, and using appropriate words written without orthographical mistake will be certainly much more clear and easy to read for the reader than a ill-structured sentence written with orthographical mistakes. In brief, when we are writing a document, whether it is a research paper, a book, a doctoral thesis, a student's work, a business report, or others, we must always keep in mind that it is very unpleasant for the reader to read a ill-structured and/or ill-written document.

SOME GUIDELINES ON "HOW" AND "WHEN" INTEGRATE THE DIMENSIONS AND ELEMENTS INTO HYPOTHETICO-DEDUCTIVE SCIENTIFIC RESEARCH

In this section, I suggest to authors in organizational sciences some guidelines on "how" and "when" integrate the dimensions and elements discussed in the previous section into hypothetico-deductive scientific research throughout the development of four scenarios respectively related to field experiment, laboratory experiment, field study, and case study research, the core of the hypothetico-deductive scientific research, in fact. First, a scenario is described. Second, some dimensions and elements to include into the research in such a situation are proposed (the answer to the "when" question). Finally, some ways to integrate them into the research are suggested throughout relevant examples of published research papers (the answer to the "how" question). But before to begin to develop the scenarios, it should be noted that given, in my view, it is fundamental in all research paper to take care of its structure and writing, the research paper form dimension and the three elements that can be integrated into it (logical organization of the ideas, construction of the sentence structures, and orthographical quality) shown in Figure 1 occur implicitly into all of the scenarios that follow. Consequently, they are not taken into account in the scenarios.

Scenario 1

In the first scenario, you are a team of researchers that much like experimental research and you decided to conduct soon a theory-driven hypothetico-deductive field experiment to verify the impact of some information and communication technologies (ICT) on students' outcomes in distributed learning environments (taking for granted that an academic environment is also an organizational one). More specifically, you want to compare the outcomes of some groups of students using a certain ICT with other groups using a different ICT or not using an ICT.

What dimensions and elements should we include in such a situation?

I believe this situation entails that at least the following dimensions of the diagram represented in Figure 1 should be included in the research: the abstract, the introduction, the literature review, the theoretical approach, the methodology, the results, the discussion, as well as the conclusion. And, with the exception of the methodology in which the description of the experimental procedure is, in my view, rather optional (in effect, given that the experiment is not supposed to be of a great complexity in this situation, that is, just to compare groups of students using ICT in distributed learning environments, the description of this procedure can be made with those of the sample), all of the elements suggested for each of these dimensions in Figure 1 should be integrated into it.

How should we integrate these dimensions and elements in such a situation?

In a field experiment (A comparative study of distributed learning environments on learning outcomes) published in Information Systems Research (5) journal in 2002, which compares the learning outcomes of groups of students using two different group support systems (GSS) in distributed learning environments, Alavi, Marakas, and Yoo articulated their research as following. In the abstract, all of the elements shown in Figure 1 are present and addressed as proposed in this article. In the introduction, the authors have further discussed both the object of study and the research objective than in the abstract, as said before, but they have not presented the research paper content. Remember, it is important to present the research paper content so that the reader can have an outline of what is discussed in the article. The literature review is organized in form of background of collaborative distributed learning. The formulation of the research problem and question is included in the next section of the paper in which is also developed the research framework (or the theoretical approach as called in this article), that is, the theoretically supported hypotheses (in this case, there is no theoretical research model to schematize the constructs, however they are well-defined). Remember, when possible and relevant, it is important to graphically show to the reader the theoretical approach developed to answer the research question; this can lead the latter to a better understanding. The methodology describes only the executive development program studied by the authors. The subjects (or the sample as called in this paper) are presented in the next section. As the experimental procedure seems to be of a relatively high level of complexity in this research, it is then described in a specific section (following those of the subjects) in which the constructs measurements are also included. Although called differently (data analysis and results) than those in Figure 1 (results), the next dimension (or section) of the research paper integrates the same elements and these ones are addressed as indicated previously. In the last section, the results and their implications in the field are addressed. Overall, the authors discuss the same elements than those in Figure 1, however the discussion is rather organized around the implications of the findings in the field as well as their comparison with existing theory. The three other elements are quite slightly addressed. Finally, the article has not a formal conclusion. In fact, I think that, in the authors' mind, the last paragraph of the discussion is considered as a form of conclusion. Remember, no matter they are integrated at the end of the discussion or into the conclusion, the authors of a research paper should always provide the reader with a synthesis of the core research procedures and results, as well as an insight into what is coming in the field involved.

In another field experiment (Videoconferencing in distance education: A study of student perceptions in the lecture context) published in Innovations in Education and Training International (6) at the end of 1999, which compares the outcomes of groups of students using videoconferencing with other groups not using it, Fillion, Limayem, and Bouchard adopted practically the same research structure than those diagrammed in Figure 1. In the abstract, all of the elements shown in Figure 1 are present and discussed as proposed in this paper. In the introduction, the authors have further discussed both the object of study and the research objective than in the abstract, as mentioned earlier, but, as Alavi et al. (2002), they have not presented the research paper content. So the same comment made in the previous paragraph in this respect is also applicable here. The literature review of the article inquires the previous treatment of the object of study and leads to the formulation of the research problem and question as said before. In the next section, the theoretical approach is developed as suggested in this paper. Although quite differently organized, the methodology includes most of the elements shown in Figure 1. Indeed, the choice and justification of the research design, the choice and justification of the data collection methods, as well as the descriptions of the sample, the data collection, and the experimental procedure are presented in a same subsection of the methodology, which is called sample and data collection. The constructs measurements are presented in a second subsection of the methodology. On the other hand, the choice and justification of the data analysis methods are found in a subsection of the results called data analysis and results interpretation. Thus, with the exception that it also includes the choice and justification of the data analysis methods, the results dimension (or section) integrates the same elements and addresses them as indicated in this article. As for the discussion section, it is quite different. Called discussion and recommendations, this one compares the results with existing theory and provides students, professors, and educational institutions with a series of recommendations. The proposition of new research ideas is addressed in the conclusion of the paper. In addition, it includes the recall of the research objective and the presentation of a results synthesis. All of these elements are discussed as suggested in this article. In other respects, the identification of the research limits, the indication of the research contribution, as well as the openness toward the future of the field concerned are not present neither in the discussion nor in the conclusion. Remember, these elements are important to inform the reader on the scope of the research and its findings, and to provide him/her with an insight into what is coming in the field.

The interested reader can also refer to a field experiment conducted by Piccoli, Ahmad, and Ives (Web-based virtual learning environments: A research framework and a preliminary assessment of effectiveness in basic IT skills training) published in MIS Quarterly (7) at the end of 2001, which compares the outcomes of students in a traditional environment with those of other students in a virtual learning environment. And to those carried out by Webster and Hackley (Teaching effectiveness in technology-mediated distance learning) published in Academy of Management Journal (8) in 1997, which compares the outcomes and perceptions of the technology (videoconferencing) of students in different environments (face-to-face versus distance learning, full-motion video versus compressed video, etc.). As for the two field experiments reviewed above, in these ones, some dimensions and elements are called and/or articulated differently than suggested in Figure 1, but, overall, most of the dimensions and elements diagrammed in Figure 1 are present and addressed as indicated in this article.

Scenario 2

In the second scenario, you are a researcher who most of the time is part of a research team, but sometimes you also like to conduct your research in solo. You much like experimental research and you plan very seriously to carry out probably on the next month a theory-driven hypothetico-deductive laboratory experiment to investigate the effect of some independant variables (for example, decisional guidance, media used, and group cohesion) on the dependant ones (for example, students' outcomes, decision making, social presence, task participation, and group consensus) in making different comparisons of groups of students being exposed to an experimental procedure with other groups being not (as for the first scenario, it is taken for granted that an academic environment is also an organizational one).

What dimensions and elements should we include in such a situation?

I think this situation implies that at least the following dimensions of the diagram depicted in Figure 1 should be included in the research: the abstract, the introduction, the literature review, the theoretical approach, the methodology, the results, the discussion, as well as the conclusion. And all of the elements proposed for each of these dimensions in Figure 1 should be integrated into it.

How should we integrate these dimensions and elements in such a situation?

In a laboratory experiment (Providing decisional guidance for multicriteria decision making in groups) published in Information Systems Research (5) journal at the end of 2000, which compares groups of students using a group decision support system (GDSS) with decisional guidance with other groups using the same GDSS without decisional guidance, Limayem and DeSanctis articulated their research as following. In the abstract, all of the elements shown in Figure 1 are present and discussed as said previously. In the introduction, the authors have further discussed both the object of study and the research objective than in the abstract, as mentioned before, but they have not presented the research paper content. So the same comment made for the two previous articles examined above is also applicable here. The literature review inquires the previous treatment of the object of study and leads to the formulation of the research problem and question, as well as to the development of the theoretical research model. Called differently (implementing and testing the guidance concept) than those suggested in the diagram drawn in Figure 1 (methodology), the next dimension (or section) is also organized quite differently. Indeed, the authors included into this section the relevant theory leading to the implementation of the "guidance" concept as well as the hypotheses testing this one. They also integrated the descriptions of the sample, the data collection, and the experimental procedure, as well as the constructs measurements. In addition, the authors included into this section all of the elements of the results dimension and one of the discussion dimension shown in Figure 1, that is, the constructs validation, the presentation of the data analysis, the results interpretation, and the comparison of the results with existing theory. All of these elements are presented in several subsections so that, in the whole, the section is logically organized. Finally, the conclusion of the article includes the other elements of the discussion dimension and those of the conclusion dimension shown in Figure 1, and these ones are addressed in the same way as suggested in this paper, even in a better way.

The interested reader can also refer to a laboratory experiment conducted by Alavi (Computer-mediated collaborative learning: An empirical evaluation) published in MIS Quarterly (7) in 1994, which compares the outcomes of groups of students whose the collaborative learning is GDSS-supported with those of other groups whose the collaborative learning is non-GDSS supported. And to those carried out by Yoo and Alavi (Media and group cohesion: Relative influences on social presence, task participation, and group consensus) published in MIS Quarterly (7) at the end of 2001, which investigates the effect of the media and group cohesion on social presence, task participation, and group consensus of triads of students using audio conferencing versus other triads using desktop videoconferencing. As for the laboratory experiment examined above, in these ones, some dimensions and elements are called and/or organized differently than proposed in Figure 1, but, overall, most of the dimensions and elements diagrammed in Figure 1 are present and discussed as indicated previously.

Scenario 3

In the third scenario, you are a team of researchers that better like the field investigation. So since sometimes you and your colleague(s) are to organize a theory-driven field study in which you want to examine some aspects of the human communication both in face-to-face and with some media, as well as the relation between different factors that might have some influence on this communication and its outcomes (in this scenario, the research environment can be either academic or organizational).

What dimensions and elements should we include in such a situation?

In my view, this situation involves that at least the following dimensions of the diagram represented in Figure 1 should be included in the research: the abstract, the introduction, the literature review, the theoretical approach, the methodology, the results, the discussion, as well as the conclusion. And, with the exception of the methodology in which the description of the experimental procedure is not relevant in this case, all of the elements proposed for each of these dimensions in Figure 1 should be integrated into it.

How should we integrate these dimensions and elements in such a situation?

In a field study (Message equivocality, media selection, and manager performance: implications for information systems) published in MIS Quarterly (7) at the end of 1987, which examines the relationship between the content of managerial communication (middle- and upper-level) and media selection, Daft, Lengel, and Trevino organized their research as following. In the abstract, all of the elements proposed in Figure 1 are present and addressed as indicated previously. In the introduction, the authors have further discussed both the object of study and the research objective than in the abstract, as said earlier, but they have not presented the research paper content. So the same comment made for the previous articles reviewed is also applicable here. Really, several authors forget to provide the reader with an outline of what is discussed in the research paper when they write the introduction. Remember, it is important and it takes only a few lines. Although called differently (research problem) than those suggested in the diagram depicted in Figure 1 (literature review), the next dimension (or section) include the same elements, that is, a review of the relevant existing theory as well as the formulation of the research problem and question. Also called differently (theory development) than those shown in Figure 1 (theoretical approach), the next dimension (or section) discusses in-depth the theoretical foundations of the study. And the hypotheses are formulated in the next section. In the research method section (or the methodology as called in this article), with the exception of the experimental procedure (which is not relevant in a field study), the authors address all of the elements proposed in Figure 1 and as mentioned previously. The results are presented in a visual way and explained by the authors as indicated in this paper. Finally, in the next section called discussion and implications, they compare the results with existing theory and propose new research ideas, but the research limits and its contribution are not present. Remember, these two elements allow the reader to better visualize the scope of the research and its findings, as well as what it adds to the knowledge of the field. The research paper has no conclusion, but two of the three elements suggested in Figure 1 are integrated into the last paragraph of the discussion, that is, the recall of the research objective and the presentation of a results synthesis. Remember, it is also important to provide the reader with an insight into what is coming in the field involved.

The interested reader can also refer to a longitudinal field study carried out by Storck and Sproull (Through a glass darkly: What do people learn in videoconferences?) published in Human Communication Research (9) at the end of 1995, which examines the communication performance and quality of groups of students in a traditional environment or face-to-face versus other groups using videoconferencing. And to a field study conducted by Lengnick-Hall and Sanders (Designing effective learning systems for management education: Student roles, requisite variety, and practicing what we teach) published in Academy of Management Journal (8) in 1997, which investigates the relationship between some influencing factors, such as the transformation process and the student co-producer role, and the student products (personal effectiveness, management, and application of material) and reactions (satisfaction with results and process), also called the outcomes, in a high-variety communication system.

Scenario 4

In the fourth and last scenario, you are a team of two researchers that better like to investigate some particular cases. Also, you and your colleague are now ready to conduct a theory-driven case study research in which you want to examine the use and outcomes of computer-based systems (as for the previous scenario, the research environment can be either academic or organizational).

What dimensions and elements should we include in such a situation?

I believe this situation entails that at least the following dimensions of the diagram depicted in Figure 1 should be included in the research: the abstract, the introduction, the literature review, the theoretical approach, the methodology, the case, the results, the discussion, as well as the conclusion. And, with the exception of the methodology in which the description of the experimental procedure is not relevant here, all of the elements suggested for each of these dimensions in Figure 1 should be integrated into it.

How should we integrate these dimensions and elements in such a situation?

In a case study research (The information age confronts education: Case studies on electronic classrooms) published in Information Systems Research (5) journal at the beginning of 1993, which investigates the use and outcomes of computer-based instructional technology in the context of graduate business education, Leidner and Jarvenpaa articulated their research as following. In the abstract, all of the elements suggested in Figure 1 are present and addressed as indicated in this paper. In the introduction, the authors have further discussed both the object of study and the research objective than in the abstract, as mentioned earlier, but they have not presented the research paper content. So the same comment made for the previous articles reviewed is also applicable here. Curiously, none of the manuscripts examined in this section provides the reader with an outline of what is discussed into it at the end of its introduction. Should we removed this even so important element? Perhaps, but I do not think! For example, when an editor publishes a book that brings together a set of chapters or articles written by different authors, in the introduction of the book, he/she provides the reader with a valuable insight of what is addressed in each of these ones. The same holds in the case of a scientific journal or review. Further, after a quick verification in one of my pile of studies, it appears that numerous authors present their content in the introduction. The literature review, as called in the diagram drawn in Figure 1, is presented in two different sections, that is, a first section called information technology use in classrooms inquires the previous treatment of the object of study and a second called research questions develops the research problem and questions. In the research design and method section (or the methodology as called in this paper), with the exception of the experimental procedure (which is not relevant in a case study research), the authors discuss all of the elements suggested in Figure 1 and in the same way as proposed in this article, even in a better way. In addition, they describe the environment where each of the three cases are studied. The next section called results brings together the two ones proposed in Figure 1 (case and results) as well as an element of the discussion dimension, that is, the comparison of the results with existing theory. In brief, for each of the three cases studied, the authors describe the case, analyse the data, as well as discuss and compare the results with existing theory. Finally, in the next section called implications, future research, limitations, and conclusions, the authors address all of the other elements of the discussion shown in Figure 1 as well as those of the conclusion, and similarly as indicated in this article. More specifically, in a first subsection, they discuss the research implications and contribution, as well as the future research possibilities. It is also in this subsection that the authors develop a theoretical research model and formulate some assumptions to be tested in future research. We have therefore a good example here of what is suggested in this paper with regard to the fact that some elements should be presented only after the data analysis in qualitative research and case study research. Remember, in these two types of research, it is most of the time through the data analysis process that the themes, the constructs, the variables, the relations between the variables, and the measures begin to emerge. The next subsection addresses the research limits. As for the conclusion, this one is presented in the last subsection.

The interested reader can also refer to a case study research conducted by Goodman and Darr (Computer-aided systems and communities: Mechanisms for organizational learning in distributed environments) published in MIS Quarterly (7) at the end of 1998, which investigates the role of computer-based systems to enhance organizational learning in a formal electronic library and an informal community that uses a variety of communication technologies.

To summarize, we can see that, overall, the published research examples I chose to help me to suggest to authors in organizational sciences some guidelines on "how" and "when" integrate the dimensions and elements of the diagram depicted in Figure 1 into an hypothetico-deductive scientific research are very coherent with my own view. Clearly, some dimensions are called and/or articulated differently, and some elements are also called differently and/or integrated into different dimensions, but it is right (we just can call this "flexibility") and fit well with the view of an hypothetico-deductive scientific research I want to share in this article. In sum, it was exactly the goal of the exercise here to review different hypothetico-deductive research papers already published in some high-ranked scientific journals in order to verify whether the dimensions and elements included into them by their authors fit well with those I propose in this article. As a result, it seems they do so. Also, it is not surprising that these manuscripts had been published in high-ranked scientific journals. All are both theoretically and methodologically strong. And the theoretical principle stated previously concerning the logical organization of the ideas or the macrostructure of a research paper, that is, the theory has to be congruent with the method, the method with the results, the results with the discussion, and all sections with each other, is rigorously applied. In other words, each of these research papers flows logically in a straight line of thought without disgression. Hence we can conclude that they are assuredly "good" models to follow.

In the last section of the article, I bring my personal view about its theoretical and practical contribution for the organizational sciences community.

THEORETICAL AND PRACTICAL CONTRIBUTION OF THE PAPER

The synthesized extended literature review on the dimensions and elements that can be included into an hypothetico-deductive scientific research I carry out in this paper as well as the guidelines on "how" and "when" they should be integrated into it I suggest to authors allow us to see some very interesting theoretical and practical implications for knowledge production and dissemination in organizational sciences. Indeed, I think this paper can bring responses to some issues raised by professors Cummings, Frost, Taylor, Deetz, Nord, Staw, and Daft with respect to the actual publication system (see contextualization section).

From a theoretical perspective, the paper provides authors with a single very concise theoretical document that brings together and examines all of the actual dimensions and elements to conduct and describe hypothetico-deductive scientific research, whether they are quantitative, qualitative, or both. Further, it can even be used by editors and reviewers as a guide to evaluate hypothetico-deductive scientific research submitted by authors for publication. These different aspects can contribute: (1) to increase quality, relevance, and rigor of the works carried out by authors; (2) to decrease high workloads of editors and reviewers; (3) to progress toward a common understanding of the role of scientific knowledge in our societies, in the sense that to show rigor and structure to conduct and describe scientific research, as well as to focus on understanding organizational problems and provide answers in scientific journals are strongly urged; and (4) to promote innovation and creativity among authors, in the sense that they can add and/or modify, and/or replace some dimensions and/or elements of the research structure diagrammed in Figure 1, as well as bring new ideas to improve it (while remaining, of course, in a rigorous and structured research environment).

From a practical perspective, the paper brings a tool which can contribute to quality control. One knows how much essential is this aspect in production and dissemination of scientific knowledge. As Gibbons, Limoges, Nowotny, Schwartzman, Scott, and Trow (1994) point out, "[...] scientific and technological knowledge production systems depend heavily and inherently on quality control" (p. 65). In short, according to Gibbons et al. (1994),

 "What counts as knowledge is,..., to a large extent, determined by
 what scientists and technologists say shall count, and this
 involves, implicitly if not explicitly the norms governing the ways
 they produce knowledge. Not only do those claiming to produce
 scientific knowledge have to follow certain general methods, but
 they also must be trained in the appropriate procedures and
 techniques. To be funded, researchers must formulate the problems
 on which they want to work in specific ways recognizable to their
 colleagues, and they must be scrupulous in reporting their results
 to a community of their peers using prescribed modes of
 communication. Science is a highly structured set of activities
 involving a close interaction between technical and social norms"
 (p. 31).

On the basis of Gibbons et al.'s (1994) view of quality control above, one can anticipate that an increase in quality of the works carried out by authors should be likely translated by a proportional increase of their acceptance and publication rates in organizational sciences journals. Clearly, I do not think that there is only "one best way" to conduct and describe an hypothetico-deductive scientific research in organizational sciences and that what I suggest in this paper agree it perfectly. On the contrary, I just believe this paper can allow us to take "one step forward" not only toward a better quality of published works, but also toward a better quality of the works carried out by authors and submitted for publication. I even ask to editors, reviewers, and authors to think about the diagram drawn in Figure 1 and eventually propose new ideas, dimensions, and elements that can improve it.

CONCLUSION

The two main objectives of this paper were to carry out a synthesized extended literature review on the dimensions and elements of an hypothetico-deductive scientific research, and to propose to authors in organizational sciences some guidelines on "how" and "when" they should be integrated into it. As for all other sciences, research is fundamental in organizational sciences. It aims to the advancement of scientific knowledge in its different fields of activities. It is therefore our responsibility, as researchers, to looking for a constant improvement of the quality of our scientific research to continuously evolve toward a better understanding of the human and technological needs of our organizations, and thus promote their development and productivity. It is with this thought in mind that I wrote this article. Is my goal reached? I hope so! Finally, I just would like to recall to editors, reviewers, and authors that all suggestions allowing the enhancement of the hypothetico-deductive scientific research structure I suggest in Figure 1 in the future are welcome.

ACKNOWLEDGMENTS

I would like sincerely to thank professors Moez Limayem (City University of Hong Kong), Michel Audet, and Francois Bergeron (Laval University, Quebec), as well as the reviewers for their helpful comments and suggestions on earlier drafts of this paper. A grateful thanks so to the Fonds pour la Formation de Chercheurs et l'Aide a la Recherche (FCAR) for its financial contribution to this project.

REFERENCES

Alavi, M. (1994). Computer-mediated collaborative learning: An empirical evaluation. MIS Quarterly, 18(2), 159-174.

Alavi, M., Marakas, G.M. & Yoo, Y. (2002). A comparative study of distributed learning environments on learning outcomes. Information Systems Research, 13(4), 404-415.

Campbell, D.T. & Stanley, J.C. (1966). Experimental and quasi-experimental designs for research. Chicago: Rand McNally.

Checkland, P. (1981). Systems thinking, systems practice. Chichester: John Wiley & Sons. Contandriopoulos, A.-P., Champagne, F., Potvin, L., Denis, J.-L. & Boyle, P. (1990). Savoir preparer une recherche: la definir, la structurer, la financer. Montreal: Les Presses de l'Universite de Montreal.

Creswell, J.W. (1998). Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks, CA: Sage Publications Inc.

Cummings, L.L. & Frost, P.J. (1995a). Conceptual perspectives: Introduction. In L.L. Cummings & P. J. Frost (Eds.), Publishing in the organizational sciences (pp. 3-12), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Cummings, L.L. & Frost, P.J. (1995b). Relevance and rigor in publishing: Introduction. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp. 79-84), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Daft, R.L. (1995). Why I recommended that your manuscript be rejected and you can do about it. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the Organizational Sciences (pp. 164-182), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Daft, R.L., Lengel, R.H. & Trevino, L.K. (1987). Message equivocality, media selection, and manager performance: Implications for information systems. MIS Quarterly, 11(3), 355-366.

D'Amboise, G. & Audet, J. (1996). Le projet de recherche en administration: un guide general a sa preparation. Document inedit, Faculte des sciences de l'administration, Universite Laval, Quebec. Retrieved September 8, 2003, from http://www.fsa.ulaval.ca/personnel/damboisg/liv1/index.html.

Davis, D. & Cosenza, Q.M. (1993). Business research for decision making. Wadsworth.

Deetz, S.A. (1995). The social production of knowledge and the commercial artifact. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp. 44-63), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Eden, C. & Ackermann, F. (1998). Making strategy: The journey of strategic management. London: Sage Publications Inc.

Eisenhardt, K.M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532-550.

Fillion, G., Limayem, M. & Bouchard, L. (1999). Videoconferencing in distance education: A study of student perceptions in the lecture context. Innovations in Education and Training International (IETI), 36(4), 302-319.

Frost, P.J. & Taylor, R.N. (1995). Partisan perspective: A multiple-level interpretation of the manuscript review process in social sciences journals. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp. 13-43), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Gauthier, B. (1992). La recherche sociale: de la problematique a la collecte des donnees, 2eme edition. Quebec: Les Presses de l'Universite du Quebec. Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P. & Trow, M. (1994). The new production of knowledge: The dynamics of science and research in contemporary societies. London: Sage Publications Ltd. Goodman, P.S. & Darr, E.D. (1998). Computer-aided systems and communities: Mechanisms for organizational learning in distributed environments. MIS Quarterly, 22(4), 417-440.

Hair, J.F. Jr., Anderson, R.E., Tatham, R.L. & Black, W.C. (1995).Multivariate data analysis with readings, 4th edition. New Jersey: Prentice Hall.

Huberman, A.M. & Miles, M.B. (1994). Data management and analysis methods. In N.K. Denzin & Y.S. Lincoln (Eds.),

Handbook of qualitative research (pp.428-444). Thousand Oaks, CA: Sage Publications Inc.

Kerlinger, F.N. (1986). Foundations of behavioral research, 3rd edition. New York: Holt, Rinehart & Winston.

Leidner, D.E. & Jarvenpaa, S.L. (1993). The information age confronts education: Case studies on electronic classrooms. Information Systems Research, 4(1), 24-54.

Lengnick-Hall, C.A. & Sanders, M.M. (1997). Designing effective learning systems for management education: Student roles, requisite variety, and practicing what we teach. Academy of Management Journal, 40(6), 1334-1368.

Limayem, M. & DeSanctis, G. (2000). Providing decisional guidance for multicriteria decision making in groups. Information Systems Research, 11(4), 386-401.

Mace, G. (1988). Guide d'elaboration d'un projet de recherche. Quebec: Les Presses de l'Universite Laval.

Miles, M.B. & Huberman, A.M. (1994). Qualitative data analysis: An expanded sourcebook, 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Morton, A., Ackermann, F. & Belton, V. (2001). Technology-driven and model-driven approaches to group decision support: Focus, research philosophy, and key concepts. Research paper no 2001/1, Management Science, Theory, Method & Practice, Management Science Department, Strathclyde Business School, Glasgow, Scotland. Retrieved September 8, 2003, from http://www.managementscience.org/research/ab0101.asp.

Nord, W.R. (1995). Looking at ourselves as we look at others: An exploration of the publication system for organization research. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp. 64-78), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Pedhazur, E.J. & Schmelkin, L. (1991). Measurement, design, and analysis: An integrated approach. New Jersey: Lawrence Erlbaum.

Peffers, K. & Ya, T. (2003). Identifying and evaluating the universe of outlets for information systems research: Ranking the journals. The Journal of Information Technology Theory and Application (JITTA), 5(1), 63-84. Retrieved September 8, 2003, from http://jitta.org.

Piccoli, G., Ahmad, R.& Ives, B. (2001). Web-based virtual learning environments: A research framework and a preliminary assessment of effectiveness in basic IT skills training. MIS Quarterly, 25(4), 401-426.

Popper, K.R. (1979). Objective knowledge: An evolutionary approach, Revised edition (1st edition: 1972). Oxford: Oxford University Press.

Robinson, W.S. (1951). The logical structure of analytic induction. American Sociological Review, 16,812-818.

Rousseau, D.M. (1995). Publishing from a reviewer's perspective. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp.151-163), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Sekaran, U. (1992). Research methods for business: A skill building approach, 2nd edition. New York: John Wiley & Sons.

Staw, B.M. (1995). Repairs on the road to relevance and rigor: Some unexplored issues in publishing organizational research. In L.L. Cummings & P.J. Frost (Eds.), Publishing in the organizational sciences (pp. 85-97), 2nd edition. Thousand Oaks, CA: Sage Publications Inc.

Stevens, S.S. (1951). Mathematics, measurement, and psychophysics. In S.S. Stevens (Ed.), Handbook of experimental psychology (pp. 1-49). New York: John Wiley & Sons.

Storck, J. & Sproull, L. (1995). Through a glass darkly: What do people learn in videoconferences?. Human Communication Research. 22(2), 197-219.

Webster, J. & Hackley, P. (1997). Teaching effectiveness in technology-mediated distance learning. Academy of Management Journal, 40(6), 1282-1309.

Yin, R.K. (1994). Case study research: Design and methods. Applied Social Research Methods Series, Volume 5. Thousand Oaks, CA: Sage Publications Inc.

Yoo, Y. & Alavi, M. (2001). Media and group cohesion: relative influences on social presence, task participation, and group consensus. MIS Quarterly, 25(3), 371-390.

Znaniecki, R. (1934). The method of sociology. New York: Farrar & Rinehart.

ENDNOTES

(1) It should be noted that d'Amboise and Audet (1996), Contandriopoulos, Champagne, Potvin, Denis, and Boyle (1990), Gauthier (1992), as well as Mace (1988) are references to french scientific research books. So the french-english translation has therefore be made as rigorously and carefully as possible by the author of this article to keep authenticity and truthfulness of the theories of their authors.

(2) Data from various sources (databases, data warehouses, reports, census, business directories, stock exchange, etc.) serve as starting point for the research. The research results are infered from these entry data.

(3) Existing theory underlies the research. It allows to identify constructs and variables that can be reused to guide the research. These constructs and variables, and the links between them are usually illustrated in a logical model.

(4) A standardized domain model (organizational, decision making, financial, optimization, simulation, etc.), which may underly a methodology, is at the basis of the research. For example, at the organizational level, Checkland (1981) seeks to model organizations using the concept of a "soft system" and Eden and Ackermann (1998) focus on using cognitive or causal mapping (quoted in Morton, Ackermann & Belton, 2001).

(5) In the large-scale online survey conducted by Peffers and Ya, from october 2002 through january 2003, 1129 IS researchers ranked Information Systems Research journal first in the 10 top ranked journals, ranked by average weighted perceived value rating as outlets for IS research.

(6) Innovations in Education and Training International (now Innovations in Education and Teaching International) is essential reading for all practitioners and decision makers who want to stay informed about the developments in education and training. It is the official journal of the Staff and Educational Development Association. Retrieved September 8, 2003, from http://www.seda.ac.uk.

(7) In the large-scale online survey conducted by Peffers and Ya, from october 2002 through january 2003, 1129 IS researchers ranked MIS Quarterly journal second in the 10 top ranked journals, ranked by average weighted perceived value rating as outlets for IS research.

(8) In the large-scale online survey conducted by Peffers and Ya, from october 2002 through january 2003, 1129 IS researchers ranked Academy of Management Journal third in the 50 top ranked allied discipline research journals, ranked by average weighted perceived value rating as outlets for IS research.

(9) Human Communication Research is one of the official journals of the prestigious International Communication Association. Retrieved September 8, 2003, from http://www.icahdq.org/. It is a top-ranked communication studies journal and one of the top two journals in the field of human communication.

Gerard Fillion, Laval University