A proposed idiographic approach to the study of entrepreneurs.

Luthans, Fred ; Envick, Brooke R. ; Anderson, Robin D. 等

INTRODUCTION

Although an idiographic approach to research methodology has a long history in the behavioral sciences (Allport, 1937), has been suggested for the study of organizational behavior (Luthans & Davis, 1982), and has been applied to the study of leaders and managers (Hemphill, 1959; Mintzberg, 1973; Luthans & Lockwood, 1984; Luthans, Hodgetts & Rosenkrantz, 1988), its application to the study of entrepreneurs has been absent. Allport (1937) broadly stated that the idiographic approach attempts to understand a particular event in nature or society. As a research method, the idiographic approach is characterized by individual-centered and naturalistic environmental contexts, and by qualitatively-based direct observation data gathering techniques (Luthans & Martinko, 1987). The idiographic approach also takes an emic (an insider's, subject's definition of the research situation) perspective (Morey & Luthans, 1984). By contrast, the more popular nomothetic approach is almost a completely opposite methodology. Nomothetic research is characterized by group-centered and controlled environmental contexts and by quantitatively-based indirect measures such as questionnaires and interviews (Luthans & Martinko, 1987). Importantly, under the nomothetic approach, an etic perspective is taken whereby the researcher defines the situation and develops the research questions (Morey & Luthans, 1984). The focus in on indirect measurement and sophisticated statistical analysis that are testing predetermined hypotheses.

The nomothetic approach has by far been the most popular research methodology in the fields of organizational behavior, management, and most recently, entrepreneurship. This is because no other methodology is even considered or known about, and pragmatically it provides the most objective and convenient way of collecting and analyzing data. A nomothesis research perspective is based on the starting assumption that individuals are more similar than different (Marceil, 1977; Luthans & Davis, 1982). In other words, the goal of nomothesis is to find the "average person." This is scientifically appealing to researchers because error and variability can be accounted for in a group of subjects. However, this dominant nomothetic approach is not without some limitations. For example, one problem with a nomothetic approach is the overdependence on questionnaire gathered data. Over the years, questionnaires used in organizationally-based studies have been heavily criticized in terms of design, usability (Van Maanen, Dabbs & Faulkner, 1982), and, especially, reliability and validity (Schriesheim & Kerr, 1977; Schriesheim, Bannister & Money, 1979).

To date, research in entrepreneurship has been done almost entirely from the nomothetic approach. Yet, especially for better understanding the nature of entrepreneurs' behavior in today's organizations, the idiographic approach may be a viable alternative methodology. One of the main problems with the field of entrepreneurship has been the varying definitions. For example, Gartner (1988) stated the view that entrepreneurship was nothing more than the creation of organizations. This suggested that growing and managing the business were not a part of what an entrepreneur is all about. However, another definitional view is that an entrepreneur may start new ventures, but is also the major owner and manager of a business not employed elsewhere (Brockhaus, 1980). This definition does acknowledge the act of creating the venture, but also the importance of sustaining or managing the business. Other scholars have used similar comprehensive definitions that describe entrepreneurship as a process of creating or seizing a venture, and then instituting activities intended to ensure survival (Carland, Hoy, Boulton & Carland, 1984; Sexton & Bowman, 1985; Sandberg, 1986; Johnson, 1990). The importance of this latter definition is that it recognizes entrepreneurship as an interactive process, and that it takes place within existing organizations. The implication is that entrepreneurial activities do not end after a venture has been initiated. The entrepreneur must engage in behaviors beyond creating or seizing the venture in order for it to be successful. When this more comprehensive definition of the entrepreneur is assumed, then the ideographic approach to studying the behavior of entrepreneurs in their naturalistic(organizational) setting becomes particularly relevant.

In order for entrepreneurs to create or seize ventures, and then institute activities to ensure success and even survival, they must interact with their internal and external environment, both manipulating it and deriving some kind of meaning from it. This interactionist perspective is not new. Mischel (1973), a noted personality theorist, asserts that researchers should not focus on situation-free environments, but examine person-situation interactions in naturalistic settings. He further states that we should not only generalize about how different individuals are alike, but also identify and understand what they do behaviorally on an individual basis. Widely recognized social learning theorist Albert Bandura has argued for research that takes an interactionist perspective. He suggested that there is a dynamic, reciprocal interaction between the person, the environment, and the behavior itself (Bandura, 1977). Applying social learning theory to entrepreneurship, the person is the entrepreneur, the environment is both external and internal to the organization, and the behavior is that of the yet to be reliably and validly identified and measured behavior of entrepreneurs.

The dominant nomothetic approach may not be the most appropriate methodology for reliably and validly identifying entrepreneurial behaviors and activities in the interactionist, naturalistic setting in which today's entrepreneurs operate. What is proposed here is an alternative idiographic approach and multi behavior-multi rater (MBMR) method to the study of the behaviors and activities of entrepreneurs in today's organizations. After a brief review of the past and present nomothetically-based research methodologies commonly used, the background and specific details of an idiographic approach and MBMR method are presented.

THE DOMINANT NOMOTHETIC APPROACH

There are two major methods associated with the nomothetic approach to research in organizational behavior, management and entrepreneurship-- trait-surveys and behavioral-surveys. The following briefly presents the background, characteristics and critique of these two widely used methods of data collection.

THE TRAIT-SURVEY METHOD

The earliest studies in leadership research attempted to identify specific traits that clearly distinguished leaders from nonleaders. This trait approach examined such things as personality, intelligence, height, weight, age, and social characteristics. However, Stogdill (1974) reviewed over two hundred of these studies and found that no set of such traits clearly predicts leaders from nonleaders. Yet, a trait-survey method has persisted through the years. For example, recently, the Constructive Thinking Inventory was developed which measures traits and emotional stability and uses regression analyses to predict academic performance and leadership ratings of undergraduates at a military academy. Similar to earlier work, this study found that personality traits were not related to leadership (Atwater, 1992). However, Kirkpatrick and Locke (1991) suggest that although research shows that the possession of certain traits alone does not guarantee leadership success, there is evidence that effective leaders differ from other people in certain key respects.

In the 1800s, John Stuart Mill is most often credited with being the first to distinguish the entrepreneur from the manager. He claimed that the main difference was the disposition of entrepreneurs to take risks. Pioneering economic theorist Joseph Schumpeter (1934) was also one of the first to recognize that entrepreneurs were distinct individuals worthy of study. Until the mid-1950s, the most prevalent view of successful entrepreneurs was that they had specific traits such as stamina, ambition, willingness to make sacrifices, and an interest in risk-taking versus an interest in security (Ginzberg, 1955).

In addition to this early emphasis on entrepreneurial traits, through the years other characteristics of entrepreneurs also began receiving attention. For example, McClelland (1961) introduced the need for achievement as being an essential entrepreneurial characteristic. Glennon, Albright, and Owens (1966) proposed that the entrepreneur's background (e.g., family, education, experience) was of central importance in determining success. They also introduced 21 personality characteristics in the form of a questionnaire survey. The characteristics included such things as creativity, energy level, and tolerance for uncertainty, along with several others. Even more entrepreneurial traits were added to the literature in the 1970s and 1980s. For example, Borland (1974) suggested that internal locus of control was relatively important in determining entrepreneurial success.

Unfortunately, parallel to the leadership field, as this list of entrepreneurial traits grows and grows, there is no consensus among researchers as to which traits clearly distinguish entrepreneurs from nonentrepreneurs. For example, Gartner (1988) argued that a startling number of traits and characteristics have been attributed to the entrepreneur, and a "psychological profile" of the entrepreneur assembled from these studies would portray someone larger than life, full of contradictions, and conversely, someone so full of traits that he/she would have to be a sort of generic "Everyman". This statement depicts the deficiencies of the trait approach to entrepreneurship and the nomethetic approach to the study of traits.

The main advantage of the trait-survey method is that it acknowledges the importance of predispositions, which for understanding can not be ignored in the study of any individual, including entrepreneurs. The main disadvantage is the inconsistencies in finding a set of traits that distinguishes a particular group of people from other individuals. Simply repeating what has been done in the past may only add to the inconsistencies instead of adding value to what we already know. Therefore, alternative methodologies seem needed to further our knowledge of particular individuals such as entrepreneurs. This is especially true as we move to the study of entrepreneurs' behaviors instead of traits.

THE BEHAVIORAL-SURVEY METHOD

The other popular nomethetic method is the behavioral-survey which attempts to indirectly measure and identify behaviors. Questionnaires and interviews are used to gather the data on individuals or about others. For example, the widely recognized questionnaire called the Leader Behavior Description Questionnaire (LBDQ) was constructed by Ohio State researchers to measure leaders' behavior described by their subordinates (Hemphill & Coons, 1957). However, as Campbell (1977) and others point out, there is a considerable reliance upon recall and the perceptions of the respondents on such questionnaires. In other words, the perceptions of behavior and actual behavior may not be the same. Questionnaires have all types of limitations and bias (e.g., recency, primacy, halo, stereotyping) affecting their reliability and validity.

Another well known leadership behavior questionnaire was developed at the University of Michigan called the Survey of Organizations (Katz, Maccoby, & Morse, 1950). This behavioral survey, like the Ohio State LBDQ, is once again reliant upon the perceptions of only one source (subordinates) completing the questionnaire. Although these questionnaires did turn attention towards leadership behaviors rather than solely traits, they depend on indirect measurement and have questionable reliability and validity. They do not make a direct systematic observation of behaviors, nor do they take situational variables into account. An approach which does take the situation into account is the path-goal theory which attempts to explain the effects of four styles of leader behavior (supportive, directive, participative, and achievement-oriented) on subordinate satisfaction, acceptance of leadership, and effort-reward expectancies. The measures for the path-goal approach are borrowed from the LBDQ and again rely upon perception. Wofford and Liske (1993) recently conducted a meta-analysis on path-goal theory and found that potential situational and artifactual moderators exist. The results suggest that much of the research testing path-goal theory has been flawed because of the measures that have been used.

Similar to the leadership field, entrepreneurial research has almost solely depended upon indirect measures from questionnaires and interviews. Entrepreneurs' behaviors such as being innovative, aggressive, energetic, and goal oriented have been measured and identified through questionnaires and interviews. However, in recent years there has been increasing concern and attention given to the need to study the interactive nature of entrepreneurs in the natural settings. For example, Starr and Fondas (1992) applied organizational sociology to the work setting of entrepreneurs, specifically studying aspiring entrepreneurs. Their model addresses the aspiring entrepreneur's journey to becoming the company founder and identifies factors that may influence the transition from a preorganization to the formation of a new organization. Gartner, Bird, and Starr (1992) differentiate entrepreneurial from organizational behavior, defining entrepreneurial behavior as the creation of the firm. They further state that generating entrepreneurial theory will lie in deriving an understanding of both entrepreneurial and organizational behavior and on probing how connections between the two can be made. Reynolds (1991) showed how sociology can contribute to understanding entrepreneurs by looking at the interactive effects of the organization on the entrepreneur. Amit, Glosten, and Mullen (1990) specifically studied how decision-making and skills of entrepreneurs affected how they started their ventures.

This increased recognition of the complexity of entrepreneurs in organizations calls for multiple measures and alternative methodologies. The proposal made here is not that the trait and behavioral surveys be eliminated. Perceptions can and should continue to play an important role in the understanding of entrepreneurs. However, depending only on perceptions gained through questionnaires and interviews, especially when measuring entrepreneurs' behavior, is no longer sufficient. As Sechrest (1969) noted many years ago, surveys relying on perception provide the greatest amount of information regarding how people feel about various real-life phenomena and situations, but they are often poor predictors of actual behavior. The time is right in the development and research of the field of entrepreneurship for other measures, such as direct observation, and other methodologies and perspectives, such as an idiographic approach, to be used.

THE USE OF THE IDIOGRAPHIC METHOD IN MANAGEMENT RESEARCH

As an alternative to traditional nomothetic research and its dependence on trait and behavioral survey methods, the idiographic approach uses direct behavioral observation from single and multiple sources in the natural setting. The background of this idiographic approach as applied to the management field is summarized in the following sections.

THE BEHAVIORAL OBSERVATION METHOD (SINGLE SOURCE)

Direct observation-single source is one data gathering method under the idiographic approach. An example would be observational or diary studies of managers in order to determine classifications of how managers spend their time (see Mintzberg, 1973; McCall, Morrison, & Hannan, 1978; Yukl, 1981) or the frequency of their activities (Luthans & Lockwood, 1984; Luthans, Hodgetts & Rosenkrantz, 1988). Mintzberg (1973) used unstructured observational methods to derive five characteristics and ten managerial roles. One distinction to Mintzberg's study is that the managerial activities were derived during and after the data collection, as opposed to the predetermined categories used in nomothetic research. However, a limitation for external validity to the Mintzberg study was that he used only five CEOs. Another well known observational study of managerial activities was done by Kotter (1982). He subjectively observed, supplemented with questionnaires and interviews, fifteen general managers to derive six job demands on managers. Like the Mintzberg study, these six job demands were developed after the data collection. The Kotter study was not as rigorous as the Mintzberg study and once again the sample of managers was quite small to make any generalizations. To date, such observational studies of entrepreneurial behaviors have not been conducted.

THE MULTI BEHAVIOR-MULTI RATER (MBMR) METHOD

The multitrait-multimethod (MTMM) approach to data gathering and analysis is most closely associated with Campbell and Fiske (1959). Because the focus of the proposed research approach is on directly observable entrepreneurial behaviors, as opposed to traits, the term multi behavior is substituted for the better known multi trait terminology associated with Campbell and Fiske. Such a multi behavior-multimethod approach has been used by Luthans and colleagues (Luthans & Lockwood 1984; Luthans, Hodgetts & Rosenkrantz, 1988) in their study of "Real Managers". Also, the term multi rater is substituted for multi method here, because the focus is on identifying behaviors through the direct observation conducted by more than one rater. The intent of MBMR is not to combine data collection methods. Researchers who have used the multi trait-multi rater method include Campbell and Fiske (1959), Bescoe and Lawshe (1959) and Lawler (1967). The idiographic approach proposed here for research on entrepreneurs' behaviors is a combination of the two, the multi behavior-multi rater method or what we call MBMR.

Similar to Campbell and Fiske's MTMM, the basic logic behind the MBMR method deals mainly with convergent and discriminant validity.

The MBMR method is a validational process utilizing a matrix of intercorrelations among tests representing at least two behaviors, each measured by at least two raters. Measures of the same behavior should correlate higher with each other than they do with measures of different behaviors involving separate raters. Ideally, these validity values should also be higher than the correlations among different behaviors measured by the same rater (Campbell & Fiske, 1959).

Such an idiographic approach using an MBMR method has not been widely used among researchers due mainly to time and cost constraints. It is much easier to administer a questionnaire or conduct interviews than it is to directly observe and record behaviors in the natural setting. However, some researchers have used a MBMR method. The multi trait-multi rater method has been used by Lawler (1967) to measure managerial job performance. He states that his primary gain from using this approach was to develop a more sophisticated understanding of his criteria. He asserts that management jobs are often multidimensional and hard to define; thus, performance in such complex jobs are difficult to quantify and make objective. This description that Lawler used for managers would certainly apply to entrepreneurs even more. Kavanagh, MacKinney, and Wolins (1971) also used the multi trait multi rater approach for measuring managerial performance. They obtained performance-trait ratings on individuals from raters at different organizational levels.

The prototype for the proposed research in this article comes from the earlier Real Managers study (Luthans & Lockwood, 1984; Luthans, Hodgetts & Rosenkrantz, 1988). This Real Managers study used an idiographic approach and the multi behavior-multi rater approach to measure and analyze observable behavioral activities of managers. Specifically, the study used multiple observers and multiple categories of behaviors. Many managers were observed over the phases of the four-year study. The subjects were real managers in real organizations (thus the title of the study Real Managers). The driving research question of this study was, "What do managers really do?" Twelve categories of behavior were derived from free observation and then were reduced into four types of broad managerial activities -routine communication, traditional management, networking, and human resource management. It was further determined that the managers studied spent 29% of their time in routine communication, 32% in traditional management activities, 19% networking, and 20% conducting human resource management activities (Luthans, Hodgetts, & Rosenkrantz, 1988). Pinder (1988) noted that this study provided the most solid empirical underpinnings for our understanding of managerial behavior to date.

This comprehensive Real Managers study of directly observable behavioral activities of managers in their natural setting can serve as a prototype for how entrepreneurs can be studied. The idiographic approach and MBMR method can lead to the better understanding of what entrepreneurs really do in their day-to-day activities. This alternative idiographic methodology is not suggesting that the traditional nomothetic approach is wrong and no longer needed, but at this point the entrepreneurial field needs an in-depth analysis of what entrepreneurs really do through directly observable behaviors in the natural setting.

APPLYING THE IDIOGRAPHIC APPROACH AND METHODS TO THE STUDY OF ENTREPRENEURS

A number of years ago Brandt (1981) noted that there has been a lack of carefully conducted, rigorously designed, empirical studies of human functioning in natural settings. Unfortunately, this observation still holds true today for research in entrepreneurship. There are to date no studies that systematically, rigorously, and empirically identify and measure the observable behaviors of entrepreneurs in their natural settings. Since the first step to any basic science is naturalistic observation, there should be emphasis placed on the systematic observation of entrepreneurial behaviors.

The work of the Barker group in the 1960s, stands out as one of the most complete overall attempts at describing the full range of behaviors of people in small towns and in the institutions of those towns. He addressed the need for purely descriptive details about human behavioral patterns in all walks of life (Barker, 1968). Without such sufficient descriptive information, the wrong research questions are developed, inappropriate hypotheses are tested, and erroneous assumptions are made (Klein, 1968).

Traits, characteristics, or human qualities are relatively meaningless unless they can be anchored to some kind of denotable behavior (Brandt, 1981). As Ackerman (1954) points out, only describing someone as intelligent helps us little unless we know how the intelligence manifests itself in behavior. For example, one common trait assumed among entrepreneurs is creativity. This term really has little meaning unless specific behaviors can be tied to it. If you write the word creativity on a sheet of paper and have two researchers write down examples of entrepreneurial creativity, you will very likely end up with very different descriptions and meanings. One may also conclude that the two researchers are actually perceiving different traits. However, if you take the same two researchers and have them observe one entrepreneur over a period of time, they are very likely to end up with the same responses. This is because behaviors can be directly observed and measured. The researchers are not relying on perceptions and subjective interpretations, only on their ability to directly observe and reliably record.

SPECIFIC PROCEDURES TO FOLLOW

Following the Luthans et al. (1984, 1988) Real Managers study, the specific procedures in an idiographic study of entrepreneurs would include: (1) free, unstructured observation, (2) post-log interviews, (3) the Delphi technique, in order to determine the categories of entrepreneurs' behavior, and (4) structured observation to determine the frequencies of the entrepreneurs' behavior in the natural setting.

PHASE I: FREE, UNSTRUCTURED OBSERVATION. Trained student observers could be used to observe and record the behaviors of target entrepreneurs. Students of entrepreneurship can be used because as Nunnally (1978) has pointed out, observations will be more accurate if the observer has an understanding of the person and context of the situation. The observers would keep an observational log of the activities exhibited by the entrepreneurs during the observation period. To avoid errors and biases, the students should be given formal training on the systematic errors commonly encountered in observations (Campbell, 1958; Thornton & Zorich, 1980). The trainer would throughly discuss each error in particular and provide detailed, relevant examples of each. The trainees would also receive an example of what an observational log should look like, as well as an example of a representative observation schedule. Observation schedules should be representative of each day of the week and each hour of the day.

After training, the students could observe and record the behaviors of the target entrepreneurs one varied hour per day over a two-week period. Thus, each entrepreneur would be observed a total of ten hours. The sample of entrepreneurs used would be selected as randomly as possible. However, as Brandt (1981) points out, although subjects can be drawn randomly in laboratory studies, the real world is generally made up of individuals who have already been preselected into groups and activities on the basis of the characteristics the researcher is interested in studying. Although the researcher does not define the research situation (an emic perspective), general criteria for inclusion in the study can be set by the researcher. For example, the criterion for selection in this study might be that the entrepreneur has been the one to create or seize the venture, and must still be actively involved in the business operations. No other criteria need be set in order to capture the most representative group of entrepreneurs as possible. However, the researchers could ensure that the entrepreneurs come from a wide variety of entrepreneurial ventures, so that they are not all retailers, manufacturers, service oriented, and so forth. The ventures could also differ in size and sales volume. The entrepreneurs themselves could be diverse in terms of gender, age, and education. This diversity would increase the generalizability of the entrepreneurs' behavioral categories.

PHASE II: POST-LOG INTERVIEWS. The second phase of the idiographic study of entrepreneurs' behaviors could be conducted after the observational logs are complete. Each entrepreneur could be given the opportunity to rate how well the observational logs are typical of his/her normal daily activities. The target entrepreneurs could also be asked to describe and give examples of any behaviors they commonly engage in that were not captured by the observational logs. This phase would help ensure representativeness of the entrepreneurs' behavior, ensure that no important behaviors are left out, and also make certain that the observational logs had at least face validity for the entrepreneurs themselves.

PHASE III: DELPHI TECHNIQUE. The third phase in identifying the behavioral categories could be the use of the Delphi technique (Delbecq, Van de Ven & Gustafson, 1975; Helmer, 1983). The goal of this procedure would be to condense the multitude of behaviors observed and logged during the free observation into a comprehensive, yet conceptually sound set of entrepreneur behavioral categories. While Kerlinger (1979) does not address the Delphi technique specifically, he states that the analysis of observational data involves the categorizing, ordering, manipulating, and summarizing of data, with the purpose of reducing large quantities of raw data into a manageable and interpretable form. The members of the Delphi panel may best include researchers actively involved in the study as well as "naive" members with respect to the study and the entrepreneurship literature. No attempt is made to distinguish between any of the members of the panel during the process.

The Delphi process involves successive rounds of anonymous input, composite feedback, and iterations. All members of the Delphi panel would receive a copy of the complete unstructured behavioral logs obtained in phase one. For the first round, no set number of categories would be specified. In subsequent rounds based on the feedback, the panel members would collapse the categories to ensure comprehensive, mutually exclusive, but conceptually sound, categories. Final editing of the categories could be conducted by the researchers actively involved in the study.

PHASE IV: STRUCTURED OBSERVATION. The final phase of the idiographic approach to entrepreneurial research could be structured observation. The purpose of this phase would be to measure the identified entrepreneur behaviors through frequency counts. Specifically, the observers would receive a checklist with all of the behaviors that were identified by the Delphi panel from the unstructured observations. A check mark could simply be placed next to the behavior each time it occurs, or left blank if it does not occur. Each target entrepreneur in this stage could be observed at a random time each hour over a two week period for a total of 80 times. The actual number of observations may be lower depending on whether or not the entrepreneur was available to be observed.

Observers used in this measurement stage of the study would have maximum visual and audible contact with the entrepreneur, and also have a good understanding of the functions and nature of the activities of the entrepreneur. It is likely that secretaries or key subordinates would best serve as insider, participant observer sources. Students of entrepreneurship would be good candidates to serve as external, outside observer sources. The observers would be trained in the same manner as observers in phase one. The training could include role playing to ensure that the trainees know and understand each of the identified categories of behaviors.

Thornton and Zorich (1980) demonstrated that observer training significantly improved observer accuracy. All observers (regardless of source type) would be trained in the same manner, by the same trainers, and for the same amount of time. Observers would first be given a general explanation of the study. The observational checklist would then be distributed for a detailed explanation of each category of behavior. The trainees would be instructed on how to deal with various practical problems. Finally, role playing could be used in order for the observer trainees to practice and demonstrate that they are able to recognize each behavioral category accurately. Bandura (1977) contends that modeling, rehearsal, and repetition can increase observer accuracy; the intention of the role playing exercise.

USE OF THE MULTI BEHAVIOR-MULTI RATER (MBMR) METHOD

The use of any measurement should undergo reliability and validity analysis. The MBMR method can analyze both reliability and validity. Interrater agreement (Bijou, Peterson, & Ault, 1968) can be assessed to determine reliability. The agreement between raters should only differ minimally in order for the measure used to be considered reliable (Nunnally, 1978). Interrater reliability can be assessed by analyzing the behaviors observed by two or more raters of one entrepreneur during the same time frame. Cohen's (1960) kappa statistic could be calculated. This kappa statistic represents the proportions of joint judgments in which there is agreement, after chance agreement has been excluded.

The validity analysis would focus on whether the behavioral checklist is measuring what it is supposed to be measuring. The two major dimensions of the MBMR validity assessment would be convergent and discriminant validities. Convergent validity is the correlation between two different raters measuring the same behavior. Discriminant validity is the extent to which raters differentiate between different behaviors, as well as converge when observing the same behaviors.

Further understanding of the MBMR validity analysis can be illustrated in Table 1. As shown, this MBMR matrix has three different behaviors measured by three different raters. The "R"s represent correlations between the same behavior measured by the same rater. The "R"s set the upper limit for the matrix, because a rater of a behavior must correlate at least as highly with him/herself as he/she does with any other rater. The "C"s represent the degree of convergent validity. These "C"s measure the correlations of using different raters to measure the same behavior. The "M"s represent the degree to which correlations among the raters in the matrix are errors made by a particular rater. Therefore, the difference between the "C"s and the "M"s represents divergent validity. This is the extent to which raters differentiate between different behaviors, as well as converge when measuring the same behavior. And finally, the "H"s represent the correlations of one behavior with a different rater of another behavior. In other words, they have neither a common rater nor behavior. Therefore, the "H"s should be approximately zero.

Campbell and Fiske (1959) suggest that the entire matrix must be evaluated in order to assess construct validity. The "R"s determine reliability. The "C"s represent convergent validity, which is not enough to establish construct validity. Discriminant validity must also be determined. This is done in three ways. First the "C"s should be higher than the "H"s. Second, the "C"s should be higher than the "M"s. And finally, the extent of discriminant validity is indicated where similar patterns of behavior intercorrelations ("M"s) are found across raters.

[ILLUSTRATION OMITTED]

CONCLUSION

This article proposes that an idiographic approach using a multi behavior-multi rater method is a badly needed alternative methodology for researching the behaviors of entrepreneurs in their natural settings. Currently, the study of entrepreneurship depends almost solely on nomothetic approaches such as trait-survey and behavioral-survey methods. These methods generally assume group-centered, standardized, and controlled environmental contexts, and the use of quantitative analyses. This nomothetic approach is more appropriate under the assumption that people are more similar than different. However, assuming a more realistic interactionist perspective of practicing entrepreneurs in today's organizations would suggest the need for an idiographic approach with a multi behavior-multi rater (MBMR) method of measurement and analysis. This idiographic methodology would seem to be a desirable alternative to nomothetic methodology because entrepreneurs realistically interact with their environment and tend to be more individually than group centered.

The few idiographic studies in leadership and management can serve as an important prototype for researching entrepreneurs' behavior in the natural setting. Specifically, the previous Real Managers study conducted by Luthans et al. (1984, 1988) can provide specific procedures for identifying and directly measuring entrepreneurs' behavior. Some key steps might include: (1) unstructured observation, to qualitatively gather data on entrepreneurs' behavior through direct observation; (2) post-log interviews, to ensure all behaviors were observed, are representative, and have face validity; (3) the Delphi technique, to systematically determine categories of behaviors; and (4) structured observation to measure the frequency of the identified behaviors. Overall, we suggest that such an idiographic approach and the MBMR method may not only be overlooked, but is a badly needed alternative (not substitute) methodology for the better understanding of entrepreneurs in today's organizations.

REFERENCES

Ackerman, W.I. (1954). Teacher competence and pupil change. Harvard Educational Review, 24, 273-289.

Allport, G. (1937). Personality: A psychological interpretation. New York: Holt.

Amit, R., Glosten, L., & Mullen, E. (1990). Entrepreneurial abilities, venture investment, and risk sharing. Management Science, 36, 1232-1245.

Atwater, L. E. (1992). Beyond cognitive ability: Improving the prediction of performance. Journal of Business and Psychology, 7, 27-44.

Bandura, A. (1977). Social learning theory. Englewood Cliffs, N.J.: Prentice Hall.

Barker, R.G. (1968). Ecological psychology: Concepts and methods for studying the environment of human behavior. Standford, Cal.: Stanford University Press.

Bescoe, R.O., & Lawshe, C.H. (1959). Foreman leadership as perceived by superiors and subordinates. Personnel Psychology, 12, 573-582.

Bijou, S.W., Peterson, R.F., & Ault, M.H. (1968). A method to integrate descriptive and experimental field studies at the level of data and empirical concepts. Journal of Applied Behavioral Analysis, 1, 175-191.

Borland, C. (1974). Locus of control, need for achievement, and entrepreneurship. Unpublished doctoral dissertation University of Texas at Austin.

Brandt, R.M. (1981). Studying behavior in natural settings. New York: University Press of America.

Brockhaus, R.H. (1980). Risk taking propensity of entrepreneurs. Academy of Management Journal, 23, 509-520.

Campbell, D.T. (1958). Systematic error on the part of human links in communication systems. Information and Control, 1, 334-369.

Campbell, D. T. & Fiske, (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.

Campbell, J. P. (1977). The cutting edge of leadership: An overview. In J.G. Hunt & L.L. Larson (Eds). Leadership: the cutting edge. Carbondale, Ill.: Southern Illinois University Press.

Carland, J.W., Hoy, F., Boulton, W.R., & Carland, J.C. (1984). Differentiating entrepreneurs from small business owners: A conceptualization. Academy of Management Review, 9, 354-359.

Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37-46.

Delbecq, A.L., Van de Ven, A.M., & Gustafson, D.H. (1975). Group techniques for program planning. Glenview, Ill.: Scott, Foresman.

Gartner, (1988). Who is an entrepreneur? is the wrong question. American Journal of Small Business, Spring, 11-28.

Ginzberg, E. (1955). What makes an executive. New York: Columbia University Press.

Glennon, J.R., Albright, L.E., & Owens, W.A. (1966). A catalog of life history items.

Greensboro, NC: The Richardson Foundation.

Helmer, O. (1983). Looking forward: A guide to future research. Beverly Hills, CA: Sage Publications.

Hemphill, J.K., & Coons, A.E. (1957). Development of the leader behavior description questionnaire. In R.M. Stogdill & A.E. Coons (Eds.), Leader behavior: Its description and measurement. Columbus, Ohio: Bureau of Business Research, Ohio State University.

Hemphill, J.K. (1959). Job descriptions of executives. Harvard Business Review, 37(5), 55-67.

Johnson, B.R. (1990). Toward a multidimensional model of entrepreneurship: The case of achievement motivation and the entrepreneur. Entrepreneurship: Theory and Practice, Spring, 39-54.

Katz, D., Maccoby, N., & Morse, N. (1950). Productivity, supervision, and morale in an office situation. Ann Arbor, Michigan: Institute for Social Research.

Kavanaugh, M.J., MacKinney, A.C., & Wolins, L. (1971). Issues in managerial performance: Multitrait-multimethod analyses of ratings. Psychological Bulletin, 75, 34-49.

Kerlinger, F.N. (1979). Behavioral research: A conceptual approach. New York: Holt, Rinehart, and Winston.

Klein, D. (1968). The teen-age driver: A research paradigm. Traffic Quarterly.

Kirkpatrick, S., & Locke, E.A. (1991). Leadership: Do traits matter? Academy of Management Executive, 5, 48-60.

Kotter, J. (1982). The general managers. New York: Free Press.

Lawler, E.E. (1967). The multitrait-multirater approach to measuring managerial job performance. Journal of Applied Psychology, 51, 369-381.

Luthans, F., & Davis, T.R. (1982). An idiographic approach to organizational behavior research: The use of single case experimental designs and direct measures. Academy of Management Review, 7, 380-391.

Luthans, F., Hodgetts, R.M., & Rosenkrantz, S.A. (1988). Real Managers. Cambridge, Mass: Ballinger.

Luthans, F., & Lockwood, D.L. (1984). Toward an observational system for measuring leadership behavior in natural settings. In J.G. Hunt, D. Hosking, C. Schriesheim, & R. Stewart (Eds.). Leaders and Managers. New York: Permagon Press.

Luthans, F., & Martinko, M. (1987). Behavioral approaches to organizations. In C.L. Cooper & I.T. Robertson (Eds.). International review of industrial and organizational psychology. London: John Wiley Ltd.

Marceil, J.C. (1977). Implicit dimensions of idiography and nonothesis: A reformulation. American Psychologist, 32, 1046-1055.

McCall, M.W., Morrison, A.M., & Hannan, R.L. (1978). Studies of managerial work: Results and methods. Greensboro, N.C.: Center for Creative Leadership.

McClelland, D.C. (1961). The achieving society. Princeton, N.J.: Non Nordstrand.

Mintzberg, H. (1973). The nature of managerial work. New York: Harper and Row.

Mischel, W. (1973). Toward a cognitive reconceptualization of personality. Psychological Review, 80, 284-302.

Morey, N.C., & Luthans, F. (1984)., An emic perspective and ethnoscience methods for organizational research. Academy of Management Review, 9, 27-36.

Nunnally, J.C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.

Pinder, C.C. (1988). Book review: Real managers, Academy of Management Review, 13, 661663.

Reynolds, P.D. (1991). Sociology and entrepreneurship: Concepts and contributions. Entrepreneurship: Theory and Practice, 16, 47-70.

Sandberg, W.R. (1986). New venture performance: The role of strategy and industry structure. Lexington, MA: Lexington Books.

Schriesheim, C.A. & Kerr, S. (1977). Theories and measures of leadership: A critical appraisal of current and future directions. In J.G. Hunt & L.L. Larson (Eds.), Leadership: The cutting edge. Carbondale, Ill: Southern Illinois University Press.

Schriesheim, C.A., Bannister, B.D., & Money, W.H. (1979). Psychological properties of the LPC scale: An extension of Rice's review. Academy of Management Review, 4, 287-290.

Schumpeter, J.A. (1934). The theory of economic development. Cambridge, Mass: Harvard University Press.

Sechrest, L. (1969). Non reactive assessment of attitudes. In E.P. Willems & H.L. Raush (Eds.). Naturalistic viewpoints in psychological research. New York: Holt, Rinehart, & Winston.

Sexton, D.L., & Bowman, N. (1985). The entrepreneur: A capable executive and more. Journal of Business Venturing, 1(1), 129-140.

Starr, J.A., & Fondas, N. (1992). A model of entrepreneurial socialization and organization formation. Entrepreneurship: Theory and Practice, 17, 67-76.

Stogdill, R.M. (1974). Handbook of leadership. New York: Macmillan.

Thornton, G.E., & Zorich, S. (1980). Training to improve observer accuracy. Journal of Applied Psychology, 65, 351-354.

Van Maanen, J., Dabbs, J.M., & Faulkner, R.R. (1982). Variables of qualitative research. Beverly Hills, CA: Sage Publications.

Wofford, J.C., & Liske, Z. (1993). Path-goal theories of leadership: A meta-analysis. Journal of Management, 19, 857-876.

Yukl, G. (1981). Leadership in organizations. Englewood Cliffs, N.J.: Prentice-Hall.

Fred Luthans, University of Nebraska-Lincoln

Brooke R. Envick, University of Nebraska-Lincoln

Robin D. Anderson, University of Nebraska-Lincoln