[meteor_slideshow slideshow=”arp1″]
Field method paper – please re-write, add recent sources and get more output
Using field methods to study and code entrepreneurial behaviour
Evaluation of methods and introduction of complementary research devices to improve research robustness
Abstract
Entrepreneurship has been conceptualized in a number of ways in literature, common ground has still not been found. Looking for the essence of entrepreneurship, social scien-tists have embarked to study effects and reasons of entrepreneurial actions. Research on ‘true’ entrepreneurial behavior, however, is scarce. This is although researchers agree that investigating this field could help solidify entrepreneurship theory, its delimitation from management and lay groundwork for studies on entrepreneurial effectiveness and efficiency.
Recurring to Edmondson & McManus (2007), this paper proposes that a research design based on field methods, in particular methods of observation, can be used to study and subsequently code and analyze behavioral phenomena in entrepreneurship. Henry Mintzberg’s 1968 method to studying managerial behavior proves a valuable starting point, however an evaluation based on Yin (1998) and Gibbert et al. (2008) shows a clear lack of rigor in his proceeding. Consequently, the research design is updated and remedies as well as complimentary methods are introduced. These measures are integrated into a research approach which enables the generation of reliable and valid data on entrepreneurial behavior.
Table of Contents
Table of Contents ii
List of Abbreviations iii
1 Introduction 1
2 Conceptual foundations 3
2.1 The job of the entrepreneur as research focus 3
2.2 Methodological fit of systematic observation in entrepreneurial settings 4
2.3 Key concepts of scientific observation 6
2.4 Using observation to generate data 7
3 Evaluation of Mintzberg’s observation approach 8
3.1 Internal validity 8
3.2 Construct validity 9
3.3 External validity 10
3.4 Reliability 10
3.5 Attainments of Mintzberg’s work 11
4 Measures to improve rigor of systematic observation 12
4.1 Clear definitions, epistemological foundations as well as pattern matching to improve internal validity 12
4.2 Social learning theory as theoretical foundation for the research approach 13
4.3 Multi-rater methods, in particular Delphi processes to code behaviour 14
4.4 Methodological triangulation to enhance construct validity 15
4.5 Larger sample size and clear rationale for case selection to enhance external validity 16
4.6 Comprehensive and comprehensible documentation to improve reliability 17
4.7 Overview of approach to studying entrepreneurial behaviour with systematic observation 17
5 Conclusion 18
5.1 Summary 18
5.2 Directions for future research 19
References 20
List of Abbreviations
Abbreviation Full term
e.g. exempli gratia (for example)
et al. et alteri (and others)
i.e. id est (that means)
p. page
pp. pages
1 Introduction
The purpose of this paper is to analyze, evaluate and improve methodological rigor of field research for deployment in entrepreneurial settings. Field research is defined as “systematic studies that rely on the collection of […] data in real organizations” (Edmondson & McManus, 2007). This paper’s focus is predominantly on systematic observation techniques as well as complimentary research devices to collect original data and a scientific process to generate activity categories. These methods fit the exploration of a research gap in entrepreneurship particularly well (see Section 2.2). Systematic observation has been used to explore and analyse a variety of subject matters, such as educational, healthcare and military institutions as well as political and economic actors (Sulsky & Kline, 2007; Yukl, 2005).
In the academic discipline of management, Henry Mintzberg pioneered the deployment of direct systematic observation in his seminal PhD thesis on managerial behaviour “The manager at work – determining his activities, roles and programs by structured observation” (Mintzberg, 1968). Since then, researchers have repeatedly relied upon systematic observation as a method to study and understand managerial behaviour (Kotter, 1982; Kurke & Aldrich, 1983; Luthans, 1987; S. Tengblad, 2001a). Still, in the much younger discipline of entrepreneurship, the actual activities, tasks and behaviours of entrepreneurs have not yet been studied thoroughly, yielding a research gap which could be addressed by systematically analyzing data from systematic observation (Brown & Hanlon, 2004; Gartner, 1989; Schwehm, 2007).
Although incorporating typical advantages of case study research, such as the coverage of events in their natural context, in real time, and the generation of detailed, voluminous evidence (R. K. Yin, 1998), observation approaches have been criticized for a variety of reasons in the literature. Limitations of observant research relate typically to validity problems, lack of reliability checks, subjective coding methodology and a number of other conceptual problems (Brown & Hanlon, 2004; Hales, 1986; Martinko & Gardner, 1985). The underlying research question for this chiefly methodological paper is thus:
R: How can we address the limitations of observational field research while preserving its methodological advantages?
The attempt to answer this question in the realm of management and entrepreneurship has three main objectives. First, direct, systematic observation has been widely used to study managers but not to study entrepreneurs (Brown & Hanlon, 2004). Thus, identifying and addressing the methodological issues in the scientific observation of managers can bring about research designs which, could validate, repel or reveal important aspects of managerial or entrepreneurial behaviour (Gartner, 1989; Schwehm, 2007). Second, the paper aims to enhance rigor of direct behaviour observation and behaviour coding approaches by systematically identifying weaknesses and strengths to subsequently introduce improvement measures to advance the methodology for future fruitful use in the social sciences. Third, based on the evaluation and development of systematic observation methods, a step-by-step research framework shall be devised which can be applied to support bridging a concrete research gap in entrepreneurial setting.
After the introduction, Section II will lay the foundation for the remainder of the paper. To substantiate the identified research gap, current issues in entrepreneurship theory will be reviewed briefly. Thereafter, methodological fit of an approach, in particular systematic observation and subsequent activity coding procedures to address the research gap shall be assessed. Finally, observation approaches as deployed in field research in management will be introduced as scientific methodologies. Subsequent to the fundamental groundwork laid out in Section II, Section III will evaluate the rigor of studies relying on structured observation by deploying Yin’s (1994) and Gibbert et al.’s (2008) suggested validity and reliability criteria. The analysis shall exemplarily focus on Mintzberg’s 1973 case studies on managerial behaviour, since they incorporate common weaknesses but also strengths of systematic observation and behavioural coding (Brown & Hanlon, 2004). Based on identified methodological and conceptual shortcomings, Section IV will suggest incremental remedies and complementary research devices to advance research designs featuring observation as a method to generate valid and reliable data on entrepreneurial behaviour for research and practice. Section V concludes the paper by providing a summary as well as directions for future research.
2 Conceptual foundations
2.1 The job of the entrepreneur as research focus
Entrepreneurship, although being a relatively young discipline, has been conceptualized in a number of different ways over the last decades (Gartner, 2008). More recently, its definition in literature has been predominantly processual: According to Fueglistaller et al. (2008) Entrepreneurship is a process which is initiated and executed by individuals to identify, evaluate and exploit opportunities. The entrepreneur is hence an individual performing these processes, succeeding with new products or production methods in the market and establishing new economic structures (Blanchflower & Oswald, 1998; Gartner et al., 1994) .
Nonetheless, academic discourse about the terms and concepts entrepreneurship and entrepreneur has not settled (Fueglistaller et al., 2008; Gartner, 2001, 2008; Schwehm, 2007). Correspondingly, researchers have embarked in a variety of directions to explain and conceptualize entrepreneurship phenomena, which can be subsumized under the following three questions (Stevenson & Jarillo, 1990)
• What happens when entrepreneurs act? (Results of entrepreneurial action)
• Why do entrepreneurs act? (Predispositions of entrepreneurial action)
• How do entrepreneurs act? (Behaviours / actions of entrepreneurs)
The first two questions are investigated in the disciplines of economics and psy-chology / sociology respectively. The third question, however, is best to be examined in the field of business management research (Gartner, 1989; Timmons, 2002). Stevenson and Jarillo (1990) postulate that management researcher should consider what entrepreneurs do behaviourally and how they succeed at being entrepreneurs. Despite almost 20 years having past, to this day little has been published on this topic, recent research still claims for a focus on direct contemplation of entrepreneurs (Schwehm, 2007). Correspondingly, a comprehensive literature search has yielded a total of five publications on that topic, with only one relying on direct observation (Man et al., 2002; Crossley & Pittaway, 2000; Envick & Luthans, 1996; Good, 1993; Kilby, 1971). This picture has been confirmed by a meta-study (Brown & Hanlon, 2004). This indicates the study of entrepreneurial behaviour to be a potentially fruitful area of investigation.
Bridging the persisting research gap by studying the actual “entrepreneurs job” or his true behaviour, contributes to theory and practice in a number of ways: First, by com-paring or injecting knowledge on real entrepreneurial tasks, activities and behaviours to available definitions and concepts of entrepreneurship, these can be validated, rejected, enriched and/or refined. Second, the proposed study lays a foundation for research into cause-effect relationships of entrepreneurial behaviour, i.e. which behaviours promote success and which promote failure. Without scientifically investigating what exactly entrepreneurs do hypotheses regarding, for example, effective or ineffective entrepreneurial behaviour are difficult to formulate. Third, entrepreneurship education can benefit from knowing which specific tasks entrepreneurs have to perform on a day to day basis. This essentially refers to the issue as to ‘how can we educate and train managers [or entrepreneurs] if we don’t know what managers [entrepreneurs] do?’ (Mintzberg, 1990). Fourth, the study supports entrepreneurs to reflect on their behaviour and potentially identify more effective or efficient ways to work.
2.2 Methodological fit of systematic observation in entrepreneurial settings
As found above, entrepreneurship research can benefit from firsthand insights on en-trepreneurial behaviour, on what entrepreneurs do, what their tasks and activities are. Thus, there appears to be a possibility to generate insights from systematic observation with relevance for theory and practice. Still, the issues concerning methodological fit and scientific rigor of systematic observation in entrepreneurial settings remain. The fit shall be assessed recurring to Edmondson & McManus’ (2007) contingency framework on methodological fit in management field research.
Since there is little research and formal theorizing on the topic of the “entrepreneur’s job” it can be classified as nascent theory research. The below table provides an overview of the archetypical research design which can be used in nascent theory research. Clearly, most of the criteria coincide, indicating that (systematic) observation is to be at least part of a methodology to explore entrepreneurial behaviour (see Table 1). This claim is further substantiated if we consider the three exemplary nascent theory studies presented by Edmondson & McManus (2007). Each of the studies relies on observation as their primary method of data collection as well as iterative content analysis as approach in data analysis phase and are exploratory in nature. Hence, we can conclude that methodological fit between research question, state of previous research and data collection method is, at least partly, fulfilled.
Table 1: The study of entrepreneurial behaviour as nascent theory research
Suggested approach towards nascent theory research Approach to study (entrepre-neurial) behaviour with sys-tematic observation
Research questions Open-ended inquiry about a phenomenon of interest Open inquiry about the phe-nomenon of entrepreneurial behaviour
Type of data collected Qualitative, initially open-ended data that need to be interpreted for meaning Quantitative and qualitative, without prior interpretation
Illustrative methods for col-lecting data Interviews; observations; obtaining documents or other material from field sites relevant to the phenomena of interest Mainly systematic observation
Constructs and measures Typically new constructs, few formal measures Constructs derived from em-pirical data
Goal of data analyses Pattern identification Pattern identification in activi-ties of entrepreneurs
Data analysis methods Thematic content analysis coding for evidence of con-structs Activity coding
Theoretical contribution A suggestive theory, often an invitation for further work on the issue or set of issues opened up by the study Theory building (e.g. entrepre-neurial roles)
Source: Partly adopted from Edmondson & McManus, 2007, p. 1160.
Furthermore, behavioural theories such as social learning theory suggest that data on (overt) human behaviour needs to be collected by direct observation (Bandura, 2006). Self reports through diaries and interviews have shown to be overly limited in detail and biased from self perceptions (Bandura, 1985).
Lastly, in terms of suitability of observation in entrepreneurial settings, literature and previous studies provide no reason for the method to not be applicable to study entrepre-neurial behaviour (Florén & Tell, 2004). In contrary, managers are active in similar, some-times even indistinguishable environments as entrepreneurs (Envick & Luthans, 1996; Gartner, 2008). However, the appropriateness of systematic observation as a method to generate scientific insights in the field of entrepreneurship hinges not only on its fit and applicability but also upon its ability to produce rigorous results, i.e. results with high validity and reliability (Eisenhardt, 1989; Punch, 2005). This issue shall be addressed in Sections 3 and 4 through an evaluation and subsequent improvement of the methodology applied by Mintzberg and replicating studies.
2.3 Key concepts of scientific observation
Observation of individuals, groups or physical environments is one of the oldest methods which researchers relied upon to obtain data for scientific analysis (Zunker, 2004). It can be defined as a planned capture of perceivable events, during which the scientific observer takes on a passive, receptive deportment (Brown & Hanlon, 2004; R. K. Yin, 1994). The non-intrusive deportment of the observer is also a key differentiator from other methods of data generation such as interviews, diaries or experiments, since it tries to minimize verbal or other stimuli which could arouse unwanted reactions (Envick & Luthans, 1996; Stewart, 1989). According to Yin (1998), systematic observation is to be allotted under the umbrella of case study research. To evade conceptual confusion, an overview of the umbrella term of quantitative field methods and its main constituents can be found in Table 2.
Table 2: Definitions of key terms or scientific observation
Denomination Definition
Quantitative field methods Sociological devices to collect original data in real organizations, the data recorded is predominantly quantitative (e.g. frequency of events), but can also features qualitative attributes (Berg, 1998; Martin, 1982).Systematic observation Umbrella term for unstructured / open, structured as well as for all distinct forms of participant observation (Blumenfeld et al., 1993).Unstructured / open observa-tion Planned capture of perceivable events with focus on thick, narrative evidence. The data is not recorded in predefined categories / types but with description of each event in several dimensions (Berg, 1998; R. K. Yin, 1998).Structured observation Planned record of events in predefined categories. Focus on measuring time and counting events, characterized by a high level of predetermined structure (Saunders et al., 2003).2.4 Using observation to generate data
As stated in the introduction, Mintzberg used observation as a sociological device in his PhD thesis observing and, based on the collected data, classifying managerial behaviour (Mintzberg, 1968). This study challenged and ultimately changed the then prevailing conception of management (Stewart, 1989; S. Tengblad, 2001a). Mintzberg and his discipleship primarily relied on the following approach in generating their data: A trained researcher observes the manager doing his job for a predefined period of time (mostly one working week), taking notes of all events and activities. This documentation was done in a number of dimensions, primarily duration and description of activity, but also purpose, participants, as well as self initiation and place of the event (Martin, 1982). Only during and ex post this observation, however, the researcher is to develop a system to categorize the observed events or behaviours. The researchers coding should not be influenced by an a priori analysis of extant literature or his own past experience but “by the single event taking place before him” (Mintzberg, 1973). Furthermore, hypotheses are derived as well as theories are developed iteratively during or deductively after (most of) the study has taken place. Mintzberg was clear in his call for an unbiased, unfolding observational approach (Mintzberg, 1973). After coding the recorded events, a multidimensional quantitative analysis was performed to identify activity patterns, time allocation and communication models. Based on this, further (theoretical) deductions were made, in particular a role theory of management was developed (Mintzberg, 1968).
3 Evaluation of Mintzberg’s observation approach
The exemplary evaluation of Mintzberg’s research design relying on systematic observation and one researcher as the activity coder is based on Yin’s (1998) and Gibbert et al’s (2008) criteria as to what passes as a rigorous case study in the positivist tradition. Those criteria are internal validity, construct validity, external validity and reliability, which are in a hierarchical relationship to one another: “construct and internal validity [represent] a condition sine qua non for external validity” (Gibbert et al., 2008). It is thus of importance to satisfy all four validity and reliability criteria to generate rigorous results through case study research.
3.1 Internal validity
Internal or logical validity considers whether the researcher’s logical reasoning is compelling. Key question is whether he relies on strong theory to formulate a clear re-search framework and a data collection design (Gibbert et al., 2008; R. K. Yin, 1998). First, since Mintzberg’s research is exploratory and excludes the use of any theoretical framework, the internal validity can only be partly assessed. Still, it is striking that he does not rely on (behavioural) theory to devise and/or justify his research design, in particular data collection. Second, it remains unclear what exactly is studied. Is it the job of the manager, managerial work or managerial behaviour? These terms are used almost synonymously throughout extant studies, although referring to distinct phenomena. Third, the epistemological foundations are unclear, as the research question is ideographic whereas stated results and method are at times nomothetic. Because Mintz-berg fails to address these issues and to provide a clear research design derived from theory, the internal validity of his study (and thus also of replicating and similar studies) is questionable.
3.2 Construct validity
Construct validity refers to the extent to which a procedure leads to an accurate observation of reality, ultimately generating results which describe what the study intends to examine (Berg, 1998; Denzin & Lincoln, 1994). Mintzberg’s primary objective was to improve understanding of “the manager’s working processes” (Mintzberg, 1990). As described in Section 2, he relies mainly on a method he calls structured, or sometimes structural, observation. Insights are generated based on observing, and sometimes ad hoc interviewing managers in their natural environments. This proceeding generates the following problems with regard to construct validity.
First, Mintzberg observes the individuals and records the data by himself. He is the only one conducting this study with only selectively relying on the judgement of other (internal) observer. This approach is prone to observant bias and mistakes in perception. Still, this criticism has been found to be weak by some scholars since the data Mintzberg collected is relatively simple in nature (e.g. time, place, participants, and activity description) (Kurke & Aldrich, 1983; Tengblad, 2001a).
More gravely, after completion of the observation, he again individually and subjec-tively judges and codes the activities. As stated above, single-rater approaches are prone to individual biases and arbitrary judgements (Martinko & Gardner, 1985). Compelling criticism can be brought up on this aspect, since it is difficult to argue for objectivity or reliability of data interpretation based solely on Mintzberg’s analytics. A categorization process for human activities is highly complex and analytically challenging. Thus, if done by one researcher alone without seeking broader consensus, it might produce capricious results.
Method triangulation is performed, but only selectively, with some background data gathered through interviews in the beginnings and ad hoc questions during the observa-tion. Still, ex post observation, Mintzberg uses no ostensive measures to increase research validity: he appears to not check for the representativeness of the collected data by asking the subjects if they omitted typical procedures because of the presence of the researcher. Furthermore, ex post observational interviews should also scrutinize if there are other relevant, but unobservable activities, such as work done at home or while travelling. Finally, however, a clear chain of evidence is given by Mintzberg’s description of his proceeding, depicting that internal validity is at least partly addressed in his method-ology.
3.3 External validity
External validity describes the degree to which the results of a study can be generalized. Apparently, Mintzberg judges the external validity of his study to be high, since he generalized what he found with 5 managers (and referring to a few prior studies) for all managers to be true. He was, however, not alone with this generalization. Until now, popular and scientific articles refer to Mintzberg’s study as factual, although it has similarly been praised (Kurke & Aldrich, 1983; Stefan Tengblad, 2001b) and criticized strongly (Envick & Luthans, 1996; Luthans et al., 1988; Martinko & Gardner, 1985).
The generalizability of case studies is difficult to substantiate, since the positivist para-digm of hypothesis testing with inferential statistics cannot be fulfilled. This makes it difficult to develop normative inferences or suggestions for managers (e.g. on managerial effectiveness) based on the observations made (Punch, 2005). Still, there are measures which can be used to provide a foundation for analytical generalization of findings, such as the use of multiple cases, a clear rationale for case selection and a detailed description of the context (Eisenhardt, 1989; R. K. Yin, 1994). Mintzberg does rely on multiple cases and he carefully describes the context, still the rationale for case selection was not explicitly provided. Thus, although being certainly overrated in terms of generalizability in management literature, Mintzberg takes a number of measures to enhance external validity of his results.
3.4 Reliability
Finally, in terms of reliability, that is, transparency and replicability of research, Mintzberg scores high. He offers careful documentation of approach and execution of research as well as a comprehensive case study database. His study has been successfully replicated many times (Kurke & Aldrich, 1983; Stewart, 1989; Stefan Tengblad, 2001b).
To conclude the above evaluation, it should be noted, that the principal objective of this section is not to criticize or defame Mintzberg’s work, which constitutes an innova-tive and outstanding scientific effort at the time of its publishing. Rather, this paper aims to encourage and enable future fruitful research based on observation techniques. Therefore, the next subsection shall evaluate Mintzberg’s approach the light of overall scientific research and focus on the attainments of the study.
3.5 Attainments of Mintzberg’s work
Is systematic observation as applied by Mintzberg, because of its deficiencies, not a scientific method? Punch (2005) states that it is scientific “to collect data about the world, to build theories to explain the data, and then to test those theories against further data” (p. 8). He adds that there are no formal requirements towards the nature of this data, it may be qualitative or quantitative, statistically measurable or not. Thus, despite its limitations, systematic observation can be considered a scientific method, since it collects data about the world and builds theory. The particular value of this method lies in its explorative nature which has elucidated important aspects about managerial work that were neglected in prior literature. It further evades the prevailing self reporting bias found in interviews or diaries (Berg, 1998; Envick & Luthans, 1996). Also, structured observation allows for an observation of the subject in their job settings, with all noise factors present. The method can hence reveal a contingent, ecological side of organizational activity (Schwehm, 2007). Finally, structured observation is sensitive to very brief tasks which has proven to be helpful in measuring rhythmic phenomena in the routine role of a manager (Martin, 1982).
Moreover, despite lack of rigour in terms of formal validity and reliability in his pro-ceeding, Mintzberg’s results have been confirmed in a number of studies, reported for example by Kurke & Aldrich’s (1983) study Mintzberg was right!, Tengblad (2001) and Stewart (1993). These authors were astonished by the amount of similarities of managerial behaviour between the studies and the robustness of Mintzberg’s propositions. It seems, however, difficult to perform direct observation of managerial behaviour which provides highly insightful and at the same time statistically generalizable data. This apparent challenge as well as the other breaches of rigorous research in Mintzberg’s and his followers’ approaches shall be addressed in the next Section.
4 Measures to improve rigor of systematic observation
In the following, concrete remedies to the above methodological limitations shall be introduced. The objective of this section is to devise a framework for rigorous deployment of systematic observation to collect primary data.
4.1 Clear definitions, epistemological foundations as well as pattern matching to improve internal validity
The research subject of study needs to be clearly defined to be able to align research design (Eisenhardt, 1989; Punch, 2005). In entrepreneurial settings this could for example be entrepreneurial behaviour, the job of the entrepreneur or entrepreneurial work. The nuances between such definitions can have great impact on the data collection method for the study (Stewart, 1989). Also, underlying behavioural theory is to be explicitly classical or methodological to enable ensuing construct validity (Bandura, 2006).
Case study research is predominantly ideographic, it focuses on the in-depth under-standing of one or a few cases (Envick & Luthans, 1996). Further to this, the method of systematic observation is used to observe an individual, an entrepreneur, which is defined to be fundamentally ideographic (Windelband, 1894). In consequence, the researcher needs to be careful not to deploy nomothetic methods, inferences or to over generalize findings.
Lastly, based on existing findings on managerial behaviour in small and medium en-terprises, pattern matching should be introduced (Gibbert et al., 2008). In this, the findings with regard to entrepreneurial behaviour should be compared to prior findings and/or theoretically predicted findings to help improve internal validity of the study. Still, this is to be done separately and ex post to the initial analysis to not compromise the unbiased, exploratory research approach.
4.2 Social learning theory as theoretical foundation for the research approach
There is a clear need for theory guiding data collection in structured observation (e.g. Stewart, 1989). More recently, ideographic, observational research in psychology takes an interactionist perspective based on social learning theory SLT (Envick & Luthans, 1996). SLT offers a potential framework for data collection design, data interpretation and evaluation (Bandura, 2006; Envick & Luthans, 1996). . According to Bandura (1977), the theory proposes that human behavior is driven by a reciprocal interaction among behavior, environment and cognitive-personal influences (p. 10): “[…] internal personal factors and behavior […] operate as reciprocal determinants [because] people’s expectations influence how they behave and the outcomes of their behaviour change their expectations” (Bandura, 1977, p. 195).
The SLT contingency approach basically analyses the interactive nature of the situa-tion (S), the organism (O), the behaviour (B) and the consequence (C). It allows for modelling and functional analysis of environmental-cognitive-behavioural events. The operationalization of the framework in behavioural research can be performed by defining and measuring variables depicting the qualities of the SOBC (Ginter & White, 1982). For entrepreneurship this implies that internal and external situational variables are to be measured (S), the motives, values, cognitive abilities of entrepreneurs are to be assessed (E), behaviours should be observed (B) and consequences recoded (C) (see figure I).
Figure I.
Through the above adaptation of SLT to entrepreneurship, three measures can be taken. First, the SLT depicts clearly which variables should be measured to explore and explain behaviour. Those comprise variables of the environment, psychological and socio-logical variable regarding the entrepreneur and entrepreneurial behaviour as such. Second,, SLT can be used to make predictions regarding differences and similarities of behaviour between groups of individuals. And lastly, it can be relied upon to control for variables driving behaviour in order to generalize smaller sample studies to their respective populations.
4.3 Multi-rater methods, in particular Delphi processes to code behaviour
To enable accurate recording of all perceivable events, the observation should be per-formed by a trained, knowledgeable researcher (Nunally, 1978). Ideally, the researcher has already practical experience in systematic observation procedures.
The unstructured and structured observation of behaviours should be performed by two or more researchers to diminish observant bias (R. K. Yin, 1998). Furthermore, multiple trained observers could be relied upon to study one entrepreneur at different times to further reduce individual biases and errors. Based on the collected multi-rater data, interrater reliability measures such as Cohen’s Kappa can be calculated (Cohen, 1960). Such statistical reliability measures are to be relied upon when building a dependable database for ensuing analysis.
Further to this, the coding of recorded behaviours should rely on content analysis using Delphi procedures (Weber, 1985). This addresses a variety of reliability and validity issues of Mintzberg’s work. In accordance with Linstone & Turoff (1975), this procedure should rely on a panel of topic experts and / or involved researchers first identifying behavioural patterns from the original data and deciding on an angle from which to analyze the data (e.g. functional, activity centred or behavioural). This panel is subsequently going through several rounds of classification and anonymous feedback. This allows for collaborative deduction of activity codifications with high validity based on the anonymously achieved consensus of the experts (Linstone & Turoff, 1975).
A Delphi process can also to be used for iterative theory building after categorization is completed and quantitative analyses have been performed, e.g. to identify entrepreneurial roles (Häder, 2002). In this context, more informal peer reviewing could also be relied upon (Gibbert et al., 2008).
4.4 Methodological triangulation to enhance construct validity
Method triangulation, in this case referring to the introduction of complementary re-search measures, should furthermore be used to improve construct validity (R. K. Yin, 1994). This comprises semi-unstructured, pre-study interviews with the subjects and close co-workers to gather information on the environment and the entrepreneurs (R. K. Yin, 1998). Also, the entrepreneurs should be questioned about the representativeness of the observed period of time ex ante and ex post to the field observation (Stewart, 1989). It is furthermore recommendable to rely on structured, post-study Likert-scale interviews with the subjects and close co-workers to ensure accurate, representative capture of events (Berg, 1998).
Finally, complementary observation methods such as diaries, behavioural interviews and activity samples could be used to generate additional data on managerial behaviour, further validate results or enable longitudinal studies. This would also address the some-times criticised mechanistic oversimplicity of the observation method (Martinko & Gardner, 1985).
4.5 Larger sample size and clear rationale for case selection to enhance external validity
External validity can be improved by relying on a larger sample of observed entrepre-neurs, allowing for inferential statistics to be deployed. To contain costs and limit time needed to study entrepreneurs, multiple researchers should work on a project. Also, third parties (e.g. students) could be specifically trained to capture information for the project. This proceeding requires, however, a use of structured observation with predefined activity or event categories being counted / measured. If larger sample data is unavailable or costs are too high, behavioural modelling and controlling for environmental and / or sociological variables via SLT could be relied upon.
Further, complementary or in lieu of large sample data, case selection should be done based on a clear rationale to produce an accurate picture of the phenomenon under investigation, an approach Eisenhardt (1989) denominates theoretical sampling (Gibbert et al., 2008). In entrepreneurship, this could for example mean the confinement of the study to a specific market (since a restaurant owner clearly has differing tasks and objectives from a biotech entrepreneur), selecting companies at a particular stage of the lifecycle or companies below / above a specific size.
At last, the collection and report of ample details on the case study context is recom-mended (Cook & Campbell, 1979).
4.6 Comprehensive and comprehensible documentation to improve reliability
Finally, the replicability of the research design should be ensured through careful documentation and the building of a replicable and/or easily expandable database. Key requirement is that the database needs to contain a narrative documentation organizing and citing all files and materials collected during the fieldwork (R. K. Yin, 1998). This procedure could be facilitated through the use of software, such as Microsoft Excel, SPSS or STATA for quantitative analysis and Atlas.ti for qualitative, theory building analyses (Muhr, 1996).
4.7 Overview of approach to studying entrepreneurial behaviour with systematic observation
Although some limitations of Mintzberg’s approach remain, such as the difficulty to capture cognitive processes, most of the initially identified weaknesses can be addressed. In the figure below, a step-by-step approach to study entrepreneurial behaviour comprising the above presented remedies shall be introduced. If followed, this research approach should enable the generation of sound evidence on entrepreneurial behaviour.
The process can be divided in three phases:
I. Preparation phase, in which mainly focuses on the researcher acquiring back-ground knowledge and piloting of the methodology,
II. Data collection phase, in which data is gathered through unstructured and structured observation
III. Analysis phase, in which results are coded, analyzed, enhanced and interpreted
Figure 2: Data capture process
Source: Partly adopted from Eisenhardt, 1989.
It should, however, be noted that the three phases of the study are not clearly delimited but should rather be overlapping (Eisenhardt, 1989). In particular the observation and consolidation phase do partly cover, since the coding and theory developing procedures should be iteratively conducted.
5 Conclusion
5.1 Summary
At the outset of this study, I documented a lack of research on entrepreneurial behaviour. This research gap can, in accordance with the methodological fit criteria developed by Edmondson and McManus (2007), be bridged through case study research relying on systematic observation and complimentary methods. Still, research based on observation has been criticised in the past for lack of methodological rigour. The consequent research question was
R: Is it possible to address the limitations of systematic observation, while preserving its methodological advantages?
The ensuing analysis focused on eliciting the drawbacks of past observational ap-proaches to develop remedies for current and future research. Based on Yin’s (1998) and Gibbert et al’s (2007) criteria for rigorous case study research, Mintzberg’s methodology to study of managerial behaviour was evaluated to identify key weaknesses. Major issues arise from lack of construct and external validity, impeding generalizability and reliability of findings.
Subsequent to this analysis, measures to improve validity and reliability of re-search relying on observation were introduced. Those measures comprise the elaboration of a clear research design based on behavioural theory and explicit epistemological assumptions, a multi-rater, multi-method data collection, potentially a larger number of cases and a rationale for case study selection. Clearly, a problem in generating those data is the amount of time required to observe a large (statistically significant) number of entrepreneurs. The use of behavioural modelling and/or theoretic sampling can be used to at least partly remedy this issue.. Ultimately, Delphi and similar peer review processes enable reliable iterative activity identification and coding and sound theory building. The identified improvements and complimentary research devices were thereafter integrated into a condensed step-by-step manual for (behavioural) research in entrepreneurial, and potentially also other, settings.
Consequently, this paper has contributed to literature by identifying a gap in entrepreneurship research and advancing a methodology to approach it.
5.2 Directions for future research
There are a number of potentially fruitful directions for future research arising from this analysis. First, the above framework should be deployed to study entrepreneurial behaviour. Second, the use of behavioural theories is still poorly developed in entrepreneurship and could be expanded (Schwehm, 2007). Third, to derive normative conclusions for managerial work, researchers could include performance variables in their studies of behaviour (Brown & Hanlon, 2004).
References
Atteslander, P. (1995). Methoden der empirischen sozialforschung (Vol. 8). Berlin: Sam-mlung Göschen de Gruyter.
Bandura, A. (1985). Social foundations of thought and action: A social cognitive theory: Prentice Hall.
Bandura, A. (2006). Psychological modeling: Conflicting theories. Chicago: Aldine.
Berg, B. L. (1998). Qualitative research methods for the social sciences. Boston: Allyn and Bacon.
Blanchflower, D. G., & Oswald, A. J. (1998). What makes an entrepreneur? Journal of Labor Economics, 16(1), 26.
Blumenfeld, S. N., Roxas, M., de los Santos, M., & Kumar, K. (1993). Systematic observa-tion in the analysis of primary health care services. In Rapid appraisal methods (pp. 136-156): World Bank Regional and Sectoral Studies. Washington, D.C.: World Bank.
Brown, T. C., & Hanlon, D. (2004). Developing behavioural observation scales to foster effective entrepreneurship. Journal of Small Business and Entrepreneurship, 17(2), 103-116.
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psycholog-ical Measurement, 20(1), 37-46.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation design and analysis issues for field settings. Chicago Rand McNally.
Crossley, B., & Pittaway, L. (2000). The self-perception of entrepreneurial competencies: Are large organisations losing the battle of corporate enterprise?, British Academy of Management Conference. Edinburgh, Scotland.
Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage Publications.
Edmondson, A. C., & McManus, S. E. (2007). Methodological fit in management field research. Academy of Management Review, 32(4), 1155-1179.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Manage-ment Review, 14(4), 532-550.
Envick, B. R., & Luthans, F. (1996). Identifying the activities of entrepreneur-managers: An idiographic approach, Academy of Entrepreneurship Conference. Maui, Hawaii.
Fayol, H. (1949). General and industrial management. London: Constance Storrs Sir Isaac Pitman & Sons Ltd.
Florén, H., & Tell, J. (2004). The emergent prerequisites of managerial learning in small firm networks. Leadership & Organization Development Journal, 25 (3 ), 292 – 307.
Fueglistaller, U., Müller, C., & Volery, T. (2008). Entrepreneurship modelle – umsetzung – perspektiven mit fallbeispielen aus deutschland, österreich und der schweiz Schweiz: Gabler Verlag.
Gartner, W. B. (1989). “Who is an entrepreneur?” Is the wrong question. Entrepreneurship: Theory & Practice, 13(4), 47-68.
Gartner, W. B. (2001). Is there an elephant in entrepreneurship? Blind assumptions in theory development. Entrepreneurship Theory and Practice, 25(4), 27.
Gartner, W. B. (2008). Entrepreneurship hop. Entrepreneurship: Theory & Practice, 32(2), 361-368.
Gartner, W. B., Shaver, K. G., Gatewood, E., & Katz, J. A. (1994). Finding the entrepreneur in entrepreneurship. Entrepreneurship: Theory & Practice, 18(3), 5-9.
Gibbert, M., Ruigrok, W., & Wicki, B. (2008). What passes as a rigorous case study? Stra-tegic Management Journal, 29(13), 1465-1474.
Good, W. S. (1993). Building a dream: A comprehensive guide to starting a business of your own. . Toronto: McGraw-Hill Ryerson.
Häder, M. (2002). Delphi-befragungen. Ein arbeitsbuch. Wiesbaden Westdt. Verlag.
Hales, C. P. (1986). What do managers do? A critical review of the evidence. Journal of Management Studies, 23(1), 88-115.
Kilby, P. (1971). Entrepreneurship and economic development. New York: The Free Press.
Kotter, J. P. (1982). What effective general managers really do. Harvard Business Review, 60(6), 156.
Kurke, L. B., & Aldrich, H. E. (1983). Mintzberg was right!: A replication and extension of the nature of managerial work. Management Science, 29(8), 975-984.
Linstone, H. A., & Turoff, M. (1975). The delphi method. Techniques and applications. Reading/Mass.
Luthans, F. (1987). Successful vs. Effective real managers The Academy of Management Executive, 2(2), 127-132
Luthans, F., Welsh, D. H. B., & Taylor Iii, L. A. (1988). A descriptive model of managerial effectiveness. Group & Organization Studies, 13(2), 148-162.
Man, T. W. Y., Lau, T., & Chan, K. F. (2002). The competitiveness of small and medium enterprises a conceptualization with focus on entrepreneurial competencies. Journal of Busi-ness Venturing, 17(2), 123.
Martin, W. J. (1982). Studying educational organizations with structured observation. Pea-body Journal of Education, 59(3), 160-169.
Martinko, M. J., & Gardner, W. L. (1985). Beyond structured observation: Methodological issues and new directions. Academy of Management Review, 10(4), 676-695.
Mintzberg, H. (1968). The manager at work – determinig his activities, roles, and programs by structured observation. Massachusetts Institute of Technology, Boston.
Mintzberg, H. (1973). The nature of managerial work. New York: Harper & Row.
Mintzberg, H. (1990). The manager’s job – folklore and facts. Harvard Business Re-view(March-April).
Muhr, T. (1996). Textinterpretation und theorieentwicklung mit atlas/ti. In W. Bos & C. Tarnai (Eds.), Computerunterstützte inhaltsanalyse in den empirischen sozialwissenschaften. Münster: Waxmann.
Nunally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.
Punch, K. F. (2005). Introduction to social research—quantitative & qualitative approaches: Constantinos N. Phellas.
Saunders, M., Lewis, P., & Thornhill, A. (2003). Research methods for business students (3 ed.). Harlow: Financial Times, Prentice Hal.
Schunk, D. H. (2008). Learning theories: An educational perspective (Vol. 5): Pearson Merrill Prentice Hall.
Schwehm, M. O. (2007). Entrepreneurship im internationalen vergleich: Erfassung, beobachtung und erklärung. Marburg: Books on Demand GmbH.
Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.
Stevenson, H. H., & Jarillo, J. C. (1990). A paradigm of entrepreneurship: Entrepreneurial management. Strategic Management Journal, 11(4), 17-27.
Stewart, R. (1989). Studies of managerial jobs and behaviour: The ways forward. Journal of Management Studies, 26(1), 1-10.
Stewart, R. (1993). Managerial behaviour. In To be published: International encyclopedia of business and management: Routledge.
Tengblad, S. (2001a). Classic, but not seminal: Revisiting the pioneering study about mana-gerial work. Scandinavian Journal of Management.
Tengblad, S. (2001b). Examining the stability of managerial behavior. Göteborg: Gothenburg Research Institute, School of Economics and Commercial Law, Göteborg University.
Timmons, J. A. (2002). New venture creation – entrepreneurship for the 21st century.
Watson, J. B. (1930). Behaviorism Chicago: University of Chicago Press.
Weber, R. (1985). Basic content analysis. Beverly Hills, CA: SAGE.
Windelband, W. (1894). Geschichte und naturwissenschaft. Straßburg.
Yin, R. K. (1994). Case study research: Design and methods. (2 ed.). Thousand Oaks, CA: Sage.
Yukl, G. A. (2005). Leadership in organizations. New Jersey: Prentice Hall.
Yin, R. K. (1998). The abridged version of case study research: Design and method. In L. R. Bickman, D. J. (Ed.), Handbook of applied social research methods. Thousand Oaks, CA: Sage.
Zunker, N. (2004). Systematische beobachtung. Munich: Grin.
Zuriff, G. E. (1985). Behaviorism: A conceptual reconstruction. New York: Columbia Uni-versity Press.
Place your order of custom research paper With us NOW. The assignment will be written from scratch by our qualified and experienced writers.
[meteor_slideshow slideshow=”arp2″]
A-Research-Paper.com is committed to deliver a custom paper/essay which is 100% original and deliver it within the deadline. Place your custom order with us and experience the different; You are guaranteed; value for your money and a premium paper which meets your expectations, 24/7 customer support and communication with your writer. Order Now
Use the order calculator below and get started! Contact our live support team for any assistance or inquiry.
[order_calculator]
