Chat with us, powered by LiveChat CU Literature The Exciting Intersection of Sustainability Discussion - Uni Pal


Understanding the Role of Theory in Research
By Wednesday: Select one of the sources that you have found in your preparation for the assignment due in Unit 5. The article should propose a theory that could be used as the framework for a study on work-life balance and online learning.
Briefly summarize the article and describe why you believe the theory is useful for developing a theoretical framework that could be used for research on work-life balance and online learning. Link to the article you used for this discussion.
Response Guidelines
By Friday: Reply to at least two fellow learners with a question about their posts that would clarify their selection.
By Sunday: Reply to each learner whose contribution you plan to use in your Unit 5 assignment. Indicate what you learned from them and state that you intend to acknowledge it in the credit section of your assignment.
( USE: UNIT 3 ( 2 , Unit 3 1 , unit 3 , 3 , unit 3 – 4 ) 
Student 1 
One of the articles chosen to write about the Unit 5 essay is called Meaningful Work on career satisfaction: a moderated mediation model of job embeddedness and work-based social support. This article by Gardener, J., Stuart, M., et al. Gardener, J., Stuart, M., et al. explained that the work of meaningfulness includes purpose, reasons, and processes influencing satisfaction to have careers. Purpose, motives, and operations develop careers of job embeddedness to create a satisfactory career opportunity from mechanism mediation. 
The Design methodology approach examined by the accountants in Thailand created the perception of a direct relationship between meaningful work and career satisfaction as comparatively arbitrated by job awareness (p.889 – 904). This design methodology approach states that communication and connections with each employee establish the significant or professional job experience and satisfying performance to develop fractional placement workers are aware of. The employees use what they can to work to implement a good relationship with each other to improve their professional experience and performance at work (Gardener, J., Stuart, M., et al., 2021, p. 889 – 904). As Gardener, J., Stuart, M., et al. (2021) stated, the viewpoints of supervisor and co-worker support had medium improvements affecting important work and job placement. Gardener, J., Stuart, M., et al. (2021) recognize profound implications for researchers and practitioners by highlighting an integrative model of organizational factors in managing human resources. When researching, there are continuing attempts to extend relevant knowledge in meaningful work and job embeddedness to identify mechanisms that amplify the structural relationship (p. 889 – 904).
I believe that the theory is helpful when developing a theoretical framework used for research on work-life balance because of the relationship between commitment to important job role tasks and improving career satisfaction through performance on the job. This commitment involves obeying new rules and regulations, performing acceptable tasks and activities at work, and communicating and cooperating with supervisors and co-workers for the best improvement in professional services and better pay.

Student 2 
Theresa Alfanza evaluated the impact of telecommuting on job performance and work life balance during COVID-19 by surveying 396 BPO company employees in the Philippines via stratified random sampling, Pearson correlation T-test and confirmatory factor analysis. The aim of this study was to determine whether a relationship between telecommuting intensity and employees job performance and work life balance exists, determine whether there was a negligible difference in employee productivity while in office versus working remotely while also testing the validity of previous frameworks related to job performance and work life balance.
The framework that this article evaluated were heuristic conceptual framework of individual work performance developed by Linda Koopmans in 2015 which focuses more on an individual’s behaviors and actions opposed to the results of those behaviors and actions and along with the quantity and quality of work completed as well as their skills and knowledge of the job when it comes to task performance. This framework is useful as it provides a clear definition of key terms relevant to the study of work life balance as well as it changes the perspective on how work performance and work life balance should be viewed.
Seminal Works: Theoretical and Empirical Foundations
By Wednesday: Select one of the foundational or seminal sources that you have found in your preparation for the assignment due in Unit 5 on your literature search on work-life balance and online learning. Briefly summarize the article and describe why you believe the source is foundational or seminal for the literature about work-life balance and online learning. Link to the article you used for this discussion.
Response Guidelines
By Friday: Reply to at least two fellow learners with a question about their posts that would clarify their selection.Sage Research Methods
How to Build Social Science Theories
For the most optimal reading experience we recommend using our website. A
free-to-view version of this content is available here, which includes an easy to
navigate and search entry, and may also include videos, embedded datasets,
downloadable datasets, interactive questions, audio content and downloadable
tables and resources.
Author: Pamela J. Shoemaker, James W. Tankard, Dominic L. Lasorsa
Pub. Date: 2011
Product: Sage Research Methods
Methods: Theory, Hypothesis testing
Keywords: theories, theory building, social science
Disciplines: Anthropology, Business and Management, Criminology and Criminal Justice, Communication
and Media Studies, Counseling and Psychotherapy, Economics, Education, Geography, Health, History,
Marketing, Nursing, Political Science and International Relations, Psychology, Social Policy and Public Policy,
Social Work, Sociology
Access Date: August 6, 2023
Publishing Company: SAGE Publications, Inc.
Online ISBN: 9781412990110
© 2011 SAGE Publications, Inc. All Rights Reserved.
Sage Research Methods
Contact SAGE Publications at
Using and Evaluating Theory
Theory building is an ongoing process. It doesn’t come to an end.
A theory may reach the point where it is useful and can be applied to solving problems. But it is still subject to
revision as further testing takes place.
Perhaps we should think of theories as moving through stages—young, middle-aged, and mature. The mature
theories are the ones that have been around awhile, that are more extensive in the development of concepts
and hypotheses, that have been most extensively tested, and that are possibly being applied to solve the
problems from which they originated.
This chapter begins with a discussion of the weaknesses of atheoretical research and a reiteration of a basic
theme of the book, the importance of theory in research. It continues with a discussion of the usefulness of
theory, a list of ten steps in building a theory, and a presentation of some criteria for evaluating theories. If we
can identify some standards for assessing theories, this should allow us to create better theories. The chapter concludes by identifying some constraints on theory building and presenting a few final suggestions that
might lead to better theories.
Atheoretical Research
Many studies in the social sciences are done without—or with little—reference to theory. Scholarly journals
offer a wide variety of descriptive studies. Even some studies attempting to explain do so with little reference
to theory. Examples would include studies that insert a haphazard set of predictor variables in a regression
analysis with the goal of seeing which ones explain the most variance. These regression studies typically offer
prediction without explanation. These kinds of atheoretical research areas have their uses, but they tend to
produce isolated studies that do not move our knowledge forward on important questions in the field. As Platt
(1964) has noted, such studies become like bricks lying around the brickyard rather than bricks that are used
to build a wall.
There may be a number of explanations for researchers conducting studies that are not based on theory:
Page 2 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
A. Expediency. It is easier to do descriptive research than theoretical research. The formulation of
hypotheses that will add to our theoretical understanding is difficult.
B. Inadequate training. Thorough training in social science research demands time.
C. Lack of clear definition or identification of theoretical problems in a particular discipline. What are
we trying to explain? In some fields, the major theoretical problems may be known, and there may
be a shared sense of urgency about solving them. Other fields may lack this kind of clarity and focus.
D. Lack of precedent. Some areas of research have few examples of good theories. In addition,
many social science theories are not well developed.
E. A field’s lack of a clearly defined and agreed-upon paradigm. Kuhn (1962) has argued that a field
moves forward most quickly when researchers identify theoretical puzzles that need to be solved. A
shared sense of what the puzzles are defines the paradigm for the field. Many fields may lack this
kind of agreement on what the puzzles are.
A major argument presented in this book is that theory-based research is essential if a field is going to advance. Theory has a number of advantages, including that it summarizes knowledge, has practical applications, and helps guide the research process.
The Usefulness of Theory
Summarizing Knowledge
A major purpose of theory is to condense and store knowledge. Theories help us to put our discoveries of
the nature of the world into statements. Dubin (1978) stated that “the ‘need’ for theories lies in the human
behavior of wanting to impose order on unordered experiences” (p. 6).
The social scientist is interested in discovering general patterns of behavior and expressing them succinctly.
This may be done in verbal statements but also through charts, graphs, or mathematical equations. Sometimes a formula, chart, or diagram can express a relationship more precisely than a sentence in the English
Page 3 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
Practical Applications
Many theories allow people to do things. The goals of science are often stated as understanding, prediction,
and control. All three of these outcomes can allow people to accomplish tasks and to bring about desirable
outcomes. In the area of television violence, for instance, research indicating that viewing television violence
can contribute to aggressive behavior led to the development of the family viewing hour and the television
rating system. The National Television Violence Study found that very few violent television programs showed
long-term negative consequences of violence and that most violent scenes showed the violence going without
punishment (National Television Violence Study, 1996). This research could lead to changes in television,
with more dramatic programs showing the consequences of violence, more punishment of violence, and so
Guiding Research
A major function of theory is to provide guidance for research. This use for theory comes directly from the
cyclical nature of the research process. Where do research hypotheses come from? Often they come from
theories. A major contribution of theory is to help guide research. Starting with theory keeps data gathering
and observation from becoming a random process or a “fishing expedition.” It also forces the researcher to
think about the value of a particular research project and how this project will relate to other projects.
Ten Steps of Building a Theory
Start with a problem, some unexpected results, an anomaly, an observation of something unusual, something
you would like to know the effects of, or something you would like to know the causes of. Bavelas (1987) suggested that the most promising ideas for new research projects often come, not from reading the literature,
but from observation.
Identify (or formulate) the key concepts involved in the phenomenon of interest (see Chapter 2). Try to come
up with concepts that are observable and that can be quantified. This step can be one of the most difficult
ones. Often you are trying to identify something and give it a name when no one else has identified it before.
On the basis of careful observation and using techniques of creativity (see Chapter 8), try to think of as many
Page 4 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
causes of the key concepts as you can. These may become important additional concepts in your theory.
On the basis of careful observation and using techniques of creativity (see Chapter 8), try to think of as many
effects of the key concepts as you can. These may become important additional concepts in your theory.
Specify theoretical definitions for all concepts (see Chapter 2).
Specify operational definitions for all concepts (see Chapter 2).
Link some of the concepts to form hypotheses (see Chapter 4). State each hypothesis in a verbal form. These
hypotheses might involve two (see Chapter 3), three (see Chapter 5), or four variables (see Chapter 6), depending on the complexity of the phenomenon and how far along you are in theory building. The hypotheses
will often state or imply causal relationships. Specify the form of the linkage—linear, curvilinear, power curve,
or other. Is the relationship bivariate or multivariate? What statistical test will be used to investigate the linkage?
Specify the theoretical rationale for the hypotheses (Chapter 4). Why should they be expected to be true?
Write this explanation out, either as a paragraph or as a list of reasons.
Try to think in terms of multiple hypotheses, as recommended by Chamberlin (1890/1965) and Platt (1964).
These multiple hypotheses are sometimes alternative explanations for the same phenomenon. In some cases, called crucial experiments, you will be trying to demonstrate that one hypothesis is true and the other is
false. But in other situations, several hypotheses might be supported (lending strength to a theory of multiple
Try to put the hypotheses in some kind of organized system. They might take the form of an ordered list.
Some hypotheses might be logically derived from others, for instance. Or hypotheses might relate to different
parts of a model, such as Lasswell’s (1948) verbal model “Who Says What, in Which Channel, To Whom,
With What Effect?” Galtung and Ruge (1965) used a metaphor of a radio broadcast station to develop a list
of 12 hypotheses about factors that influence international news flow.
Evaluating Theories
Not all theories are created equal. It is important to have a set of criteria for evaluating theories. Authors have
listed many criteria for evaluating theories. Some of the most important are the following.
A theory needs to be testable. Basically, it needs to be stated in terms of concepts or variables that can be
Page 5 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
Theorists need to be thinking constantly about how their theories can be tested. For instance, a theorist might
have an idea that younger people have a different means of taking in information than older people because
the younger group has grown up watching television. In what ways could this idea be made testable? Part of
the task for the theorist in this example would be to try to think of those ways in which the taking in of information might be different.
Karl Popper (1959) introduced the idea that a theory should be not only testable but also falsifiable. One difficulty in attempting to falsify a theory is that evidence can often be interpreted in a way that supports the theory
(p. 107 n.). Popper (1965a) argued that the refutation of any theory is a step forward that takes us nearer to
the truth (p. vii).
Stinchcomb (1968) added the idea that a theory must be stated specifically enough that it could be falsified:
“That is, a theory to be useful must be specific enough that it might be disproved” (p. 5). Stinchcomb went so
far as to argue that “social theorists should prefer to be wrong rather than misunderstood” (p. 6).
The simplest theory is the best. If two or more theories have the same explanatory and predictive power but
one is simpler than the other, the simpler one is to be preferred. A theory that is so complex that an individual
cannot remember its main points may be suffering from a lack of parsimony.
There are a number of trade-offs in theory building. One is between the simplicity of the theory and the precision of the prediction or explanation. A more elaborate theory will probably allow more precise prediction and
better understanding, but it may do so at the cost of being more complex. All in all, the simplest theory that
provides the desired level of precision is to be preferred over more complex theories that achieve the same
Some theories of human behavior are undoubtedly too simple. For instance, the notion that the growth in digPage 6 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
ital communication is going to mean the death of print is probably too simple. Or a theory that stated that all
people are influenced equally and in the same ways by television violence would be too simple.
Explanatory Power
The primary goal of theory is to provide explanations. The better the explanation, the better the theory. Likewise, the more phenomena a theory explains, the better the theory. Explanatory power is closely related to
parsimony. A theory that can be stated succinctly but can also explain a great deal is high in explanatory
power. Another important consideration is that the explanation should be true (Hage, 1972). For instance, we
might think that one variable causes another because they are correlated when the relationship is actually
spurious. Our explanation will be more correct if we identify that initial relationship as spurious.
Predictive Power
An important goal of science is prediction. Knowledge that allows people to make accurate predictions is important and useful. In general, the more precise the prediction, the better the theory. Interestingly, a theory
could be high in predictive power while at the same time low in explanatory power. Dubin (1978, p. 19) noted
that in the behavioral sciences understanding and prediction are not often achieved together. For instance, a
regression equation with a large number of variables may let us predict some outcome with great accuracy,
but we may not understand the mechanisms or processes that are taking place.
Even simple two-variable relationships can have great predictive power. For instance, knowledge of a person’s income will often allow one to predict with some accuracy whether the person will vote Democratic or
Republican (Table 9.1).
The more phenomena that a theory helps us understand, the better the theory. A theory that helped us understand the effects of television violence would be more useful than a theory that helped us understand the
effects of one particular show such as The Sopranos. Hage (1972) suggested that one measure of scope is
the number of basic problems of a discipline that are handled by a theory. In the social sciences, most theoPage 7 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
ries deal with a limited range of behaviors and therefore are low in power.
Scope can also be thought of as generality. A theory that is high in scope will apply to a number of different
Cumulative Nature of Science
Theory is not static but is changing and growing. Research is cumulative, with later studies building on earlier
studies. Through this process, theory is continuously refined as we test hypotheses with appropriate evidence. New studies probe the unanswered questions left by earlier studies. In this way, theories move to a
closer approximation to the truth.
Table 9.1 Relationship between income and party vote in the 1998 congressional election
In addition, no single study can provide adequate support for a hypothesis. Every study has strengths and
weaknesses. It is important to examine a hypothesis with a variety of tests using different methods. Campbell
and Fiske (1959) have argued for a multimethodological approach to measuring concepts.
And the progress may not take place without effort and setbacks. As Stephen Jay Gould (1988) noted, “Science is not a linear march to truth but a tortuous road with blind alleys and a rubbernecking delay every mile
or two” (p. 17).
Page 8 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
Degree of Formal Development
Theories range greatly in their degrees of formal development. We can visualize a continuum of degree of
formal development of theories. At the lower end, we would have “areas of research” in which concepts are
being developed, hypotheses are being formulated, and data are being gathered, but there is not an effort to
be exhaustive about the parts of a theory or to arrange propositions in a logical system. At the higher end,
we would have theories made up of systems of propositions, with some logically deduced from others. These
more formal theories would also include most or all of the elements of theory discussed in this book—concepts, theoretical definitions, operational definitions, hypotheses, theoretical linkages, operational linkages,
limits, and assumptions. Some theories in the social sciences, such as Miller’s (1955, 1995) general system
theory, are at the higher end of the scale (Table 9.2). Many areas of research in the communication field appear to be at the lower end.
Table 9.2 Sample propositions from general systems theory
When we are thinking about the degree of formality of a theory, it is useful to examine how authors present
Page 9 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
their own work. Is it presented as a formal theory? Is it called a theory? When should an area (or hypothesis)
be called a theory? Maybe some good theoretical work is going on in an area to which the label theory has
not yet been applied. Most of our theories in the social sciences are not finished—they are works in progress.
Many of them are in early stages of development. They may not be written down in one place as completed
theories, but they may have many of the components that we have presented in this book—concepts, theoretical statements, theoretical rationales, theoretical and operational definitions, and so forth.
Social science theories often consist of verbal statements or hypotheses. It may be desirable to move to a
more formal, mathematical form for these theories. Often theories that consist of verbal statements can be
recast as path models. A path model can be created by identifying the variables and representing them as
boxes, then specifying the linkages between them with lines or arrows (see Chapter 7 for a discussion of path
Heuristic Value
A theory is valuable when it helps us generate ideas for research and when it leads to other theoretical ideas.
The more new hypotheses that can be generated from a theory, the better the theory.
Although it is perhaps not an essential criterion, some writers suggest applying an aesthetic principle to the
evaluation of theories. Kaplan (1964, p. 318) claimed that a theory can be beautiful. In a similar way of thinking, physicists sometimes talk about an idea or explanation being “sweet.”
Unfortunately, it is likely that all these criteria cannot be fulfilled simultaneously. For instance, there is probably
a trade-off between generality, simplicity, and accuracy. “It is impossible for an explanation of social behavior
to be simultaneously general, simple, and accurate” (Thorngate, 1976, p. 126).
The parts of a theory presented in this book—concepts, theoretical definitions, operational definitions, hypotheses, theoretical linkages, operational linkages, limits, and assumptions—give us one more way of evaluating a theory. How many of the parts have been developed, and how well have they been developed? The
more parts have been explicitly dealt with and included, the better the theory.
Page 10 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
Other discussions of criteria for evaluating theories can be found in Hage (1972), Agnew and Pyke (1978),
Littlejohn (1983), and Burgoon and Buller (1996).
Sample Evaluation of a Theory
It may help the reader to understand the 10 criteria for evaluating a theory if they are applied to a particular
theory. In the following paragraph and Table 9.3, the 10 criteria are applied to cultivation theory, a theory developed by George Gerbner and his associates (Gerbner, Gross, Morgan, & Signorielli, 1986, 1994) to explain the effects of television watching on viewers’ values and worldviews. This theory initially presented one
main hypothesis, which stated that watching television shapes viewers’ beliefs, ideologies, and worldviews.
The hypothesis was modified later in the face of research evidence indicating that cultivation effects did not
hold up across the board for all viewers.
Page 11 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
Table 9.3 An evaluation of cultivation theory
Page 12 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
The revised cultivation hypothesis introduced the concepts of mainstreaming and resonance. These concepts
specified the conditions under which cultivation would be more or less likely to occur. But the added concepts
may have also limited the scope and explanatory power of the theory. Some writers have also suggested that
the revised theory incorporating mainstreaming and cultivation allows for so many outcomes that it has rendered the theory unfalsifiable.
Scholars have also debated whether the hypothesis in its original form, which looked at television viewing
as the major independent variable, is too general. Researchers have suggested that cultivation effects may
differ depending on the kind of television program being watched—that is, cultivation effects may be program
specific. Further revisions of the theory have separated the effects into two types—first-order beliefs (realitybased perceptions, such as rate of occurrence of a particular crime) and second-order beliefs (for instance,
the idea that the world is a dangerous place).
Although cultivation theory has been one of the most active areas of communication research, it still seems to
be an area of research more than a theory. It is difficult to say whether it is an area showing a healthy cumulative growth and refinement of hypotheses or an area in trouble because some of the main propositions have
not been supported by evidence and the scope of the theory is being diminished by the new propositions.
Platt (1964) noted some time ago that certain areas of science, including molecular biology, have moved forward much more rapidly than others. He suggested that this rapid progress has occurred because these areas have used a particularly effective approach to research that he called strong inference.
Strong inference is a specific way of thinking about research. The method involves suggesting multiple hypotheses, subjecting the hypotheses to tests that permit sharp exclusion of certain hypotheses, and then formulating new multiple hypotheses to move the process forward another step. The process allows research to
progress along logical branches, with certain branches being eliminated because they are not supported by
evidence. Then the process moves to the next branching point.
Strong inference and falsifiability of hypotheses seem to go together well. Hypotheses and theories that are
capable of falsification will allow the sharp exclusion of wrong hypotheses that is required by strong inference
thinking. Strong inference implies a linear progress in research that may be difficult to achieve in the social
Page 13 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
sciences but is still worth striving for.
Constraints on Theory Building
Several forces, some obvious and some not so obvious, act to constrain the building of theory. Four of the
most important are listed here.
1. Availability of funding causes some theoretical problems to receive a great deal of attention from researchers and others to be ignored (Martin, 1982, p. 27). In the mass communication research field, funding
has frequently been made available for research on the effects of television violence. As a consequence,
researchers who might have studied other theoretical problems ended up conducting studies of television violence.
2. There are constraints on the operationalizations of variables, or the choices of measurements (Kulka, 1982,
p. 58). A familiar example is the secondary analysis of survey data, where the researcher is forced to use
questions developed by other researchers for other purposes. Researchers are also constrained by their tendency to use the operationalizations of variables that have been used by prior researchers. Even beyond
these considerations, if something is not easy to measure, it is less likely to be studied. “Much of what is considered an acceptable area of investigation is that for which there exists the possibility of operationalization”
(DeCarufel, 1976, pp. 336-337).
3. Choices must be made in the formulation of theory. One choice has to do with the scope or generalizability
of the theory. It might be possible to create a theory that would allow nearly perfect prediction of a very specific
phenomenon. Such a theory might actually be close to a simulation. This would be a theory that was deliberately narrow in scope. Or the theorist might choose to go in the other direction and create a theory that would
explain a much larger class of phenomena, such as a theory of aggressive behavior. McGrath (1982) pointed
out that a theory that is high in generalizability will probably be low in realism of context and in precision of
measurement. This might be called the “precision versus scope dilemma” (p. 93).
4. Certain statistical models direct researchers toward certain kinds of problems. For instance, many researchers think in terms of statistical tests, and a common approach in statistics is to try to explain as much
variance as possible. But this way of looking at research may lead the theorist to ignore problems that do not
involve variation. For instance, much of the content of the mass media may be similar from day to day. How
Page 14 of 15
How to Build Social Science Theories
Sage Research Methods
Contact SAGE Publications at
could a researcher investigate the effects of this similarity? “The pursuit of variance leads researchers to pick
certain problems and ignore others” (Lieberson, 1985, p. 91). The “explaining the variance” approach has
other limitations. It implies that causes are being found, when in fact the relationships being uncovered are
often rather weak. For instance, a study that yields a correlation (or a multiple correlation) of .40 is explaining
only 16% of the variance. What about the other 84%?
Final Suggestions
The following suggestions should lead to better theory building:
Be aware of the various constraints on theory building, and try to minimize their impact.
Operationalize variables so that they have a number of points on a continuum, not just two.
In testing propositions, use multiple methods (see Campbell & Fiske, 1959).
Be wary of the explaining the variance approach. It has its uses, but it also has its limitations. One alternative
is the process of exclusion of possible explanations advocated in the strong inference approach.
Be prepared for failure and ready to move on anyway. Darwin and Faraday said that most hypotheses turn
out to be wrong (Beveridge, 1957, p. 79). Campbell and Stanley (1963) said we should inoculate graduate
students to expect failure in testing hypotheses. Beveridge (1957, p. 97) said that not all intuitions turn out to
be correct.
Editors of journals and reviewers of manuscripts should look for contributions to theory in the manuscripts
they review and be more willing to publish manuscripts that are aimed at theory building.
Many of the social sciences may be held back by a lack of attention to theory building. Much research has
shown an overemphasis on collection of empirical data without a clear sense of theoretical purpose. One
of the easiest ways many researchers could improve their research would be through increased attention to
theory building. The payoff for the individual researcher and the field overall would be more efficient summarization of information and a clearer sense of direction about future research.
Page 15 of 15
How to Build Social Science Theories
JOMXXX10.1177/0149206316647102Journal of ManagementShepherd, Suddaby / Theory Building
Journal of Management
Vol. 43 No. 1, January 2017 59­–86
DOI: 10.1177/0149206316647102
© The Author(s) 2016
Reprints and permissions:
Theory Building: A Review and Integration
Dean A. Shepherd
Indiana University
Roy Suddaby
University of Victoria
Newcastle University
Building theories is important for advancing knowledge of management. But it is also a highly
challenging task. Although there is a burgeoning literature that offers many theorizing tools, we
lack a coherent understanding of how these tools fit together—when to use a particular tool and
which combination of tools can be used in the theorizing process. In this article, we organize a
systematic review of the literature on theory building in management around the five key elements of a good story: conflict, character, setting, sequence, and plot and arc. In doing so, we
hope to provide a richer understanding of how specific theorizing tools facilitate aspects of the
theorizing process and offer a clearer big picture of the process of building important theories.
We also offer pragmatic empirical theorizing as an approach that uses quantitative empirical
findings to stimulate theorizing.
creativity; grounded theory; philosophy of science; entrepreneurship theory
Management scholars have been highly attentive to the role of theory. A prerequisite for
publication in elite management journals is that papers make a contribution to theory (Colquitt
& Zapata-Phelan, 2007; Hambrick, 2007; Rynes, 2005; Sutton & Staw, 1995). While some
scholars question the extent of this preeminence of theory (Hambrick, 2007; Pfeffer, 2014),
there is little argument about the importance of building theories for advancing knowledge of
management (Suddaby, 2014a). For example, business scholars have called for new theories
of organization (Suddaby, Hardy, & Huy, 2011), entrepreneurship (Shepherd, 2015),
Acknowledgments: The authors would like to acknowledge, for comments on the previous versions of the article,
J. Craig Wallace (action editor), two anonymous reviewers, and the participants of both the Australian Centre for
Entrepreneurship Research Exchange and the QUT boot camp.
Corresponding author: Dean A. Shepherd, Indiana University, Kelley School of Business, 1309 E. Tenth St.,
Bloomington, IN 47405, USA.
60   Journal of Management / January 2017
management (Barkema, Chen, George, Luo, & Tsui, 2015), work (Okhuysen et al., 2013),
compassion (Rynes, Bartunek, Dutton, & Margolis, 2012), and so on. Despite the widespread
recognition of the importance of building theory, doing so is a highly challenging task (Weick,
1995). As a result, there is a growing literature in management on the process of theorizing—
that is, how to build theories. This burgeoning literature offers many tools and approaches to
theorizing, for example, engaged scholarship (Van de Ven & Johnson, 2006), metaphor
(Cornelissen, 2005), and finding the balance between novelty and continuity (Locke &
Golden-Biddle, 1997). These papers have made important contributions by offering different
insights into select aspects of the theorizing process—that is, different ways to stimulate the
creation of a new theory, different ways to build new explanations of management phenomena, and different notions of what represents a theoretical contribution, respectively.
But where does this leave budding theorists? It seems to leave them with an array of potential tools for theorizing without a coherent understanding of how these “theorizing tools” fit
together; there is little information about when to use a particular theorizing tool vis-à-vis a
different theorizing tool (i.e., substitutes) and which combination of tools can be used in the
theorizing process (i.e., complements). Therefore, while these approaches address discrete
and often isolated questions about “how” to construct specific aspects of theory, they fail to
offer a coherent explanation for how and when to engage the various tools that facilitate theorizing. Thus, our intent is to integrate the various threads of how to build theory. We then
extend that integration to a specific theorizing approach—pragmatic empirical theorizing.
Our systematic review of the literature on theory building in management integrates the various individual components of theory building into a coherent whole. Our reading of this growing
literature reveals the distinct importance of narrative or storytelling in theorizing (Pollock &
Bono, 2013; Van Maanen, 1995)—that is, compelling theories are at their core compelling stories. Compelling stories are built around main characters who engage in a struggle with a powerful entity (narrative conflict) within a narrative setting. The story is held together by the
sequence of events and made comprehensible by the plot. The narrative arc concludes with a
resolution of the problem of the story and/or the problem faced by the main character(s) of the
story. Accordingly, we organize our review of theory building around the five key elements that
inform every great story: conflict, character, setting, sequence, and plot and arc.
By reviewing and organizing the literature on theory building, we hope to make three
primary contributions. First, organizing the literature on theory building provides the opportunity to integrate “like tools” to provide a richer understanding of how these like tools facilitate a specific aspect of the theorizing process. Second, organizing the literature provides the
opportunity to connect different aspects of the theorizing process. With a deeper understanding within and across theorizing aspects, we gain a clearer “big picture” of the process of
building interesting theories. Finally, we offer a theorizing tool—pragmatic empirical theorizing—that we believe has potential for advancing theories of management. In short, pragmatic empirical theorizing uses quantitative empirical findings to stimulate theorizing as part
of an abductive process of inquiry.
To select the articles for review, we used keyword searches in general management journals (consistent with other recent review articles; Shepherd, Williams, & Patzelt 2015; Surdu
Shepherd, Suddaby / Theory Building   61
& Mellahi, in press; Wang & Rajagopalan, 2015) publishing work on theory building. These
journals include the Academy of Management Journal, Academy of Management Review,
Academy of Management Annals, Administrative Science Quarterly, Journal of Management,
Journal of Management Studies, Organization Science, Management Science, and Strategic
Management Journal. We choose these journals because, according to the Web of Knowledge,
they are the highest impact general management journals in the category of “Management”
that are not journals focused on psychology, operations management, research methods, or
international business (Thomson Reuters), with the exception of Management Science, which
has the reputation as a top journal (despite a lower impact factor). To provide an initial list of
papers on theory building, we searched for papers that included in their title the word(s)
theory or theorizing or theories. Not surprisingly, this generated a large number of
papers—973. We further refined this list by reading the abstract of each of these papers (and
when necessary the full paper) to determine their appropriateness given the purpose of the
review. Specifically, we excluded papers that did not have theory building at their core (788
papers) and excluded papers that were commentaries, research notes, and book reviews (127
papers). Furthermore, in the process, we necessarily considered some contributions in books.
The remaining 58 papers (marked with an asterisks in the reference section) were categorized
into theory-building topics arranged based on the key elements that inform every great story:
conflict, character, setting, sequence, plot, and arc.
1. Theorizing Trigger—the Narrative Conflict
Perhaps the most challenging aspect of theorizing is identifying an anomaly or tension to
motivate and guide the process. Doing so is a creative process that requires both considerable
imagination (Mills, 1959) and acute powers of observation, skills that, according to March
(1970), can be best learned by attending to the observational habits of outstanding
In stories, narrative conflict represents the struggle between two powerful entities—
human versus human, human versus nature, or human versus god. In theory, narrative conflict reflects a struggle between two realms of knowing—the empirical world of phenomena,
on one hand, and the scholarly world of theoretical literature that attempts to describe the
empirical world, on the other. Conflict can arise from within either of these worlds and, perhaps more typically, can arise from gaps that occur between them. We examine each in turn
to identify the various techniques used by management scholars to “trigger” the theorization
process. In Table 1 we detail each form of narrative conflict, its function, the key cites, and
an example.
Conflict in the Literature
Immersion in the literature can reveal paradoxes, problems, challenges, and puzzles. A
paradox involves “contradictory yet interrelated elements that exist simultaneously and persist over time” (Smith & Lewis, 2011: 382). Recognizing the underlying tension between two
sets of relationships that appear to make sense when considered independently but contradictory when considered simultaneously can trigger theorizing as an attempt to resolve the paradox. Paradoxes arise from changes in system(s), differences in individual and collective
Following an actor’s activities can reveal
relationships across practices, the
connections between people and tools,
and events that disconnect individuals
from their activities, all of which may
highlight paradoxes and problems of
practical importance.
Searching, observing, and questioning
everyday events provides abundant
opportunities to theorize and searching,
observing, and questioning absurd events
challenges conventional wisdom.
Collaborating with practitioners provides
the academic access to a different
perspective as a basis for identifying
complex real-world problems.
Practice logic
or absurd
Observing through the senses can reveal
data and findings that would not
otherwise be expected, which requires
theorizing for an explanation.
Recognizing the tension between two sets
of statements that on their own make
sense but together are contradictory
triggers theorizing to resolve the paradox.
Challenging the value of a theory and/or
focusing on its weaknesses highlights the
need for new thinking on the topic.
Empirical surprise
Narrative Conflict
“ ‘[O]n the one hand, conflict improves decision quality; on the other, it may
weaken the ability of the group to work together.’ … How can top management
teams use conflict to enhance the quality of their decisions, without sacrificing
consensus and affective acceptance among their members?” (Amason, 1996: 123).
“An idea from an ethic of care perspective that is important to the construction of
people’s struggles is the problematization of the division between public and
private spheres (Held, 2005; Tronto, 1993). An ethic of care draws attention to
the ways in which ostensibly private problems and issues are the result of public,
political processes” (Lawrence & Maitlis, 2012: 646).
“This study of the complete life-spans of eight naturally-occurring teams began with
the unexpected finding that several project groups, studied for another purpose,
did not accomplish their work by progressing gradually through a universal series
of stages, as traditional group development models would predict. Instead, teams
progressed in a pattern of ‘punctuated equilibrium,’ through alternating inertia and
revolution in the behaviors and themes through which they approached their work.
The findings also suggested that …” (Gersick, 1988: 9).
A nine-year ethnography is used to show how two investment banks’ controls,
including socialization, targeted bankers’ bodies, how the bankers’ relations
to their bodies evolved, and what the organizational consequences were. The
banks’ espoused and therefore visible values emphasized autonomy and worklife balance; their less visible embodied controls caused habitual overwork that
bankers experienced as self-chosen. This paradoxical control caused conflict
between bankers and their bodies, which bankers treated as unproblematic objects.
“If one watches people ride on escalators, he will observe that there are times when
they walk on the escalator in order to speed up their ride. Now the question is, is
there any regularity to this pattern of walking? Informal observation suggests that
… the closer they are to where they want to get, the stronger is their tendency to
approach it” (Weick, 1974: 488).
“To explore change and managerial sensemaking, we conducted action research
at the Danish Lego Company. … Through collaborative intervention and
reflection, we sought to help managers make sense of issues surfaced by a major
restructuring. Results … a process for working through paradox and explicating
three organizational change aspects—paradoxes of performing, belonging, and
organizing” (Luscher & Lewis, 2008: 221 [Note: Luscher is a practitioner]).
Poole & Van de Ven,
1989; Smith &
Lewis, 2011
Van de Ven &
Johnson, 2006
Weick, 1974
Feldman &
Orlikowski, 2011;
Sandberg &
Tsoukas, 2011
Alvesson & Karreman,
2007; Locke &
1997; Shepherd &
Sutcliffe, 2011
Locke, 2007; Shah
& Corley, 2006;
Turner, 1983
Key Cites
Table 1
Narrative Conflict as a Trigger for Theorizing
Shepherd, Suddaby / Theory Building   63
identity, competing organizing modes/designs, and different stakeholder goals (Smith &
Lewis, 2011). Paradoxes also exist across the categories of learning, belonging, organizing,
and performing and represent (or create) a tension that can stimulate theorizing that is more
encompassing as an attempt to reconcile the apparent paradox (Poole & Van de Ven, 1989).
Problematization is another way to engage the literature to stimulate theorizing. To problematize means to “challenge the value of a theory and to explore its weaknesses and problems in relation to the phenomena it is supposed to explicate” (Alvesson & Karreman, 2007:
1265-1266). This problematization highlights the need for rethinking existing theory and
perhaps the need to head in a new direction. To problematize requires an understanding of the
literature. However, it also requires an open-minded approach to that literature. Theorists can
approach the literature with an open mind to allow the literature (as data consistent with a
grounded theory approach) to “speak to them” to reveal (in a bottom-up way) problems in or
across literatures (Shepherd & Sutcliffe, 2011). Problematizing also involves considerable
rhetorical skill in constructing the “gap” between the literature and the real world or describing a logical flaw in existing theory (Locke & Golden-Biddle, 1997) because it is likely
(hopefully) not a simple case of incremental gap-spotting but a substantial gap that challenges important assumptions (Sandberg & Alleveson, 2011). Contrastive questions can help
problematize a situation or explanation by referring to different aspects of the event (i.e., an
allomorph) or highlighting the fact to be explained and contrasting it with an alternative(s)
(i.e., fact and foil) (Tsang & Ellsaesser, 2011). The notion underlying contrastive explanation
is that by asking better questions the theorist can begin the process towards offering better
explanations (Tsang & Ellsaesser, 2011). Indeed, Abbot (2004) proposes a number of heuristics that can facilitate discovery by changing the way the budding theorist conceptualizes a
problem or solution. For example, problematization can be stimulated through reversing a
well-known proposition, switching figure and ground, using emotional language, and as we
elaborate on below, “putting things in motion” (Abbot, 2004).
Conflict Revealed Through Empirical Phenomena and Practice
Although the data stimulating theorizing can come from the literature (as detailed above),
it can also come from the phenomena of interest: knowledge discovery starting with “observation by the senses” (Locke, 2007: 888). Again, however, the theorist needs to approach the
phenomena and the associated data with a somewhat open mind; otherwise, the data and/or
its interpretation will simply be forced to fit existing theories. With an open mind (i.e., withholding as best one can prior expectations), collecting and analyzing data can reveal interesting research problems—namely, “the high potential for an empirical response and a novel
insight that adds significantly to—or against—previous understandings” (Alveson &
Karreman, 2007: 1268) and, in the case of grounded theory, can “elicit fresh understandings
about patterned relationships” and social interactions (Shah & Corley, 2006; see also Glaser
& Strauss, 1967; Turner, 1983).
One important source of empirical material for stimulating theorizing on management
phenomena can come from an orientation toward practice—how organizational activities are
constituted and enacted by actors (Sandberg & Tsoukas, 2011: 339). Because recurrent
actions represent the building blocks of a social understanding for those in or affected by
organizations (Feldman & Orlikowski, 2011), theorizing triggered by practice helps reveal
paradoxes and problems of practical value to managers. To do so might require the theorist
64   Journal of Management / January 2017
to zoom in on the specific activities in context or zoom out to attend to the relationships
across practices to gain a deeper understanding of the connections and possibilities of activities, tools and interactions (Bechky, 2011; Nicolini, 2009; Sandberg & Tsoukas, 2011).
Indeed, in performing organizational activities, managers and/or employees are often one
with the task (Dreyfus, 1995) but it is when they experience a temporary breakdown in the
effectiveness of the activity—a momentary disconnection of the individual from others and/
or things—that they detach from the task and engage in deliberate reflection (Sandberg &
Tsoukas, 2011). These temporary breakdowns reveal problems for the manager and by extension an opportunity to theorize to gain a deeper, richer, and practically useful understanding
of the situation and/or task. Such theorizing helps to “explore new terrain and develop novel
ideas, thus potentially overcoming the inherent conservatism in well-established frameworks” (Alvesson & Karreman, 2007: 1267).
Indeed, Weick (1974) suggests a theorist focus on everyday events, everyday places,
everyday questions, micro-organizations, and absurd organizations. By searching, observing,
and/or questioning everyday events in everyday places, theorizing itself can become more
commonplace rather than tied to Fortune 500 companies or the “armchair.” It starts by
observing a pattern and building more and more robust explanations for the pattern of the
focal task (and organizing tasks more generally). Similarly, a focus on micro-organizations
reduces the emphasis on the centrality of the thing—the organization—and more on the process—the organizing. Studying the absurd organizations—almost by definition (of absurd)—
challenges the theorist’s fundamental assumptions, which is an important step toward
theorizing to open up new terrain (Weick, 1974) and generate contributions to knowledge.
Using engaged scholarship can also stimulate new theorizing. Engaged scholarship is “a
collaborative form of inquiry in which academics and practitioners leverage their different
perspectives and competences to co-produce knowledge about a complex problem or phenomenon that exists under conditions found in the world” (Van de Ven & Johnson, 2006:
803). Engaged scholarship is likely to be most useful when the associated projects are
designed to address complex real-world problems, to be a collaborative learning environment, to operate for an extended duration, and to employ multiple frames of reference (Van
de Ven & Johnson, 2006). This problem-driven research requires the researcher to be at least
somewhat engaged with the practitioner performing his or her activities, to be open to new
(vis-à-vis existing theories) experiences, and to be self-reflective of his or her engaged scholarship role (Van de Ven & Johnson, 2006). In doing so, the researcher is taking a step toward
addressing what has been argued as a large gap between theory and practice (Anderson,
Herriot, & Hodgkinson 2001; Rynes, Bartunek, & Daft 2001). By collaborating with practitioners throughout the process, the theorist is able to formulate a problem grounded in the
experiences of those engaged in the task (Van de Ven, 2007)—a real-world problem, whose
solution can make a contribution to academic and practitioner knowledge.
Conflict Between Literature and Phenomena
We have described how a trigger for theorizing arises when the researcher encounters an
unexplained puzzle resulting from an unexplained phenomenon that defies extant knowledge. Considerable effort has been devoted in management theory to debating the relative
importance of phenomenal gaps over gaps in the literature. Advocates of the former tend to
Shepherd, Suddaby / Theory Building   65
grant primacy to empirical facts (Hambrick, 2007; Pfeffer, 2014). They are supported by
intellectual giants in social theory, such as Durkheim (1895/1964: 15), who argues that
researchers should move from “things to ideas” not from ideas to things. However, the pragmatic consensus—supported by a long procession of writers beginning with Peirce (1934),
extending to Merton (1967), and advancing today with Weick (2014)—is that effective theorizing is a process in which the researcher moves iteratively between the gaps observed in the
phenomenal world and those observed in the extant literature. Indeed, it is often the tension
created by a gap between the literature and the phenomenal world that ultimately triggers the
need for new theory.
Having triggered the theorizing process by discovering or generating a conflict—a paradox, problem, or challenge—the theorist conceives of a research idea, perhaps first as a
simple construct or guess, that is then constructed into a theory.
2. Conceiving and Constructing Theories—Building Stories
We use a narrative framework to organize research on conceiving and constructing theories because it reinforces the notion that powerful theorizing involves skillfully weaving
together prior knowledge (i.e., existing literature) and emerging knowledge (i.e., new empirical observations). As illustrated in Table 2 and detailed below, building stories is facilitated
by storytelling that involves main characters, a narrative setting, an event sequence, and a
Identifying Core Constructs: The Main Characters
Effective stories are built around main characters (Pentland, 1999)—actors whose behavior
best captures the narrative of interest. In storytelling, a character is an actor—a person, animal,
or entity—whose experience is the focal point of the story. Just as stories are built around main
actors, so too are theories built around core constructs (Pentland, 1999). The act of naming a
core construct early in the process of theorizing is a critically important step because even
though the theoretical narrative is not yet clear and the construct itself is still somewhat fuzzy,
the act of putting a formal name to the phenomenon of interest is an essential step in conceptually separating one’s phenomenon from the mass “noise” of our everyday empirical experience
and/or separating one’s core construct from the mass “noise” of prior research.
Theorists have adopted a range of different strategies for naming constructs. The most
common strategy, perhaps, is to simply use a common everyday word that most closely captures the phenomenon of interest. So, for example, the somewhat generic word performance
has been used to describe the range of activities by which we evaluate organizations. Noted
sociologist Max Weber (2001: 63) endorsed this approach, advocating the use of “the nearest
and most descriptive words” from common language to name constructs. However, there are
clear risks to using dictionary definitions. Foremost is the risk that adopting a term in common everyday use will burden the construct with too much “surplus meaning” (Cronbach &
Meehl, 1955). Thus, the use of the term performance invites theorists to infer, consciously or
otherwise, a range of meanings of performance drawn from individuals, machines, sports
teams, and a range of other entities and activities, which substantially reduces the analytic
precision of the construct.
Narrative’s event sequence
Considering time from different perspectives—e.g.,
how time is experienced, bracketed, categorized
as periods of stability and change, considered
in terms of rate, magnitude, and pattern, and
the interrelationship between the past, present
and future—can allow theorizing to extend the
boundary conditions of existing theories.
Corley & Gioia,
2011; Dansereau,
et al., 1999;
George & Jones,
2000; Langley,
1999; Zaheer
et al., 1999
Dyer & Wilkins,
1989; Eisenhardt
& Graebner,
Klein et al., 1994;
Morgeson &
Hofman, 1999;
Shepherd &
Sutcliffe, 2015
Entering the field with a research question and
perhaps focal constructs, selecting cases that
are extreme or highly revelatory, and pattern
matching data and theory enables the theorist
build a story that bridges rich qualitative
evidence with deductive research.
Making explicit the mechanisms by which
constructs and relationships are influenced by
lower and/or upper level constructs can provide
new insights at the focal level or cross-levels
and a basis for theorizing on the emergence of,
stability in, and changes to collective constructs.
Back and forth
between data and
Choice of levels
Gioia & Pitre,
1990; Lewis &
Grimes, 1999;
Ofori-Dankwa &
Julian, 2001
“‘Team temporal leadership’ orients teams toward managing the time-related
aspects of their work. We examine how perceived time pressure affects
team processes and subsequent performance under weak versus strong team
temporal leadership” (Maruping et al., 2015: 1313).
“To label these five as ‘not theory’ makes sense if the problem is laziness and
incompetence. But ruling out those same five may slow inquiry if the problem
is theoretical development still in its early stages. Sutton and Staw know
this. But it gets lost in their concern with theory as a product rather than as a
process. To add complication and nuance to their message, I want to focus on
the process of theorizing” (Weick, 1995: 385).
“[B]y adopting a multi-paradigm approach that integrates insights from the
OB and OT literatures to study multiteam systems … we suggest that IFD
and vertical coordinated action are intertwined in a complex manner, with
vertical coordinated action determining whether IFD’s advantages (increased
horizontal coordination) or disadvantages (decreased aspirational behavior)
will prevail” (de Vries et al., in press).
In commenting on the finding of the use of semistructures, Brown and
Eisenhardt (1997: 15) engaged the literature to note: “Another reason may
be that these limited structures help people to make sense of a fast-changing
environment. In such environments, it is easy to become confused, make
mistakes, and fall behind. Previous research indicates that structure helps
people to make sense of change. For example, Weick’s (1993) …”
“This study focuses on emotional contagion, ‘a process in which a person or
group influences the emotions or behavior of another person or group through
the conscious or unconscious induction of emotion states and behavioral
attitudes’ (Schoenewolf, 1990: 50), in particular, the contagion of every-day
moods in work groups. … [which] will lead to greater cooperativeness on both
an individual and group level” (Barsade, 2002: 646, 651).
Compassion organizing refers to “when individuals in organizations notice, feel,
and respond to human pain in a coordinated way” (Dutton et al., 2006: 59).
Donaldson et al.,
2013; Pentland,
1999; Whetten,
et al., 2009
Kilduff, Mehra,
& Dunn, 2011;
Thompson, 2011
Key Cites
Moving up the ladder of complexity can provide
abstraction necessary for a meta-paradigm
perspective, whereas moving down the ladder of
complexity provides a more concrete perspective
of the phenomenon.
Shifting the way a theorist conceptualizes the
nature of phenomena (ontology) can provide a
new perspective from which to theorize but also
requires a corresponding shift in epistemology.
Identifying and naming a core construct(s) helps
to separate the phenomenon of interest from the
mass noise of everyday experience and prior
Narrative setting
Labeling constructs
Main Characters
Table 2
Building Stories Through Conceiving and Constructing Theories
Analogically connecting concepts from target
and source domains at a structural level,
transferring concept information between
domains, and blending the concepts provides for
an emergent understanding of both domains. For
anthropomorphizing the source is human and the
target is nonhuman.
Provides a basis for transforming constructs and
relationships in both the target and source to
generate new insights for both literatures.
Flexibly and responsively deploying whatever
research strategies, methods, or empirical
materials at hand and assembling these
knowledge elements in unique combinations to
generate fluid constructs for theorizing.
Combining contextual, structural, and strategic
factors to offer ideal types based on the same set
of dimensions and making explicit the weighting
of those dimensions enables theorists to explore
multiple patterns.
Posing problem statements, making conjectures on
solutions to the problem, trialing conjectures, and
selecting and retaining those that show promise
enable the theorist move through disciplined
imagination to build a theory.
Metaphor and
Plot and theme
Thought experiments
Main Characters
Doty & Glick,
1994; Fiss, 2011;
Payne, 2006
Boxenbaum &
Rouleau, 2011;
Denzin &
Lincoln, 1994
Oswick et al.,
2011; Zahra &
Newey, 2009
Cornelissen, 2005;
Morgan, 1996;
Shepherd &
Sutcliffe, 2015
Davis, et al., 2007;
Folger & Turillo,
1999; Weick,
Key Cites
“One could imagine a thought experiment in which there are two groups and
the average individual attributes conducive to creativity (creative personality,
experience in creative work, etc.) of one team is greater than the other; another,
similar sort of question might concern whether groups starting out with a
more creative initial ‘working idea’ require even more adherence to these
coordinative interactions. We think these differences would matter in our
model, but perhaps in a counterintuitive way” (Harrison & Rouse, 2014: 1278).
“The meaning … is that the act of managing is framed as involving
improvisation and as offering considerable degrees of freedom to managers
within organizations in interpreting, expressing, responding, and performing
in a given situation. … This image is stretched even further by writers …
who suggest that managers in fact ‘author’ their own script; they become
the writers and play-wrights of their own actions and are fully accountable
for them. … This image of organizational life that is evoked through the
‘organization as theatre’ metaphor underlines …” (Cornelissen, 2004: 716).
“That modern work organizations either ‘have’ or might ‘be like’ a culture
similar to other human groups requires us to draw connections that mirror the
clan and workgroup. … The modern workgroup and the clan then yield the
richer idea of organizational culture, which can then be tested in relation to
other referents (motivation, morale, performance, etc.)” (Oswick et al., 2011:
“The building blocks for organizations come to be littered around the societal
landscape; it takes only a little entrepreneurial energy to assemble them into a
structure. And because these building blocks are considered proper, adequate,
rational, and necessary, organizations must incorporate them to avoid
illegitimacy. Thus, the myths built into rationalized institutional elements
create the necessity, the opportunity, and the impulse to organize rationally,
over and above pressures in this direction created by the need to manage
proximate relational networks” (Meyer & Rowan, 1977: 345).
“On the basis of an interdisciplinary literature review, Part I introduces four
basic types of process theories that explain how and why change unfolds
in social or biological entities: life-cycle, teleological. dialectical. and
evolutionary theories. … [W]e will call them motors— to explain how and
why changes unfold. Part II arranges these four ideal-type process theories
into a typology by distinguishing the level and mode of change to which each
theory applies. Part III considers how the typology is useful for understanding
a variety of specific theories of change processes in organizations” (Van de
Ven & Poole, 1995: 511).
Table 2 (continued)
68   Journal of Management / January 2017
A related strategy is to borrow a construct from a related discipline. Thus, in organizational theory, population ecologists borrowed words like niche and species from the adjacent
field of evolutionary biology (Freeman & Hunnan, 1989; Hannan & Freeman, 1977). While
a term from a related scientific discipline partially addresses the issue of a lack of definitional
precision associated with using everyday language, it does not completely resolve the problem of surplus meaning. Population ecology, thus, has been soundly criticized for using terms
like species, which has a much more precise meaning when applied to living organisms (i.e.,
capable of interbreeding and producing a viable offspring) than it does when applied to organizations. As Whetten, Fellin, and King (2009) observe, borrowing terms from other disciplines often introduces more confusion (in levels of analysis, boundary conditions, etc.) in
understanding a phenomenon than clarity. An alternative approach is to create a completely
new term to describe the phenomenon of interest. A useful example of this in management
theory is Weick’s use of the term sensemaking, which is a portmanteau of preexisting common terms but, as a result of Weick’s theorizing, has acquired a unique and highly specific
Regardless of the technique used, identifying and naming constructs is an essential part of
theorizing because constructs are a source of agency or causality. That is, greater clarity in
describing constructs and their relationship to the phenomenon of interest helps to clarify the
motivations or causal relationships in the theoretical argument (Suddaby, 2010; for other
aspects of rigor on theory building, see Donaldson, Qiu, & Luo, 2013). Clearly defined constructs in theory require precise definitions and specific boundary conditions or contexts in
which they do or do not apply. Constructs help the reader understand a theoretical argument
because if they are accurately captured, the reader can quickly grasp their history, their motivation, and the implications of their role in the causal relationships that the theorist is presenting. We note, however, that there are limits to construct clarity. As Kaplan observes, the
process of enhancing definitional clarity inevitably produces even finer-grained distinctions
that fall outside our understanding. The “more discriminations we make, the more opportunities we create for classification errors between borderlines” (Kaplan, 1964: 65).
Choosing a Perspective for Theorizing: Determining the Narrative Setting
All stories occur in a narrative setting—namely, a time and place within which events
occur. In a way, the setting becomes as important in explaining causality as the broad conflict
that defines the story and the motivations of the central characters. Skilled storytellers understand that context is not merely a backdrop but can also play a determinative role in their
argument; it is essential both to the credibility of the theoretical argument and to the reader’s
appreciation of the causal logic of the theory, and by shifting the context, the theorist may
open up new conceptual terrain. In this section, we review a range of strategies used by theorists to adopt new perspectives by adjusting the philosophical setting within which the theory
is presented, namely, shifting ontology, shifting the position on the ladder of theory complexity, shifting back and forth between data and theory, and shifting level of analysis.
First, shifting ontology can provide a new perspective. Scholars often adopt a specific theoretical lens such that one philosophical perspective dominates a particular research topic, or
the research topic is bifurcated by streams of research that progress in parallel based on their
different philosophical underpinnings (e.g., research anchored in either a structural realist or a
Shepherd, Suddaby / Theory Building   69
social constructivist perspective; Hassard, 1993). Importantly, rather than a theorist choosing
one philosophical approach to use consistently, he or she can use an ontological shift to generate creative insights for the development of midrange theories. An ontological shift refers to
“changes in the ontological emphasis that maintain epistemic-ontological alignment”
(Thompson, 2011: 755), with ontology referring to the nature of phenomena and epistemology
referring to the nature of knowledge about the phenomena (Gioia & Pitre, 1990). It is important when engaging an ontological shift to also change the epistemology; otherwise, it can lead
to ontological drift, in which the construct is compromised (Thompson, 2011).
One example of shifting ontology for theorizing is shifting from an entity-based ontology
to a process-based ontology (or vice versa). Theories in management have focused more on
entities (e.g., organizations, entrepreneurs, and institutions) than processes (e.g., organizing).
For example, take the notions of entrepreneur and institution (i.e., entities) and start to think
about them in terms of processes, such as entrepreneuring and institutionalizing, respectively. Such a theorizing approach does not replace the entity construct but involves a complexification of the established construct and can lead to different research logics of action
that are reflective of different assumptions and orientations, which tackle different research
questions (Kilduff, Mehra, & Dunn, 2011; Morgan, 1980).
Second, conceiving and constructing theory can also be facilitated by moving up and/or
down the ladder of theory complexity. Ofori-Dankwa and Julian (2001) emphasize two
dimensions in establishing the level of theory complexity: (1) relative endurance, which
captures the extent to which the core concepts of the (proposed) theory are represented as
relatively stable (high endurance) or unstable (low endurance); and (2) relative exclusivity,
which captures the extent to which a single core concept (high exclusivity) or several core
concepts (low exclusivity) form the model. As a 2 × 2, this sets up four levels of theoretical
complexity: Level 1 (simple complexity) involves high endurance and high exclusivity to
offer theories of contingency, Level 2 (medium complexity) involves low endurance and high
exclusivity to offer theories of cycles, Level 3 (high complexity) involves high endurance and
low exclusivity to offer theories of competing values, and Level 4 (very high complexity)
involves low endurance and low exclusivity to offer theories of chaos.
Indeed, abstracting one’s theorizing (by moving up the ladder of theory complexity) can
provide the basis for a meta-paradigm perspective that allows disparate approaches to theory
building to be considered together as a way to bridge across paradigm boundaries (Gioia &
Pitre, 1990; for an epistemological approach [evolutionary naturalist] to unify diverse perspectives, see Azevedo, 2002). As Kaplan (1964) observes, theorists move from observable
indicators (i.e., the “individual”) to higher levels of abstraction that involve unobservable
categories or concepts (i.e., “social classes” or “society”). The process of building theories,
as Stinchcombe (1968) notes, requires skillful abstraction, or selectively moving up or down
the ladder of abstraction to create propositions (generated at higher levels of abstraction) or
operationalize hypotheses (generated at observable levels of abstraction).
The abstraction is needed for the theorist to broaden her or his view (from one anchored in
the assumptions of one paradigm) to juxtapose, and perhaps link, previously different views
to provide a broader perspective of organizational phenomena (Gioia & Pitre, 1990; Lewis &
Grimes, 1999). Theorizing across paradigms may appear difficult given that each paradigm
has a different set of assumptions, but the boundaries between these paradigms are often
blurred (Bochner, 1985; Geertz, 1980) and can be usefully conceived as “transition zones”
70   Journal of Management / January 2017
(Gioia & Pitre, 1990). Through abstraction, the theorist can generate second-order concepts
(Van Maanen, 1979). Second-order concepts describe scientific understanding as opposed to
first-order concepts, which describe how people experience the phenomena. Second-order
concepts, as an abstraction of first-order concepts, facilitate the recognition of related or analogous concepts as the basis for a bridge across the transition zones of two or more paradigms
(Gioia & Pitre, 1990; Lewis & Grimes, 1999). A meta-paradigm perspective moves beyond
the “agree to disagree” approach of disparate paradigms to gain an understanding of why disagreement exists and to theorize on similarities and interrelationships to understand management phenomena, which broadens the “conception of theory and the theory-building process
itself” (Gioia & Pitre, 1990: 600; Lewis & Grimes, 1999). For example, Pfeffer and Fong
(2005) argue for theorizing that uncovers core, fundamental constructs and linking them to
build a broad understanding that explains a range of behaviors. Therefore, both abstraction
and complexification can serve as a basis for new theories (Thompson, 2011).
Third, moving back and forth between the empirical evidence and the literature helps to
build a theoretical story. Eisenhardt (1989) suggests that a theoretical narrative is best constructed through comparisons between multiple case studies. The theorist enters the field
with a clear research question (possibly one drawn from the literature or focused on elaborating specific constructs), carefully selects cases that build tension or contrast around the focal
research question (“theoretical sampling”), and identifies key patterns that match data with
theory to build “bridges from rich qualitative evidence to mainstream deductive research”
(Eisenhardt & Graebner, 2007: 25; see also Eisenhardt, 1989; Hallier & Forbes, 2004). An
alternative approach, suggested by Dyer and Wilkins (1991) places even greater emphasis on
the narrative elements of a single case study in which the researcher constructs theory by
moving between the thick description of data and the extant literature. In both approaches,
however, the theoretical narrative emerges as the result of abductive iteration between theory
and literature in an effort to address an “unmet expectation.” As Van Maanen, Sorenson, and
Mitchell (2007: 1149) observe, an unmet expectation is a mystery or a clue that, “like the dog
that did not bark in the fictional world of Sherlock Holmes,” motivates theorizing by requiring the research to construct a robust explanatory narrative by giving “primacy to the empirical world, but in the service of theorizing.”
Finally, building a story can be facilitated by changing assumptions through crossing levels of analysis. Klein, Dansereau, and Hall (1994) highlight three key assumptions underlying multilevel theorizing that should be made explicit—namely, (1) homogeneity, which
refers to group members’ being sufficiently similar on the focal construct such that they can
be categorized as a whole (i.e., the “group as a whole”); (2) independence, which refers to
group members’ being independent of the group’s influence and others in the group (between
individual variance); and (3) heterogeneity, which refers to individuals’ being nested within
the group such that the “group context is not only informative but necessary to interpret an
individual’s placement or standing in the group” (Klein et al., 1994: 202). Indeed, by theorizing across levels of analysis, we can gain a deeper understanding of the mechanisms at levels
of analysis different from those used in the initial theories or topics that explain the “why” of
existing relationships (and theories) (see also Shepherd & Sutcliffe, 2015).
In particular, Morgeson and Hofmann (1999: 251) highlight the multilevel nature of constructs in a collective context where collective refers to “any interdependent and goal directed
combination of individuals, groups, departments, organizations, or institutions.” Under such
Shepherd, Suddaby / Theory Building   71
collective contexts, constructs can exist at both the individual and group level and can be
explored in terms of their function—the causal output of the system (or component of the
system)—and/or in terms of their structure—the system of interaction among members of the
collective. Exploring the function and structure of collective constructs can facilitate theorizing on the emergence of, stability of, and changes in collective constructs. Emergence, stability, and change involve notions of time to which we now turn.
Set Time to Establish Boundary Conditions: The Story’s Event Sequence
The event sequence is the order in which events occur and brings together the different
components of the story. Although time is implicitly or explicitly a boundary condition of
most theories, theorizing can involve shifting the perspective on time to change the ontological nature of constructs and the relationships between constructs (George & Jones, 2000;
Zaheer, Albert, & Zaheer, 1999). Indeed, in Whetten’s (1989; see also Dubin, 1978) description of the criteria of theory—“what,” “how,” “why,” “who,” “where,” and “when”—the
“when” is a direct reflection of the importance of time in theory. Specifically, George and
Jones (2000) highlight how time can be used in theorizing by considering (1) how the past
and future can impact the present and how time can be experienced differently (i.e., subjective time) within or across individuals; (2) how time is aggregated into chunks, such as with
defined episodes (for different time scales see Zaheer et al., 1999); (3) how the duration of
periods can be categorized as periods of stability and of change, (4) how the nature of change
can be considered in terms of its rate (over time), its magnitude (e.g., incremental or discontinuous), and its pattern (e.g., frequency, rhythm, and cycles); and (5) how the interplay
between constructs over time can be reflected in mutual causation (e.g., positive or negative
spirals) and change intensity (Dansereau, Yammarion, & Kohles, 1999; Mitchell & James,
2001). For example, Corley and Gioia (2011) suggest theorists direct attention to future problems in order to anticipate problems and thereby inform future thought and action, generate
vibrancy, and ensure usefulness in a rapidly changing external environment. Such theorizing
(labeled prescient theorizing) is informed by either projective futurism—a sound theoretical
basis for arguing and predicting—or prospection—the use of informed projections into the
future to anticipate issues, act as if those issues are manifest, and then infer domains requiring attention or invention (Corley & Gioia, 2011: 25).
For theorists who consider time to develop process theories (as opposed to theories of
variance; Mohr, 1982), Langley (1999) offers different theory construction strategies, that is,
by (1) constructing a detailed story anchored through time (narrative strategy); (2) coding
qualitative incidents into predetermined categories for statistical analysis (quantification
strategy); (3) proposing and assessing alternate theoretical templates of the same events
using different theoretical premises (alternate templates strategy); (4) constantly comparing
sets of data to gradually build a system of categories that can be linked to explain the process
(grounded theory strategy); (5) graphically or otherwise visually displaying multiple representations of “precedence, parallel processes, and the passage of time” (Langley, 1999: 700)
(visual mapping strategy); (6) bracketing and labeling periods of an event and detailing the
continuities within that period and the discontinuities at or outside its borders (temporal
bracketing strategy); and (7) constructing global measures of a process as a whole to compare different processes (synthetic strategy).
72   Journal of Management / January 2017
Disciplined Imagination: Plot and Theme
The plot is what holds a story together (Jameson, 2001), makes it comprehensible (Garud
& Giuliani, 2013), and, along with the main character, provides coherence (Ibarra &
Barbulescu, 2010); the plot provides the discipline for the imaginative aspects of the story. In
a similar way, theorizing to create something new—a new explanation, new insights, and a
new story—not only requires imagination, but it also requires discipline. Theorizing as disciplined imagination can involve thought experiments—abstract hypothetical scenarios
(Folger & Turilo, 1999) or simulations—“a method for using computer software to model the
operation of real world processes, systems, or events” (Davis, Eisenhardt, & Bingham, 2007:
481)—as part of a process of artificial selection (Weick, 1989). Indeed, Weick (1989) notes
that when theorists build theory through imaginary experiments, their activities resemble an
evolutionary model of variation, (artificial) selection, and retention.
These processes of disciplined imagination begin with the creation of a research question
in the form of problem statements. Problem statements specify a need that requires a solution
and are formulated and posed by the theorist. The theorist specifies a problem to be solved
(explained), details assumptions that can be confirmed and disconfirmed, offers a set of concepts that can be connected differently, implies a plot that may be implausible, and asks a
question that has not been asked yet (Weick, 1989: 521; see also Davis et al., 2007). After
constructing problem statements, the theorist engages in thought trials—that is, trialing
(competing) conjectures of a solution to the problem statements (see also Kaplan, 1964;
Stinchcombe, 1968). Theorizing is enhanced by thought trials that are more numerous and
more diverse (heterogeneous thought trials will provide more information to inform the theorizing process) that facilitate progress in refining the conjectures. Finally, the theorist must
choose and use selection criteria for the thought trials. Theorizing is more promising when
the selection process consistently applies a set of criteria (Weick, 1989), when it activates
access to tacit knowledge through embodied or vicarious participation (Folger & Turillo,
1999), and when it invokes the related properties of a system’s interrelated links (Folger &
Turillo, 1999). Although thought trials can be conducted in the theorist’s head (or through
simulation software), knowledge production typically has a social component such that conjectures are tested when they are communicated to others (i.e., via stories) and receive feedback (Jacques, 1992; Weick, Sutcliffe, & Obstfeld, 2005).
The discipline of theorizing can come from metaphors including the specific case of
anthropomorphizing, from other forms of blending, from the knowledge resources at hand,
and from patterns in the form of typologies, to which we now turn.
Theorizing through an interaction metaphor (Cornelissen, 2005, 2006) starts with the
development of a generic structure that connects a source and a target domain such that the
theorist can begin to map the correspondences and transfer “instance-specific” information
about concepts between these domains. This provides the opportunity to elaborate on the
emerging story by blending the concepts of the source and the target, which provides for new
insights not only about the target but also about the source domain (Cornelissen, 2005, 2006)
(more on blending in the sections that follow). Specifically, metaphors can help theorizing by
(1) providing a vocabulary to “express, map, and understand” the complexity of a particularly phenomenon, which provides a more concrete basis for understanding (and discussing)
underlying constructs (Cornelissen, 2005: 753; Lakoff & Johnson, 1980; Tsoukas, 1991); (2)
encouraging an open-minded approach with “multiple ways of seeing, conceptualizing, and
Shepherd, Suddaby / Theory Building   73
understanding” the phenomena of interest (Cornelissen, 2005: 753); and (3) allowing for new
insights that may have previously been inconceivable (Morgan, 1980, 1983, 1996; Oswick,
Keenoy, & Grant, 2002).
Anthropomorphizing represents a special case of theorizing through metaphor.
Anthropomorphizing refers to “imbuing the imagined or real behavior of nonhuman agents
with humanlike characteristics, motivations, intentions and/or emotions” (Epley, Waytz, &
Cacioppo, 2007: 864). Shepherd and Sutcliffe (2015) highlight how anthropomorphizing has
been critical to the creation and development of many important management theories, including those of organizational knowledge and organizational identity. Anthropomorphizing can
be an effective theorizing tool when the theorist uses his or her rich understanding of himself
or herself and other people to (1) take a leap of faith to make a guess at an explanation of an
anomaly, (2) provide insights into the mechanisms underlying the “how” and the “why” of key
relationships and insights into organizing, and (3) facilitate sensemaking as well as tap into the
audiences’ knowledge of themselves and others as a communication strategy for sensegiving
to tell robust stories. Anthropomorphizing, as a tool for theorizing, provides the potential for
theorists to generate, build, and communicate creative theories of organizations and organizing as well as other nonhuman management entities or processes (and perhaps theories of
themselves). Moreover, it gives junior scholars the confidence to theorize.
Metaphor, at least in the interaction model of metaphor, involves blending but not all
blending involves metaphor. Oswick and colleagues (2011) offer four types of blending: (1)
orthodox domestic theory (i.e., narrow focus in terms of theoretical contribution and consumed largely with the domain of production) provides incremental extensions to a focal
subarea of management; (2) innovative domestic theory (i.e., broad focus in terms of theoretical contribution and consumed largely within the domains of production) “challenges
existing knowledge and ways of thinking but does so from an insider’s perspective” (p. 323);
(3) novel traveling theory (i.e., narrow focus in terms of theoretical contribution and consumed across domains) offers “quirky insights into non-management disciplines yet largely
reinforces, builds upon, or resonates with prior knowledge” (p. 324); and (4) radical traveling
theory (i.e., broad focus in terms of theoretical contribution and consumed across domains)
represents a “significant challenge to and departure from the contemporary and conventional
pre-existing insights in a particular discipline” (p. 322) but requires considerable “repackaging, refining, and repositioning” (p. 323) in order for it to be taken up by management scholars. It is important when using blending to theorize about how the generated insights impact
the source discipline (over and above the impact on the target discipline), potentially including how existing source theories need to be refined and boundary conditions need to be
reconsidered (see also Zahra & Newey, 2009).
While blending provides a basis for transforming constructs and relationships in both the
target and source literatures (i.e., bidirectional flow of information), bricolage largely combines subelements from a source discipline to application in management to create a unique
combination (i.e., unidirectional flow of information). Bricolage is an important theorizing
tool. Indeed, knowledge production can be conceptualized in terms of evolution, differentiation, and bricolage. Although evolution (i.e., knowledge accumulation through “trial and error
toward an increasingly robust view of the world”) and differentiation (i.e., attempts to “generate knowledge that is discontinuous with existing knowledge”) predominate in management
(Boxenbaum & Rouleau, 2011: 279-280), bricolage has considerable potential to be a source
of novel theories. In theorizing, bricolage refers to “the assembly of different knowledge
74   Journal of Management / January 2017
elements that are readily available to the researcher” into fluid knowledge constructs
(Boxenbaum & Rouleau, 2011: 281). This approach requires the theorist to be “flexible and
responsive … to deploy whatever research strategies, methods, or empirical materials, at
hand, to get the job done” (Denzin & Lincoln, 1994: 2). Indeed, perhaps bricolage’s role in
theorizing is more common than it seems because while scholars might use bricolage to theorize, they communicate the outcome of the process in terms of an evolution or differentiation
Boxenbaum and Rouleau (2011) propose that theorists engage bricolage by (1) focusing
on combining various elements (e.g., ideas, concepts, experiences) they have at hand rather
than engaging in endless search of the literature or creating a theory from “scratch”; (2)
choosing elements that are local (to the theorist) and sufficiently diverse such that their combination can provide novel (and hopefully useful) insights; (3) using common sense in selecting the items and combining them such that further theorizing can generate a coherent, broad,
and useful explanation of management phenomena; (4) remaining flexible and responsive to
new combinations by approaching the elements (to be combined) as fluid concepts and their
combinations as potentially transformative (in terms of new insights); and (5) reflecting on
how one is using (and/or has used) bricolage to theorize.1
Finally, typologies are another way of combining constructs; typologies offer a way to
theorize by representing complex explanations of causal relationships involving contextual,
structural, and strategic factors for explaining a focal outcome (Doty & Glick, 1994; Fiss,
2011). These explanations are not classification schemes—“systems that categorize phenomena into mutually exclusive and exhaustive sets with a series of discrete decision rules” (Doty
& Glick, 1994: 232; see also McKelvey, 1982; Pinder & Moore, 1979) for describing phenomena—but are complex theories (Doty & Glick, 1994). Theorizing through typologies
requires the theorist to make explicit her or his grand theoretical assertions (Doty & Glick,
1994: 235), define each ideal type, describe each ideal type using the same set of dimensions,
and make explicit the assumptions underlying the weighting of the dimensions (e.g., core and
peripheral elements; Fiss, 2011) that describe the ideal types (Doty & Glick, 1994). Typologies
can provide important insights for knowledge accumulation because they enable the theorist
to move beyond the linear to explore multiple patterns (Miles, Snow, Meyer, & Coleman,
1978), emphasize the importance of how multiple factors fit together to offer a more holistic
story (Fry & Smith, 1987; McKelvey, 1982), allow for equi-finality (i.e., organizations can
reach the same outcome [e.g., high performance] through alternate paths; Katz & Kahn,
1978; Payne, 2006; Van de Ven & Drazin, 1985), and offer a “form of social scientific shorthand” (Ragin, 1987: 149) for explaining multiple causal relationships (Fiss, 2011).
3. Evaluating a Theory: The Narrative Arc
Narrative arcs typically conclude with a resolution of the problem of the story and/or the
problem faced by the main actor of the story. Despite the importance of developing theories
and making a theoretical contribution, the resolution of the story (i.e., what constitutes a
theory) varies widely as does the interpretation of what represents a good story (i.e., a theoretical contribution). The range of understandings of what constitutes theory is, as Suddaby
(2014b) observes, a reflection of the wide variety of understandings of what theory should be
used for. Some (most perhaps) see theory as a means of accumulating knowledge. Others,
Shepherd, Suddaby / Theory Building   75
however, see theory as useful for legitimating some forms of knowledge over others. A third
group sees a powerful normative value in theory, less important in summarizing existing
knowledge than in directing the attention of a research community to explore issues of
importance for the future. In each case, however, some theories seem to be preferred over
others because of their narrative attributes (Van Maanen, 1995). Our interest in this section is
in reviewing the rhetorical attributes of successful theories and, more particularly, identifying
the narrative elements of what constitutes a contribution to theory. These attributes—in terms
of story completion, compelling story, and the next story—are illustrated in Table 3.
A theory can be conceptualized as a statement of concepts and their relationships that
specifies who, how, and/or why a phenomena occurs within a set of boundary assumptions
conditions (see Bacharach, 1989; Gioia & Pitre, 1990). The general purpose of a theory is to
organize (parsimoniously) and communicate (clearly) (Bacharach, 1989), and it does this by
offering a coherent explanation of a phenomenon, making assumptions and building on those
assumptions to logically derive predictions, offering conjectures that allow for refutation or
falsification, and testing (Shapira, 2015).
Although these attributes of the notion of theory are useful, it is not necessarily clear
whether the outcome of a specific piece of a scholar’s work is a theory. Sutton and Staw
(1995) acknowledge the difficulty in specifying an outcome as a theory and approach the
issue by specifying what theory is not: theory is not references to prior work, is not data capturing the phenomenon, is not a list of variables or constructs, is not a diagram with boxes
and arrows, and is not a set of hypotheses. Indeed, Bacharach (1989) also attempts to explain
what is not a theory by detailing how theory is not a description or the what of a relationship
absent the how, why, and when.
Weick (1995) largely agrees with Sutton and Staw (1995), and for that matter Bacharach
(1989), about what theory is not. However, he recognizes that it is rare to offer a full-blown
theory and that what scholars can often hope to do is contribute to knowledge by offering
their work as an interim struggle (Runkel & Runkel, 1984) for which the outcome can be
evaluated in terms of a continuum rather than a dichotomy (a theory or not). There is considerable comfort in thinking about theory as a continuum because it sets more realistic expectations about what is (or should be considered) a theoretical contribution. Therefore, while the
Sutton and Staw (1995) list of what is not theory is appropriate given the notion of theory as
a dichotomy, theorizing outcomes can be important as part of an emerging story and/or as an
input to further theorizing. That is, to the extent that theorizing as interim struggles informs
subsequent work, it is useful and salutary and perhaps a contribution worthy of publication
(despite not yet having reached the status of full-blown theory).
The question then becomes what represents a theoretical contribution. A theorizing outcome
can be considered to make a contribution to the extent that it bridges a gap between two theories
as a basis for explaining something between two domains (Bacharach, 1989) and generates
new insights that lead to a reevaluation of existing theories (Bacharach, 1989) that are useful
(Corley & Gioia, 2011; Kilduff, 2006; Whetten, 1989). Therefore, to be a contribution, the
theorizing outcome needs to be original and useful. In terms of being original, the theorizing
outcome should reveal something that we previously did not know (Corley & Gioia, 2011),
surprise us by making us reconsider something we thought we knew (Mintzberg, 2005; Rynes,
2002), and be sufficiently novel and/or counterintuitive (Corley & Gioia, 2011; Davis, 1971).
The theorizing outcome is useful to the extent that it offers scientific utility—facilitates
The next story
Compelling story
Original and
Theory as
Story Completion
To reflect, to take account of the research process, and to recognize the situated nature of
knowledge and knowledge creation, theorists can use a different perspective, voice, and
positioning, and can problematize the process and the outcome to stimulate subsequent theorizing.
To be a contribution, the theorizing outcome needs to reveal something that we previously did
not know, surprise us by making us reconsider something we thought we knew, and/or be
counterintuitive but also advance conceptual rigor and addresses problems facing practitioners.
A theoretical contribution is greater for those that offer a broader and simpler theory that is explicit
about the underlying mechanisms and has fewer alternate explanations.
While a theory must be sufficiently novel to capture attention, it must be similar enough to what
is known to be comprehensible and theorists can do this by imbuing novelty with meaning to
provide both novelty and continuity.
Although a theory represents a statement of concepts and relationships that specifies who, how,
and/or why a phenomena occurs within a set of boundary conditions (where and when) and while
there are some indications of what a theory is not, there remains debate about the threshold upon
which a work becomes a theory.
By focusing on theorizing, rather than theory, research is considered on a continuum of “theory”
that acknowledges the emerging nature of the story and the interim struggles on the way to
advancing knowledge.
Table 3
Narrative Arc to Achieve a Theorizing Outcome
Alvesson et al., 2008;
Michailova et al., 2014;
Pentland, 1999
Shepherd & Sutcliffe, 2011;
Thagard, 1989
Locke & Golden-Biddle,
1997; McKinley et al.,
Corley & Gioia, 2011;
Davis, 1971; Pfeffer, 1993
Langley, 1999; Shepherd
& Sutcliffe, 2015; Weick,
Bacharach, 1989; Gioia
& Pitre, 1990; Sutton &
Staw, 1995; Whetten, 1989
Main Cites
Shepherd, Suddaby / Theory Building   77
advances in conceptual rigor and specificity and/or enables operationalization and testing—or
practical utility—can be applied to the problems facing practitioners (Corley & Gioia, 2011)
(i.e., problems that matter; Pfeffer, 1993). Therefore, while the theory must be different from
received wisdom to warrant a second look, it must be similar enough to what is known to be
comprehensible (McKinley, Mone, & Moon, 1999). By linking a theory with what is already
known, the theorist imbues novelty with meaning and thus sets up a dynamic tension and interplay between novelty and continuity (McKinley et al., 1999: 638).
Building on the importance of coherence to a theory contribution (Azevedo, 2002),
Shepherd and Sutcliffe (2011) offer the following principles for assessing the theoretical contribution of theorizing outcomes: (1) A broader theory is a better theory. A broader theory is
one that explains more facts and, in doing so, provides a more coherent explanation than one
that explains fewer facts. The breadth of a theory is the range of phenomena encompassed by
the theory (Bacharach, 1989: 509). (2) A simple theory is a better theory. A simpler theory is
one that requires the fewest assumptions (Read & Marcus-Newhall, 1993) and is more parsimonious. A theory is less parsimonious when factors can be deleted because they add little
additional value to our understanding (Dubin, 1978; Whetten, 1989). A good theory finds a
balance between being overly exhaustive and overly exclusive (Feldman, 2004: 566). (3) A
theory with explicit mechanisms is a better theory. Mechanisms offer an explicit explanation
for proposed relationships (Davis & Marquis, 2005). Anderson and colleagues (2006: 102)
define social mechanisms as “theoretical cogs and wheels that explain how and why one thing
leads to another.” In describing a good theory, Whetten (1989) suggests the theory must offer
an explanation of why. (4) A theory with fewer acceptable alternative explanations is a better
theory. The evaluation of a theory is partly comparative in that a judge is partially influenced
by the availability of alternate explanations and how good they are (Read & Marcus-Newhall,
1993; Thagard, 1989). A better theory is one that loosens “the normal science straightjacket”
(Daft & Lewin, 1990) to offer something new (Feldman, 2004; Mone & McKinley, 1993) that
challenges and extends existing knowledge (Davis, 1971; Whetten, 1989).
However, a theory (or other form of theorizing outcome) itself may stimulate additional
theorizing. For example, theorists can be reflexive, that is, reflect and take account of the
Purchase answer to see full

error: Content is protected !!