All American Career College Science Worksheet

Hell, I hope you are doing well. I attached the questions word document that you can find the instructions and my note highlighted in yellow; maybe it could help answer the question.

I’m attaching my answer for the first and third questions with the model….maybe give you some hint of your answer or as a guide.

Please answer and follow the instructions carefully. Also, please keep it simple, have no complicated sentences or vocabulary, and be careful with the plagiarism

I attached the two books and the article besides the word documents.

Please let me know if there are any questions.

Thank you a lot.

Requirements: 5 pages and 300 words per question.

you may use reference sources, including your text, notes, and other books and papers.
Restrict your answer to 5 pages or less.
The final exam folder for this course contains the paper, “Work Motivational Challenges
Regarding the Interface Between Agile Teams and a Non-Agile Surrounding
Organization: A Case Study,” by Gren, Torkar, and Feldt.
Based on data from one interview and one focus group, the authors draw conclusions
about introducing agile teams into traditional (especially, stage-gate driven)
organizations. They suggest that future research on this topic should include quantitative
studies.
According to the authors, the main contribution of their study is “that in order to combine
agile teams with a stage-gate/traditional project organization the organization
surrounding the teams must understand how the agile team is different and adapt their
feedback to their way of working. If they do not, the agile team’s earlier increased job
motivation will somewhat decrease” (p. 14).
For this question, design a quantitative study that is the best follow-up you can design
to the Gren, Torkar, and Feldt paper.
·
Assume a scientific realist perspective, rather than a purely positivist one.
·
Assume that you will have all the necessary resources needed (money,
time, access, staffing…) to accomplish your design.
·
Do not assume, however, that you will be allowed to manipulate anything
about the organizations or people that might be in your study.
·
Center your study on the main contribution of the Gren, Torkar, and Feldt
study, mentioned above.
Do each of the following things:
1. Draw a conceptual model for your new study.
a.
Use the concepts (or variables) from the paper related to the main
contribution.
b.
If you use other concepts, from the paper or not, provide a
justification for each.
c.
Show the relationships among the concepts in #a-b above.
d.
Use appropriate notation (i.e., ovals, arrows, +/- to show directions
of influences).
The PLS-SEM model is the best used for exploratory study. This model includes a
structural model and a measurement model. I include the fifth latent concept in the model
regarding the contribution and the finding. The latent is agile practice, organization
management, feedback, satisfaction, and the last latent motivation, which is the target. In
the outer model of exogenous latent variables. The agile team participates in the
interview and supports each other to finish the task with their previous knowledge, which
leads to team group satisfaction. The structural model contains four latent: organization
management, satisfaction, and feedback. So we can interpret the organization could
obligate the team members to follow their own methods or approach and not apply agile
in the task, which impacts the efficiency and effectiveness. Thus, this may affect their
satisfaction, which increases if they work in the agile team, and then face the challenge
of their motivations. In contrast, the team member may stress the feedback from the
organization, which will result in dissatisfaction and then will impact their motivations.
We can see the target latent in the outer endogenous latent variable.
2. Explain in detail how you could include the concepts that cannot be measured directly
in your model.(300 words)
(book: A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM))
(chapter 1)(construct or latent) (principles of structural equation modeling) page 12
– both way if you include construct or latent from the current article or from other articles
please explain why we use it. Justification for each
3. Explain in detail how you could acquire measures for the concepts in your model that
can be measured directly? How could you determine if they are good?(300 words)
The purpose of indicator measurement is to have reliable data and creating or including
the indicators in the construct is to get perfect explanations. When I designed the
indicators I relied on a theoretical approach and previous literature [1], [2], [3], [4] which
is related to the main article in order to indicate the acceptable indicator to consider in the
model. For example, 1- The agile practice has three indicators: collaboration with the
team, structured, and unstructured. 2- Organization management has three indicators:
management support, agile methods, and decisions. 3- Satisfaction described by the
expert with previous work, positive feeling, and workload. 4- Feedback has two
indicators: unsynchronized, odd of response, and meaningful feedback. 5- Motivation is
the target latent with three indicators: fitting the work, accomplishing the task, and
adapting the feedback.
There are two kinds of models that explain the indicator: the reflective measurement
model and the formative measurement model, and each of them has a procedure to
evaluate. If the indicator is reflective, thus, we need to evaluate: the indicator
reliability, which is testing the outer loading. We must be careful when we test the
indicator and be significant. There are three test cases of the outer loading: first, If the
outer loading is less than 0.4, we eliminate the indicator. Second, if the outer loading is
higher than 0.7 then we retain the indicator. Third, if the outer loading is between 0.4 and
0.7 then we test the internal consistency and convergent validity if it meets the thresholds
then retain or eliminate the indicator if not. The second evaluation is internal
consistency. In this evaluation, we use Cronbach’s alpha which is a statistic test to
provide the test score of the reliability. The third evaluation that could use for analysis of
the outer loading is convergent validity. This evaluation is a degree of a measure
associated positively with other reflective or formative of the same latent. Also, it can
assess by the average variance extracted (AVE), which needs to be above 0.50 and not
related to the same latent. The fourth evaluation is discriminant validity which is
distinct between two latent and recommend to not be related. We could test it by using
the Fornell-Larcker criterion and heterotrait-monotrial ratio (HTMT). The HTMT is the
ratio or the mean of all the indicators. If the value of HTMT is one that is mean a lack of
discriminant validity. Most of the research used a threshold value of 0.85. So, If the
construct is less than the threshold value, it is reliable for discriminant validity. The
bootstrapping help to assess the HTMT if it is different from the threshold for all latent.
We could solve the lack of discriminant validity by following the two ways: first,
increase the latent’s average monotrial-heteromethod correlation. This could be done by
eliminating items that have a lower correlation with the same latent and splitting the
construct into homogeneous sub latent. Second, decrease the average of heteromethodheterotrait correlation. Also, this could be by eliminating any items that have a strong
correlation with other items and reassigning the indicator to other latent.
4. Explain in detail how could you acquire the survey(s) you would need to collect your
data? How could you determine if they are good?(300 words)
J&J book (chapter 13: classical test theory) (chapter 14: construction theory of selfreport)
Explain in detail what quantitative data analysis technique(s) you would use to test your
model. (300 words) (book:A Primer on Partial Least Squares Structural Equation
Modeling (PLS-SEM))
5.
Explain in detail how you would collect data. (300 words)
(book: A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM))
6.
Explain in detail how you would interpret different likely results. (300
words)
(book: A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM))
(chapter 4)
arXiv:1904.02439v1 [cs.SE] 4 Apr 2019
Work Motivational Challenges Regarding the
Interface Between Agile Teams and a Non-Agile
Surrounding Organization: A case study
Lucas Gren
Richard Torkar
Robert Feldt
Chalmers University and
Gothenburg University
Gothenburg, Sweden 412–92
Email: lucas.gren@cse.gu.se
Chalmers University and
Gothenburg University
Gothenburg, Sweden 412–92 and
Blekinge Institute of Technology
Karlskrona, Sweden 371–79
Email: richard.torkar@cse.gu.se
Blekinge Institute of Technology
Karlskrona, Sweden 371–79
Email: robert.feldt@bth.se
Abstract—There are studies showing what happens if agile
teams are introduced into a non-agile organization, e.g. higher
overhead costs and the necessity of an understanding of agile
methods even outside the teams. This case study shows an
example of work motivational aspects that might surface when
an agile team exists in the middle of a more traditional structure.
This case study was conducted at a car manufacturer in Sweden,
consisting of an unstructured interview with the Scrum Master
and a semi-structured focus group. The results show that the
teams felt that the feedback from the surrounding organization
was unsynchronized resulting in them not feeling appreciated
when delivering their work. Moreover, they felt frustrated when
working on non-agile teams after have been working on agile
ones. This study concludes that there were work motivational
affects of fitting an agile team into a non-agile surrounding
organization, and therefore this might also be true for other
organizations.
Index Terms—Agile Development Processes, Large Organizations, Work Motivation, Empirical Study
I. I NTRODUCTION
There are many success stories of companies that have
transitioned to an agile way of working. In complex projects
where a clear goal and finish line are hard to define and everchanging, a more flexible managerial style is often needed
[1]. With increasing success in “saving” projects in crisis and
with these projects being of different sizes and having diverse
circumstances there was a more widespread acceptance and
use of agile approaches to software development. Over time
the concept also saw increasing use as a more general approach
to project management. Agile thinking and methods are not an
isolated phenomenon though. According to [2] several books
covering different management schemes and theories have
been written that relates to and touches on the ideas underlying
agile project management practices, such as: Critical Chain
Theory [3] and Lean Production [4].
The benefits of introducing agile methods in organization
have been proven to be mostly positive for many organizations,
e.g. [5]. It has also been shown that job satisfaction increase on
agile teams [6]. However, motivational aspects of agile teams’
interface to a surrounding non-agile organization have not been
found. This study aims to show an example of what could
happen to agile team members’ motivation when working on
an agile team in a larger non-agile environment. Therefore,
the research question is “is there job motivational aspects
regarding the interface between agile teams and a non-agile
surrounding organization?”.
II. L ARGE O RGANIZATIONS
AND
AGILE M ETHODS
A. Traditional Project Management
In traditional project management, trade-offs are often made
between time, cost, and quality, i.e. it is impossible to prioritize
all three [7]–[9]. In order to choose from different projects
many organizations select according to a set of financial
decision methods. These are often simply based on cash
flows (which has evident drawbacks; for example, how to put
monetary value on other resources [10]), but are widely used.
a) Net Present Value: The most common method is the
Net Present Value (NPV) approach. This method is based
on the assumption that money is worth more today than in
the future (the time value of money). This means that future
earnings are worth less today so its value reflects a discount.
Therefore, this rate is referred to as a discount rate r. This
means that the sum of all cash flowsPdiscounted for today is
N
Cn
the present value of a project: PV = n=0 (1+r)
n . Where Cn
is future value for the investment at year n (and C0 is present
day). All the present values of all the cost and earnings for
the project is thereby calculated. The Net Present Value is
then: NPV = PV(benefits) − PV(costs). This means that
if the NPV > 0 the project is worth running. An interesting
fact is also that a firm’s total value is the NPV of all assets
in it [11]. One critique against the NPV approach is that
it assumes only one decision point in the beginning of the
project. A part of a solution could be a Real Options Approach.
In this case, different paths and cash flows are calculated and
weighted according to their probabilities [11], [12]. This is
also connected to the stage-gate methods described below.
The above approach is considered to be a good way to get
an overview of project activities and is also fairly simple to
draw and understand. The simplicity will bring disadvantages
such as the difficulty to make updates when many changes are
needed, and they do not help in optimizing resource allocation.
That in itself would imply that they could bring a false sense of
certainty about the project, often connected to time estimation.
Time estimation is affected by many different factors. One
of them is learning effects that can be described with a learning
curve: Yx = Kxn where x is the number of times the task
has been carried out, Yx is the time taken to carry out the task
the xth time, K is the time it took the first time, and n is
ln b
ln 2 where b is the learning rate [10], [13]. In addition, one
intimate factor to consider when performing time estimation
is risk.
There is a basic stage-gate system for splitting a project
into parts (or stages), as defined by [14]. This way a project
must pass through a gate before proceeding to the next stage.
The purpose is to solve problems where they pop up and not
to pass them on to the next stage. A disadvantage is that a
new part of the process cannot start before the previous one
is done. Concurrent engineering, i.e. to make the processes
overlap, can solve this partially. The basic idea when planning
a project would then be to break it down into small tasks. This
is often done through a work breakdown structure. The second
step is to construct a time plan according to this structure in
order to estimate how long time the project will take.
There is a set of techniques to conduct Risk Management,
but they should all include identification, quantification and
mitigation. In order to quantify the risks one can assess the
severity, hide-ability, and likelihood of a risk in order to get
a Risk Priority Number (by simply multiplying these scores).
Other, more quantitative, methods are expected value, Monte
Carlo simulation, and PERT. The latter is a simple way to add
optimistic and pessimistic times to all estimates in order to
more accurately get a time approximation that is not based on
a single guess. This adds probability to the time estimation
process, which will, hopefully, make it more realistic [15].
B. Agile Project Management
The basic idea of agile project management is that complex
projects need to combine the traditional approach to managing
projects and the need to be able to respond to change. The
agile community has, thus, defined a set of principles that
they summarize in The Agile Manifesto [16]:
• Individuals and interactions over processes and tools.
• Working software over comprehensive documentation.
• Customer collaboration over contract negotiation.
• Responding to change over following a plan.
Many customers have business needs that change over time,
reflecting not only new needs but also the need to respond to
a change in the marketplace. There are many agile practices
such as eXtreme Programming (XP), Crystal, and Scrum,
which try to take this into account. In Scrum the project
has a prioritized backlog of requirements and use iterative
development (called ‘sprints’) to get basic working software
for the customer to view as soon as possible. Scrum uses selforganizing teams that get coordinated through daily meetings
called ‘scrums’. Agile development, in general, is customerfocused, which means that the customer is preferably on site.
This means that the project is not strictly planned up front, but
changes continuously throughout the project. Instead of having
activities planned exactly the project maintains a flexibility
that is needed in order to rapidly respond to change. The
managerial culture of agile methods is trust, commitment,
teamwork, equality, and fair treatment. This means that agile
methods will probably work best in flat organizations and have
aligned decision-making on all levels [17]. Further, agility
must be present at all levels including the strategic one [18].
The idea is to have evidence-based decisions, goal-focus (with
change built in), independence with responsibility, and longterm thinking also known as sustainable pace (i.e. a 40-hour
workweek). The manager of an agile team tries to generate
group effectiveness by being a facilitator and not a supervisor,
and transparency is key for this process to work [19], [20].
C. Agility and Discipline
In software engineering the traditional approach to software development projects is usually considered to be ‘PlanDriven’. These methods come from the systems engineering
and other disciplines, and were established to coordinate large
inter-operating components. Software does not function as
hardware and, therefore, different standards were introduced.
The basic assumption is that software engineering is a process
of formal mathematical specification and verification. The
process is divided into different steps (i.e. a waterfall), which
are thoroughly documented. The process is standardized, and
incrementally improved to control and manage the work-flow
[21]. When changing to an agile method, where cooperation
and self-organizing team are central, some aspects of the
modern workplace might cause problems. If group members
are unable to, e.g., be physically present during meeting,
the aspect of human interaction becomes harder to achieve
and problems concerning communication, culture, trust, and
knowledge management appear [22]. There are also some
indications that people that does not have programming responsibilities in a large organization think that agile methods
is unsuitable in general [23]. There is also an aspect of
integrating flexibility in fixed and large organizations. Agile
methods can give a traditional stage-gate model a powerful
micro-planning tool and increase the change response time.
If the whole organization has not embraced the agile principles, an agile team that adapts to a stage-gate system can
synchronize their development with other teams and functions
of the organization. In order to make this feasible, the agile
team must be prepared to interface with the traditional stagegate system around it. The important part is that the team
is aware of these extra overhead costs. Agile methods are
generally more accepted by team members and more feared by
management. However, in order to make this work a universal
acceptance in the organization is much needed [24]. However,
it has been some evidence showing that agile teams have
higher job motivation than non-agile teams [6].
III. M ETHOD
The methodology used for this study consisted of an interview with the Scrum Master and a focus group with two teams
participating.
A. Case and Subjects Selection
The teams in this study were two teams with the same
Scrum Master at Company X in Sweden. Company X is a
part of a larger firm, which provides world-wide supply chain
expertise to a set of automotive companies. The IT part is, of
course, essential for the company to function. Many organizations, independent of field, need an efficient IT department
to provide good solutions for the whole organization. The
organization took a decision to implement agile methods and
was conducting a first pilot study to later diffuse the methods
to more parts of the organization.
The teams that were a part of this study had the task of
developing an extension of a corporate software system used
for supply chain management. In their work process they
integrated agile methods and Scrum specifically. The reason
why this case is from software engineering is that they have
the most experience with agile methods and were easier to
find. This software project included many teams, but two
of these teams were using Scrum and had the same Scrum
Master. The groups were a mix of business and programming
focused employees and external resources. The reason for this
mix was to assert that the business aspects of the project
were considered and to create a method that more areas
in the organization could use. Many of the team members
had therefore management tasks. Since there were unclear
separation between the two teams and the fact that they had the
same Scrum Master we chose to meet both teams collectively.
B. Data Collection Procedures
The first contact with the company was via an unstructured
40-minute interview the Scrum Master of these new agile
projects. During the interview one researcher were taking notes
carefully. The Scrum Master then set up a meting inviting
all members from both teams (N = 23). A subset of these
team members attended the meeting/focus group (N = 10).
The team members were informed that they would evaluate
their new process in a focus group with a researcher from
university. We had a set of questions to start the discussion
(semi-structured group interview/focus group), however, the
team had a lot to say about their new ways of working and its
connection to the rest of the organization. The topics covered
were:
• The teams’ experience with/opinions of their new agile
process.
• A comparison with their other current projects.
• Differences between this project and others they have
experienced.
One researcher participated during the one-hour focus group
and carefully wrote down what being said. The interviews
were not recorded since we wanted participant to be able to
speak as freely as possible regarding their emotions connected
to their participation on the team. The tradeoff is then, of
course, that we cannot say exactly how many times each
individual agreed on a topic lifted by one of their colleague.
The researcher who participated in the focus group wrote down
aspects the team focused on during the session instead.
C. Analysis Procedures
After both the interview with the Scrum Master and the
focus group the notes were carefully reviewed and summarized
by one author. The summaries were thematically analyzed
only keeping statements regarding work motivation. After
this, the statements were categorized, and compared to other
research. For example, unsynchronized feedback loops were
mentioned by several individuals and no other participants
expressed disagreement. Therefore this aspect was interpreted
as important and presented below.
IV. F INDINGS
A. Summary of Interview with the Scrum Master
The Scrum Master of the two teams describes the system
they are developing and the first enterprise system project so
far for them. The purpose is to integrate this new system
into the rest of the organization and the system is safetycritical. The organization traditionally has a stage-gate project
management method that is very strict. This framework is fixed
and they have to adapt to it and deliver what is needed at
certain milestones. Both these milestones and a budget for
the whole project must be predefined. The idea with agile
methodology is to work agile in between the gates at different
stages. They use a plug-in iteration process of agile that is not
exactly what they expressed that they wanted in the beginning
of the project. The business part of the project had been going
on for half-a-year already, and they have two-week sprints
with systems specifications to each sprint. The total amount
of sprints is nine, and they have a meeting at day five in
every sprint. The project uses a more strict way of writing
requirements and they do not apply user stories. The get their
requirements from the product owner, and this person decides
on the requirements and their priority. The get the requirements
documents to the teams by a standard called “Business Rules
Description”. The got a prototype up and running fast with
basic functionality.
B. Summary from the focus group
Some members expressed stress connected to the feedback
system from the surrounding part of the organization. If
they had struggled to reach a deadline internally within the
group, the effort was not recognized by other parts of the
organization since they had other milestones to follow. They
would have preferred to stop and celebrate somewhat and
then move on. At other times, they received positive feedback
from managers without them being even close to a delivery.
This was described and odd and unsynchronized. Then the
members had a discussion about how working agile had helped
them in their group development process. The Scrum practices
had given them a forum and a place to discuss solutions and
conflicts on a regular basis before they become more infected.
The members who were not 100% dedicated to the team but
had other concurring projects said they really felt a difference
between the two. The other projects felt slow and unresponsive
and they had gotten used to rapid responses and quick progress
with issues. They all agreed that job satisfaction was higher
for them when working on the agile team. They also compared
their result to another non-agile team, and stated that they were
way ahead of them considering what they had delivered.
V. D ISCUSSION
To use agile methods instead of traditional project management has its advantages [19], [25]. It was also mentioned by
the focus group that the agile work group had better results
than other groups within the company. However, they had to
create a project plan and a budget before the project started,
and had to adapt to the organization’s surrounding stage-gate
project management tools. There are often problems when
trying to scale up or use agile in a larger context [21], [26] and
the whole organization must accept and know the difference
in how the agile team is working [24].
One of the advantages with agile development compared
to other traditional methods is that decisions about the final
product can be made underway. One of the critiques of the Net
Present Value described is that it only has one decision point in
time [10]. Agile methods make decisions possible underway.
This can also be argued as being more honest, since this mostly
happens anyway but often shadowed by cover-up explanations
[1]. The aspect of stakeholder analysis is also different in
agile project management where the customer has to make
new decisions about the product underway. Since the product
owner in the studied organization owned the requirements and
sent them to the teams, the stakeholder analysis seemed to still
be traditional and not as much part of the team as suggested
in agile development.
The studied groups probably would get more positive effects
out of working agile if the organization around it would not
have been maintaining a stage-gate system. Taking a budget
decision before the project has started, which then is not
possible to change, locks the project into a certain way of
resisting flexibility in order to deliver what was expected from
the beginning. The Scrum Master also described this as a
problem, since the end-cost had to be decided beforehand.
This seems to be a dilemma when this large company tries to
implement flexibility, but does not dare to be flexible about
budgets and goals to conduct a project. The problem is then
that, if the project is very complex and no final goal can be
decided with almost any certainty, the final result will not be
as good as it could have been. [1] even states that the dream of
the perfect goal is futile, and the organization probably lacks
the trust needed to let go of some control.
However, there are reasons for companies not to implement agile fully in all aspects of a project. Aspects such
as, that plans drive funding, different people do architecture
and design, documents are needed to mitigate risks, and
development is not a part of the requirements process are all
reasons to maintain some traditional methods [27]. All of these
were reasons why the studied teams did not fully implement
the agile concepts. However, the main contribution of this
study is that in order to combine agile teams with a stagegate/traditional project organization the organization surrounding the teams must understand how the agile team is different
and adapt their feedback to their way of working. If they
do not, the agile team’s earlier increased job motivation will
somewhat decrease. An agile team, just like any other team,
expects feedback when they have worked hard and delivered
a good result. As this case study shows, the motivation of the
team will, of course, decrease if the surrounding organization
lacks the understanding of the agile team’s different ways of
working. Furthermore, if the positive or negative feedback
is given to the team at the wrong time, the team will feel
that their efforts are not appreciated. Since most organizations
combine agile with traditional projects management this study
shows that they should be aware of the interface between
the agile teams and the surrounding organization, not just for
overhead costs reasons, but also from a work motivational
perspective. In order to mitigate this lowered work motivation
risk, companies could make sure the stages and gates of the
surrounding organization are, at least somewhat, synchronized
with the iterations of the agile teams. This way, the most
evident feedback disappointments could be avoided.
In this specific case study the teams seemed to be content
with their methodology and made comparisons with other
non-agile teams in the same department. Compared to them,
the agile projects had delivered more value and faster, which
confirms earlier success stories from agile software development [19]. The focus group result shows that team members
were more motivated on the agile teams than when they
were working on other teams. The confirms job satisfaction
and team spirit research already conducted by for example
[6]. This study adds the perspective of employees getting
frustrated when working on non-agile teams after being on
an agile one within the same organization, due to a different
pace. This problem is harder to address, but making the
employees aware of these effects beforehand might decrease
their disappointment and frustration.
VI. L IMITATIONS
This case study only shows one example of what happened
to agile teams in a larger non-agile organization. In connection
to research of what motivates employees, it is most likely that
unsynchronized feedback loops will have the same effect in
other organizations. However, we, of course, cannot conclude
that the problem of unsynchronized feedback loops or frustration when returning to non-agile teams, are occurring in
other organizations, since we only studies one. Therefore, this
study only presents what happened on these specific teams.
Furthermore, the methodology is not thorough. It would have
been a good idea to check the reliability of the thematic
analysis by having more researchers code what was said in
the interview and the focus group.
VII. C ONCLUSION
AND
F UTURE W ORK
In conclusion, this study has shown that challenges when
integrating agile teams into a surrounding non-agile organization, did not only regard overhead costs, but were also of
job motivational nature in this specific case. This was due
to unsynchronized feedback loops between the agile teams’
delivery points and the surrounding stage-gate milestones.
Furthermore, employees reported being frustrated with the
slow pace when working on a non-agile team after having
been on an agile team.
These issues might occur on other organizations as well,
and if it does, it is most likely that the agile teams will feel
unappreciated since e.g. positive feedback will not be given
to the teams when they expect it. This result was shown as a
case study including an interview with the Scrum Master and
a focus group with a subset of two agile teams.
This means that organizations that implement agile methods
within traditional organizations should, not only expect higher
overhead costs, but also be aware of the different feedback
loops needed to the agile teams, and expect lower motivation
and frustration from employees on non-agile teams after have
participated in an agile project.
This study was just a first step in studying work motivational
aspects regarding the interface between agile teams and a
non-agile surrounding organization. The most obvious future
work is to see if these findings are true in more organizations,
which is preferably investigated using both qualitative (e.g. by
conducting more interviews using a more thorough method)
and quantitative data (e.g. distributing a survey to see how
agile teams in large non-agile organizations end up on work
motivation scales connected to this subject).
ACKNOWLEDGMENT
We would like to thank Pasi Moisander, Karin Scholes, and
Kristin Boissonneau Gren (without your goodwill this work
could not have been done).
R EFERENCES
[1] M. Engwall, “The futile dream of the perfect goal,” Beyond project
management, vol. 1, pp. 261–277, 2002.
[2] D. Fernandez and J. Fernandez, “Agile project management–Agilism
versus traditional approaches,” Journal of Computer Information Systems, vol. 49, no. 2, pp. 10–17, 2008.
[3] E. Goldratt, Critical chain. Great Barrington: North River Press, 1997.
[4] W. Feld, Lean manufacturing: Tools, techniques, and how to use them.
Boca Raton, Fla.: St. Lucie Press, 2001.
[5] M. Laanti, O. Salo, and P. Abrahamsson, “Agile methods rapidly
replacing traditional methods at Nokia: A survey of opinions on agile
transformation,” Information and Software Technology, vol. 53, no. 3,
pp. 276–290, 2011.
[6] G. Melnik and F. Maurer, “Comparative analysis of job satisfaction in
agile and non-agile software development teams,” in Extreme Programming and Agile Processes in Software Engineering. Springer, 2006,
pp. 32–42.
[7] R. Turner, The handbook of project-based management: improving the
processes for achieving strategic objectives, 2nd ed. London: McGrawHill, 1999.
[8] A. Babu and N. Suresh, “Project management with time, cost, and
quality considerations,” European Journal of Operational Research,
vol. 88, no. 2, pp. 320–327, 1996.
[9] D. Khang and Y. Myint, “Time, cost and quality trade-off in project
management: a case study,” International journal of project management, vol. 17, no. 4, pp. 249–256, 1999.
[10] O. Granstrand, Industrial innovation economics and intellectual property, 5th ed. Göteborg: Svenska kulturkompaniet, 2010.
[11] J. Berk and P. DeMarzo, Corporate finance, 2nd ed. Harlow, Essex:
Pearson, 2011.
[12] T. Copeland, F. Weston, and K. Shastri, Financial theory and corporate
policy, 4th ed. Boston, Mass.: Pearson Addison-Wesley, 2005.
[13] H. Maylor, Project management, 4th ed. Harlow, England: Financial
Times Prentice Hall, 2010.
[14] R. Cooper, “Stage-gate systems: a new tool for managing new products,”
Business Horizons, vol. 33, no. 3, pp. 44–54, 1990.
[15] H. Hartley and A. Wortham, “A statistical theory for pert critical path
analysis,” Management Science, vol. 12, no. 10, pp. B–469, 1966.
[16] M. Fowler and J. Highsmith, “The agile manifesto,” In Software Development, Issue on Agile Methodologies, last accessed on December
29th, 2006, Aug. 2001.
[17] N. Moe, A. Aurum, and T. Dybå, “Challenges of shared decisionmaking: A multiple case study of agile software development,” Information and Software Technology, vol. 54, no. 8, pp. 853–865, 2012.
[18] A. Roth, “Achieving strategic agility through economies of knowledge,”
Strategy & leadership, vol. 24, no. 2, pp. 30–36, 1996.
[19] K. Schwaber, Agile project management with Scrum. Redmond, Wash.:
Microsoft Press, 2004.
[20] G. Chin, Agile project management: how to succeed in the face of
changing project requirements. New York: AMACOM, 2004.
[21] B. Boehm and R. Turner, “Management challenges to implementing
agile processes in traditional development organizations,” Software,
IEEE, vol. 22, no. 5, pp. 30–39, 2005.
[22] S. Jalali and C. Wohlin, “Agile practices in global software engineering–
a systematic map,” in International Conference on Global Software
Engineering (ICGSE). IEEE, 2010.
[23] K. Korhonen, “Adopting agile practices in teams with no direct programming responsibility–A case study,” Product-Focused Software Process
Improvement, pp. 30–43, 2011.
[24] D. Karlström and P. Runeson, “Combining agile methods with stagegate project management,” Software, IEEE, vol. 22, no. 3, pp. 43–49,
2005.
[25] K. Petersen and C. Wohlin, “The effect of moving from a plan-driven
to an incremental software development approach with agile practices,”
Empirical Software Engineering, vol. 15, no. 6, pp. 654–693, 2010.
[26] ——, “A comparison of issues and advantages in agile and incremental
development between state of the art and an industrial case,” Journal of
Systems and Software, vol. 82, no. 9, pp. 1479–1490, 2009.
[27] D. West et al., “Water-scrum-fall is the reality of agile for most
organizations today,” Forrester Research, July, vol. 26, 2011.
Theory Construction and Model-Building Skills
Methodology in the Social Sciences
David A. Kenny, Founding Editor
Todd D. Little, Series Editor
www.guilford.com/MSS
This series provides applied researchers and students with analysis and research design books that
emphasize the use of methods to answer research questions. Rather than emphasizing statistical
theory, each volume in the series illustrates when a technique should (and should not) be used and
how the output from available software programs should (and should not) be interpreted. Common
pitfalls as well as areas of further development are clearly articulated.
REC ENT VOL UME S
PRINCIPLES AND PRACTICE OF STRUCTURAL EQUATION MODELING, FOURTH EDITION
Rex B. Kline
HYPOTHESIS TESTING AND MODEL SELECTION IN THE SOCIAL SCIENCES
David L. Weakliem
REGRESSION ANALYSIS AND LINEAR MODELS: CONCEPTS, APPLICATIONS,
AND IMPLEMENTATION
Richard B. Darlington and Andrew F. Hayes
GROWTH MODELING: STRUCTURAL EQUATION
AND MULTILEVEL MODELING APPROACHES
Kevin J. Grimm, Nilam Ram, and Ryne Estabrook
PSYCHOMETRIC METHODS: THEORY INTO PRACTICE
Larry R. Price
INTRODUCTION TO MEDIATION, MODERATION, AND CONDITIONAL
PROCESS ANALYSIS: A REGRESSION-BASED APPROACH, SECOND EDITION
Andrew F. Hayes
MEASUREMENT THEORY AND APPLICATIONS FOR THE SOCIAL SCIENCES
Deborah L. Bandalos
CONDUCTING PERSONAL NETWORK RESEARCH: A PRACTICAL GUIDE
Christopher McCarty, Miranda J. Lubbers, Raffaele Vacca, and José Luis Molina
QUASI-EXPERIMENTATION: A GUIDE TO DESIGN AND ANALYSIS
Charles S. Reichardt
THEORY CONSTRUCTION AND MODEL-BUILDING SKILLS:
A PRACTICAL GUIDE FOR SOCIAL SCIENTISTS, SECOND EDITION
James Jaccard and Jacob Jacoby
Theory Construction
and Model-Building Skills
A Practical Guide for Social Scientists
SECOND EDITION
James Jaccard
Jacob Jacoby
Series Editor’s Note by Todd D. Little
THE GUILFORD PRESS
New York  London
Copyright © 2020 The Guilford Press
A Division of Guilford Publications, Inc.
370 Seventh Avenue, Suite 1200, New York, NY 10001
www.guilford.com
All rights reserved
No part of this book may be reproduced, translated, stored in a retrieval system,
or transmitted, in any form or by any means, electronic, mechanical,
photocopying, microfilming, recording, or otherwise, without written permission
from the publisher.
Printed in the United States of America
This book is printed on acid-free paper.
Last digit is print number:
9
8
7
6
5
4
3
2
1
Library of Congress Cataloging-in-Publication Data
Names: Jaccard, James, author. | Jacoby, Jacob, author.
Title: Theory construction and model-building skills : a practical guide
for social scientists / James Jaccard, Jacob Jacoby.
Description: Second edition. | New York : The Guilford Press, [2020] |
Series: Methodology in the social sciences | Includes bibliographical
references and index.
Identifiers: LCCN 2019030892 | ISBN 9781462542437 (paperback) |
ISBN 9781462542444 (hardcover)
Subjects: LCSH: Social sciences—Research—Methodology. | Theory
(Philosophy)
Classification: LCC H62 .J29 2020 | DDC 300.72—dc23
LC record available at https://lccn.loc.gov/2019030892
For Marty Fishbein—
a brilliant and inspiring theorist and mentor
—James Jaccard
For Renee and Dana—
in appreciation for the balance and joy they create
—Jacob Jacoby
Series Editor’s Note
Theory is an analyst’s best friend, while data are the fodder of good theory; but which
comes first, the data or the theory? Similarly, some folks never let data get in the way
of good theory and some folks never let theory get in the way of good data; but is it this
either/or taming-­of-­the-­shrew-­like scenario? Constructing a theory is more like crafting
an elegant ensemble of logically connected ideas that depict the world and allow knowledge to leap forward. As Jaccard and Jacoby point out, 90% of our graduate training is
on methods for collecting and techniques for analyzing data and only 10% is spent on
identifying and crafting the ideas into theories that can be tested.
I have always been a student of good theory and this book is a veritable bible on
how to craft testable theories, even before the wonderful enhancements to this second
edition. Enhancements include, for example, what constitutes a theoretical contribution
and how do you craft one? How do you use a logic model to generate ideas and avoid
pitfalls in the theory construction process? How do you use mixed methods and data
mining to craft good theory? How do you test a theory and revise it if need be? Jaccard
and Jacoby answer these questions with practical wisdom born of extensive experience
and uncommon insights.
In addition, the expanded discussion of moderator variables, counterfactual causality, and their new 10-­step method for generating theory is simply invaluable. Oh, and the
new chapters on measurement are absolutely essential because good theory demands
good measurement. The operationalization of good theory is way too often neglected in
training programs and many seasoned veterans have yet to learn how to do good measurement. Last but not least, they offer insight and wisdom on both interpreting others’
theories as well as clearly expressing your own theories.
Their book is transdisciplinary, with many useful examples spanning fields such
as anthropology, business, communications, education, economics, health, marketing,
organizational studies, political science, psychology, social work, sociology, and so on.
vii
viii
Series Editor’s Note
And the companion website is a tremendously useful and helpful resource for students
and instructors alike! I will be teaching a course on model building and theory construction. Jaccard and Jacoby is unequivocally the only choice for such a course.
Sadly, Jacob “Jack” Jacoby has departed our worldly sphere. Brilliant thinkers challenge our worldview on humanity and they leave indelible marks that shape us and how
we think. Jack was one of our field’s finest.
Todd D. Little
Honeymooning at Casa Montana
Preface
Theory construction is at the heart of the scientific process. The strategies that social
scientists use to generate and develop ideas are important to understand and foster
in young academics and investigators as they prepare for a research-­oriented career.
Although books have been written about theory construction, there are surprisingly few
books on the topic that tackle the problem of teaching students and young professionals,
in a practical and concrete way, how to theorize. Students, especially graduate students,
take one or more courses on research methods and data analysis, but few experience
more than a lecture or two, or read a chapter or two, on theory construction. It is no
wonder that students often are intimidated by the prospect of constructing theories.
This book provides young scientists with tools to assist them in the practical
aspects of theory construction. It is not an academic discussion of theory construction
or the philosophy of science, and we do not delve too deeply into the vast literature on
these topics. Rather, we take a more informal journey through the cognitive heuristics,
tricks of the trade, and ways of thinking that we have found to be useful in developing
theories—­essentially, conceptualizations—­that can advance knowledge in the social
sciences. By taking this journey, we hope to stimulate the thinking and creative processes of readers so that they might think about phenomena in new and different ways,
perhaps leading to insights that might not otherwise have resulted. The intent of this
book is to provide a practical, hands-­on, systematic approach to developing theories and
fostering scientific creativity in the conceptual domain. Relative to the majority of books
on theory construction, this book is unique in its focus on the nuts and bolts of building
a theory rather than on an analysis of broad-­based systems of thought.
We have used the book both as a stand-­alone text in a course on theory construction
and as one of several texts in graduate courses on research and research methodology.
In terms of the latter, almost all traditional research methods books include a section
or chapter on the nature of theory and/or theory construction. However, the treatment
ix
x
Preface
of theory construction usually is brief and of limited practical value. The present book
is intended to provide the instructor with a useful source for helping students come up
with ideas for research and for fine-­tuning the resulting theories that emerge from such
thinking. It provides more detail and more practical knowledge than what is typical of
chapters in books on research methodology. The social psychologist William McGuire
often lamented about how research training with graduate students focuses at least 90%
on teaching methods to test ideas but no more than 10% on how to get those ideas in
the first place. Despite this difference in emphasis, the process of theory development is
fundamental to successful scientific research. Indeed, many would say that there can be
no theory testing without theory. An objective of this book is to move toward a needed
balance in the emphases given to theory construction and theory testing.
The book can be used in many different disciplines. We draw on examples from the
fields of anthropology, business, communications, education, economics, health, marketing, organizational studies, political science, psychology, social work, and sociology,
to name a few. Some instructors may prefer more detailed examples in their particular
field of study, but we believe that using examples from multiple disciplines helps students appreciate the commonalities and value of multidisciplinary perspectives.
The book has several pedagogical features that enhance its use as a textbook and as
a source of learning. First, each chapter includes a section on suggested readings with
commentary, where we direct the reader to key references for further study on the topics
covered in the chapter. Second, each chapter has a list of key terms that highlights the
most important jargon and terminology. Third, each chapter has a set of exercises that
encourages the reader to think about the material that was presented in the chapter. We
include exercises to reinforce concepts and exercises to apply the concepts to problems
of interest. Finally, each chapter has a highlighted box that covers an interesting topic
that applies the concepts covered in the chapter or that shows important uses of them.
We also created a website that contains supplemental materials to support the book (see
the box at the end of the table of contents). The website is intended for use by students,
professors, and professionals alike.
CHANGES IN THE SECOND EDITION
In the first edition, we downplayed issues surrounding data collection and data analysis, preferring to keep discussion at a conceptual level. This orientation still dominates
the current edition, but we felt it important to more fully recognize that theory often
emerges from data collection and data analysis. In the first edition, the emergence of
theory from data was front and center in the chapter on grounded and emergent theory because in qualitative research, emergent theory is prominent. However, the emergence of theory from exploratory data analysis was absent for quantitative research. We
have added a new chapter, “Emergent Theory: Quantitative Approaches” (Chapter 11),
to address this and have retitled our chapter on qualitative approaches (Chapter 10)
“Emergent Theory: Qualitative/Mixed-­Methods Approaches.” The quantitative chapter
on this topic prioritizes novel, exploratory methods of quantitative analysis that can
Preface
xi
help readers generate new theory through data mining. Readers who are less interested
in quantitative research may find the core material in this chapter of lesser relevance.
However, we have kept this material conceptual as opposed to being steeped in statistical theory and have provided supplemental materials on our website that walk readers
through the execution of the methods on popular software at a more practical level.
Another nod to the fact that data often lead to new theory is the addition of a chapter on theory revision (Chapter 15). When we collect data to test a theory, disconfirmatory results can emerge that lead us to revise the theory or abandon it altogether. When
faced with disconfirming or only partially supportive data, one can use critical thinking processes to make decisions about whether and how to revise a theory. Chapter 15,
“Theory Revision,” highlights these processes. Theory revision in light of disconfirming
data is as relevant to qualitative researchers as it is to quantitative researchers, so this
chapter should be of interest to all.
Measurement and observation are core to science. When we formulate measures, we
invoke theory to link measures to the underlying construct the measures are assumed
to reflect. When we address measurement error in research, measurement theory also
is front and center. As such, measurement theory is a core part of science. Measurement
is typically viewed as the province of methodology, but we seek to build a case with two
new chapters showing that the practice of measurement is firmly entrenched in theory
and that measurement-­oriented theory construction is essential for the social sciences.
The first chapter on measurement, “General Frameworks” (Chapter 13), emphasizes
the concepts of metrics, reliability, validity, and measurement facets. The second chapter, “Types of Measurement Strategies” (Chapter 14), focuses on self-­reports, observer
reports, and “objective” measures, strategies that form the backbone of social science
research. We provide readers with theory construction principles that guide how one
thinks about and conceptualizes such measures. Readers who want to contribute to
measurement theory will learn useful conceptual tools for doing so. Readers who are
not so inclined will still learn about the importance of measurement theory and how to
apply that theory to the specific research projects they pursue. This will be true for both
qualitative and quantitative researchers because both traditions ultimately use measurement in one form or another.
In the first edition, we wrote each chapter so that it could generally “stand on its
own.” The idea was that if instructors wanted to change the reading order of chapters,
omit certain chapters based on their own or their students’ substantive interests, or
browse different topics rather than read every chapter, the book would be amenable to
these approaches. The new chapters have this same quality.
We also have added new material to most chapters from the first edition. As examples, in Chapter 3, we added a discussion of what constitutes a theoretical contribution
and what strategies social scientists can use to make theoretical contributions. This
helps orient readers to the remainder of the book. We added a section on conceptual
logic models to the chapter on generating ideas (Chapter 4) because such models are key
to idea refinement. We expanded Chapter 6, on thought experiments, to give a better
appreciation of their role in science and reworked several of the example experiments.
We made clearer how thought experiments can be used both by confirmatory-­oriented
xii
Preface
and emergent-­oriented theorists. For the chapter on causal thinking (Chapter 7), we
expanded our discussion of moderator variables, added a brief discussion on counterfactual causality, and added two ways of generating theory, a 10-­step method and a
“binder” method. In this chapter, we also added a section on common mistakes made
during the theory construction process. We expanded the material on grounded and
emergent theory in Chapter 10 to discuss mixed-­methods approaches in more depth and
to develop the actual thought processes theorists use to construct theories from qualitative data. We also expanded Chapter 16, on reading about and writing theories, and
generally updated the book to include practical perspectives that have evolved in theory
construction since the first edition.
Our book should be useful (1) in theory construction courses, (2) in proseminars
for doctoral students to help them develop their thesis research, and (3) as a supplement
to methods courses where instructors can select a subset of chapters for students to read.
Young researchers and professors also should find the book of interest independent of
courses. We hope that even seasoned researchers will walk away from most chapters
with at least one or two new “nuggets of knowledge” they will find useful in their work.
In this sense, our intended audience is broad.
As noted, we have created a web page for the book (see the box at the end of the
table of contents). This contains useful supplemental information for readers as well as
instructional aids for professors.
ACKNOWLEDGMENTS
My dear friend and colleague Jacob Jacoby passed away before writing commenced on
the second edition. Despite this, his presence remains even in the new chapters. Jack
was an important influence on my life both professionally and personally, and I will
always be grateful for the opportunity to have known and learned from him. His legacy
will be with us for years. I miss you, Jack.
As with the first edition, a large number of people contributed in diverse ways to the
development of the second edition. I again would like to thank students and colleagues
who provided feedback on earlier drafts, including David Brinberg, Department of Marketing, Virginia Polytechnic University; Miriam Brinberg, Department of Human Development, Pennsylvania State University; Wendy J. Coster, Department of Occupational
Therapy, Boston University; Cynthia G. S. Franklin, Steve Hicks School of Social Work,
University of Texas, Austin; Liliana Goldin, Silver School of Social Work, New York
University; Guillermo Grenier, Department of Global and Sociocultural Studies, Florida
International University; Sean Patrick Kelly, School of Education, University of Pittsburgh; Hailin Qu, Spears School of Business, Oklahoma State University; Rick Sholette,
Director, Paraclete Ministries; Michael Slater, School of Communications, Ohio State
University; and Weiwu Zhang, Department of Public Relations, Texas Tech University.
At The Guilford Press, C. Deborah Laughton, as always, was insightful and supportive
in her role as editor. She once again improved the book immensely. It is rare one gets the
opportunity for a “do-­over” of a project, and I am grateful to C. Deborah for giving me
Preface
xiii
the opportunity. Although all of these individuals contributed significantly to the book,
I alone am responsible for any of its shortcomings.
I dedicated the first edition to Marty Fishbein and I do so again here. Marty was an
amazing scientist, teacher, and professor who positively impacted my life and those of
his many students. It is rare that one has the opportunity to study under and work with
a true genius. Jack, I am sure, would once again dedicate the book to his spouse, Renee,
and his daughter, Dana, two incredible women. My spouse, Liliana, again contributed
to the book in many meaningful ways, and I remain amazed after all these years at her
great intellect and breadth, which have been of such immense benefit to me in the form
of lively discussions of theory, method, and substance. She is my role model in every
way. And a special note of recognition to my daughter, Sarita, who inspires me and
serves as my role model every bit as much as her mother.
James Jaccard
Brief Contents
PART I. BASIC CONCEPTS
1 · Introduction
3
2 · The Nature of Understanding
7
3 · Science as an Approach to Understanding
22
PART II. CORE PROCESSES
4 · Creativity and the Generation of Ideas
51
5 · Focusing Concepts
95
6 · Thought Experiments for Variable Relationships
111
PART III. FRAMEWORKS FOR THEORY CONSTRUCTION
7 · Causal Models
151
8 · Mathematical Modeling
196
9 · Simulation as a Theory Development Method
248
10 · Emergent Theory:
Qualitative/Mixed‑Methods Approaches
267
xv
xvi
Brief Contents
11 · Emergent Theory: Quantitative Approaches
307
12 · Historically Influential Systems of Thought
340
PART IV. THEORY AT THE LEVEL OF MEASUREMENT
13 · Theory and Measurement: General Frameworks
375
14 · Theory and Measurement:
Types of Measurement Strategies
402
PART V. CONCLUDING ISSUES
15 · Theory Revision
435
16 · Reading and Writing about Theories
461
17 · Epilogue
485
References
491
Author Index
511
Subject Index
518
About the Authors
522
Extended Contents
PART I. BASIC CONCEPTS
1 · Introduction
3
Organization of the Book / 4
Theories and Settings / 5
2 · The Nature of Understanding
7
The Nature of Reality / 8
Concepts: The Building Blocks of Understanding / 11
Conceptual Systems: The Bases for Deeper Understanding / 15
BOX 2.1.
Concepts, Cultures, and Values / 15
Communication / 17
Summary and Concluding Comments / 19
Suggested Readings / 20
Key Terms / 21
Exercises / 21
3 · Science as an Approach to Understanding
22
Approaches to Understanding / 22
BOX 3.1.
The Fringes of Science / 25
The Essentials of Scientific Endeavor / 26
Science and Objectivity / 27
The Process of Theory Construction / 28
Characteristics of a Good Theory / 32
What Is a Theoretical Contribution? / 33
Summary and Concluding Comments / 45
Suggested Readings / 46
Key Terms / 47
Exercises / 47
xvii
xviii
Extended Contents
PART II. CORE PROCESSES
4 · Creativity and the Generation of Ideas
51
One Small Step for Science / 52
Creativity / 52
Choosing What to Theorize About / 56
Literature Reviews / 58
Heuristics for Generating Ideas / 59
BOX 4.1.
The Nacirema / 60
Scientists on Scientific Theorizing / 81
Conceptual Logic Models / 84
Summary and Concluding Comments / 89
Suggested Readings / 90
Key Terms / 91
Exercises / 91
APPENDIX 4.1.
Examples of Weak Argumentation / 93
5 · Focusing Concepts
95
The Process of Instantiation / 95
Shared Meaning, Surplus Meaning, and Nomological Networks / 98
Practical Strategies for Specifying Conceptual Definitions / 99
BOX 5.1.
Etic and Emic Constructs / 100
Multidimensional Constructs / 103
Creating Constructs / 104
An Example of Specifying Conceptual Definitions / 105
Operationism / 107
Summary and Concluding Comments / 108
Suggested Readings / 109
Key Terms / 109
Exercises / 110
6 · Thought Experiments
for Variable Relationships
Thought Experiments for Relationships in Grounded
and Emergent Theory / 113
Describing Relationships with Different Types of Variables / 114
Thought Experiments for Relationships between Nominal Variables / 115
Thought Experiments for Relationships between Quantitative Variables / 118
Thought Experiments for Relationships between Nominal
and Quantitative Variables / 129
BOX 6.1.
Simpson’s Paradox / 130
Thought Experiments for Moderated Relationships / 136
Broader Uses of Hypothetical Factorial Designs in Thought Experiments / 143
Summary and Concluding Comments / 146
Suggested Readings / 146
Key Terms / 147
Exercises / 147
111
Extended Contents
xix
PART III. FRAMEWORKS FOR THEORY CONSTRUCTION
7 · Causal Models
151
Two Types of Relationships: Predictive and Causal / 152
Causality and Grounded/Emergent Theory / 155
Types of Causal Relationships / 155
Constructing Theories with Causal Relationships / 158
Identifying Outcome Variables / 159
Identifying Direct Causes / 159
Indirect Causal Relationships / 160
Moderated Causal Relationships / 165
Reciprocal or Bidirectional Causality / 169
Spurious Relationships / 173
Unanalyzed Relationships / 175
Expanding the Theory Yet Further / 176
BOX 7.1.
Finding Sources for a Literature Review / 184
The Binder Method / 187
Common Mistakes during Causal Theory Construction / 189
Perspectives on the Construction of Causal Theories / 189
Summary and Concluding Comments / 192
Suggested Readings / 192
Key Terms / 193
Exercises / 194
8 · Mathematical Modeling
Types of Variables: Nominal, Discrete, and Continuous / 197
Axioms and Theorems / 198
Functions / 198
Linear Functions / 199
Deterministic versus Stochastic Models / 203
Model Parameters / 204
Rates and Change: Derivatives and Differentiation / 205
Describing Accumulation: Integrals and Integration / 208
Just‑Identified, Overidentified, and Underidentified Models / 209
Metrics / 210
Types of Nonlinearity / 210
BOX 8.1.
Reading Mathematical Models / 211
Functions for Nominal Variables / 222
Advanced Topics: Manipulating and Combining Functions / 224
Multiple Variable Functions / 227
Phases in Building a Mathematical Model / 228
An Example Using Performance, Ability, and Motivation / 228
An Example Using Attitude Change / 232
Chaos Theory / 235
Catastrophe Theory / 237
Additional Examples of Mathematical Models in the Social Sciences / 238
Emergent Theory Construction and Mathematical Models / 238
Summary and Concluding Comments / 239
196
xx
Extended Contents
Suggested Readings / 240
Key Terms / 241
Exercises / 242
APPENDIX 8.1.
APPENDIX 8.2.
SPSS Code for Exploring Distribution Properties / 243
Additional Modeling Issues for the Performance,
Ability, and Motivation Example / 245
9 · Simulation as a Theory Development Method
248
Defining Simulations / 249
The Uses of Research Simulations / 250
The Difference between Simulations and Laboratory Experiments / 250
Basic Simulation Varieties / 251
The Analysis of Criterion Systems as a Basis for Theory Construction / 253
Simulations and Virtual Experiments / 260
Agent‑Based Modeling / 260
BOX 9.1.
Agent-Based Modeling of Segregation / 262
Resources for Conducting Simulations / 263
Summary and Concluding Comments / 263
Suggested Readings / 264
Key Terms / 265
Exercises / 265
10 · Emergent Theory:
Qualitative/Mixed‑Methods Approaches
267
Grounded and Emergent Theory: An Overview / 268
Qualitative Research and Ways of Knowing / 269
Framing the Problem / 270
The Role of Past Literature / 271
Collecting Qualitative Data / 271
Field Notes and Memo Writing / 277
BOX 10.1.
Anthropology and the Ethnographic Tradition / 278
Theoretical Sampling / 279
Analyzing and Coding Data / 279
Constructing Theory in Qualitative Research / 285
Mixed‑Methods Research / 291
Products of Qualitative and Mixed‑Methods Research / 293
Summary and Concluding Comments / 294
Suggested Readings / 296
Key Terms / 297
Exercises / 298
APPENDIX 10.1. The Limits of Information Processing / 300
11 · Emergent Theory: Quantitative Approaches
Exploratory Analyses of Direct Effects / 309
BOX 11.1.
Robust Statistics / 318
Exploratory Analyses of Moderated Relationships / 320
Cluster Analysis / 325
Factor Analysis / 329
307
Extended Contents
xxi
Big Data, Data Mining, and Machine Learning / 331
Chance Results / 335
Summary and Concluding Comments / 336
Suggested Readings / 336
Key Terms / 337
Exercises / 338
12 · Historically Influential Systems of Thought
340
Grand Theories / 341
Frameworks Using Metaphors / 352
Frameworks Emphasizing Stability and Change / 357
Psychological Frameworks / 359
BOX 12.1.
Collaboration / 360
Frameworks Inspired by Methodology / 364
Summary and Concluding Comments / 369
Suggested Readings / 369
Key Terms / 371
Exercises / 372
PART IV. THEORY AT THE LEVEL OF MEASUREMENT
13 · Theory and Measurement: General Frameworks
375
Defining Measurement / 376
Conceptual Definitions and Measurement / 377
Classic Test Theory: The Basics / 379
The Facets of Measurement and Theory Construction / 382
Measurement Theory Construction / 383
Emergent Theory Building in Measurement: Cognitive Testing,
Expert Interviews, and Pilot Research / 390
BOX 13.1.
Measuring Sex and Gender / 391
Using Extant Measures / 393
Other Criteria for Measure Evaluation / 395
Classical Test Theory and Qualitative Research / 396
Writing Reports to Justify Measurement Choices / 396
Empirical Tests of Measurement Theory / 397
Summary and Concluding Comments / 397
Suggested Readings / 399
Key Terms / 400
Exercises / 400
14 · Theory and Measurement:
Types of Measurement Strategies
Constructing a Theory of Self‑Reports / 403
BOX 14.1.
Quantifying Love / 413
Constructing a Theory of Observer Reports / 423
Constructing a Theory of “Objective” Measures / 425
The Importance of a Theory Construction Mindset to Measurement / 426
Measurement and Qualitative Research / 426
402
xxii
Extended Contents
Summary and Concluding Comments / 428
Suggested Readings / 429
Key Terms / 430
Exercises / 431
PART V. CONCLUDING ISSUES
15 · Theory Revision
435
Disconfirmation and Theory Revision / 436
Boundary Conditions and Theory Revision / 438
Replication and Theory Revision / 440
Disconfirmation and Qualitative Research / 441
BOX 15.1.
Strong Inference / 442
Bayesian Perspectives in Theory Revision / 443
A Look to the Future: Computer Automation and Theory Revision / 451
Theoretical Desiderata for Revised Theories / 456
Paradigm Shifts / 456
Summary and Concluding Comments / 457
Suggested Readings / 458
Key Terms / 458
Exercises / 459
16 · Reading and Writing about Theories
461
Reading about Theories / 461
BOX 16.1.
PowerPoint Presentations of Theories / 466
Writing about Theories / 468
Grant Proposals, Technical Reports, and Presentations / 473
Summary and Concluding Comments / 474
Suggested Readings / 475
Key Terms / 475
Exercises / 476
APPENDIX 16.1. Inferring Theoretical Relationships
from the Choice of Statistical Tests / 477
17 · Epilogue
485
A Program of Self‑Study / 488
Concluding Comments / 490
References
491
Author Index
511
Subject Index
518
About the Authors
522
The companion website (www.theory-­construction.com) includes PowerPoint slides
of all of the book’s figures, methodological primers and video demonstrations,
supplemental exercises, and other resources.
Part I
BASIC CONCEPTS
1
Introduction
Few people dispute the central role of theory in the social sciences. Scientists formulate
theories, test theories, accept theories, reject theories, modify theories, and use theories
as guides to understanding and predicting events in the world about them. A great deal
has been written about the nature and role of theory in the social sciences. These writings have spanned numerous disciplines, including anthropology, economics, history,
philosophy, political science, psychology, sociology, and social work, to name but a few.
This literature has described, among other things, broad frameworks for classifying
types of theories, the evolution of theories over time, the lives and scientific strategies
of great scientific theorists, and general issues in the philosophy of science. Although
this literature is insightful, much less has been written to provide social scientists with
practical guidelines for constructing theories as they go about the business of doing
their science. Most students are intimidated by the prospect of constructing their own
theories about a phenomenon. Theory construction is viewed as a mysterious process
that somehow “happens” and is beyond the scope and training of a young scientist trying to find his or her way in the field. Whereas most graduate programs in the social
sciences require multiple courses in research methodology so that students will become
equipped with the tools to test theories empirically, the same cannot be said for theory
construction. In contrast to focusing on methods for testing theory, the current work
focuses on methods for generating theory.
The fundamental objective of this book is to provide students and young scientists
with tools to assist them in the practical process of constructing theories. It does so via
describing in some detail the strategies, heuristics, and approaches to thinking about
problems that we have found to be useful over the more than 70 collective years that
we have been doing social science research. This book is not an academic discussion of
the literature on theory construction or the philosophy of science. We do not delve too
deeply into the vast literature on these topics. Rather, we take a more practical journey
through the cognitive heuristics, tricks of the trade, and ways of thinking that we have
found to be useful in developing theories.
3
4
Basic Concepts
ORGANIZATION OF THE BOOK
The book is organized into five parts. Part I presents the basic concepts that form the
backdrop for later chapters. In these early chapters, we consider the nature of science
and what it means to understand something. We develop the notion of concepts and
highlight the central role of concepts in theories. We lay the foundations for communicating to others the concepts in one’s theory and then describe what separates science
from other ways of knowing.
With this as background, we turn to developing core strategies for constructing a
theory, the topic of Part II. In Chapter 4, we focus first on strategies for generating ideas
and for stimulating creative thinking. Once you have a set of rough ideas, they need
to be refined and focused to meet the criteria of a rigorous scientific theory. Chapter 5
describes strategies for thinking through your constructs and discusses how to develop
clear and communicable conceptual definitions of them. We provide numerous strategies for making fuzzy constructs more precise and not overly abstract. Chapter 6 focuses
on relationships between variables and develops strategies for making explicit the relationships you posit between variables. We show how to derive theoretical propositions
based on a careful analysis of relationships.
Part III considers different frameworks for theory generation. Chapter 7 considers
one of the most dominant approaches to theory construction in the social sciences, the
framework of causal thinking. This approach elaborates the causes and consequences
of different phenomena and views the identification of causal linkages as a central goal
of science. Chapter 8 describes strategies for building mathematical models of different
phenomena. Our intent here is to make clear the sometimes seemingly mysterious ways
in which mathematics and social science theorizing interface. Chapter 9 describes the
potential that simulations—­in particular, the development of simulations—­have for
theory construction. Chapters 10 and 11 develop emergent approaches to the construction of theory. Chapter 10 focuses on qualitative methods to identify constructs and
relationships on which to focus a theory, whereas Chapter 11 focuses on exploratory
quantitative/statistical approaches that allow theory to emerge from data. Some social
scientists might argue that these chapters belong in the previous section, where one
initially identifies constructs and relationships to include in a theory. As we emphasize
throughout this book, theory construction is not a set process, and a case like this
could be made for almost every chapter in the current section. This book provides you
with key ingredients for constructing a theory. How you choose to mix those ingredients to form your theoretical recipe depends on your predilections and the domains
that you are studying. Chapter 12 summarizes 13 broad-based theoretical frameworks
that may help in the idea-­generation process. These frameworks include materialism,
structuralism, functionalism, symbolic interactionism, evolutionary perspectives, postmodernism, neural networks, systems theory, stage theories, reinforcement theories,
humanism, multilevel modeling, and person-­centered theorizing. The idea is to think
about a phenomenon you are interested in from two or more of these perspectives in
order to help generate fresh ideas and perspectives.
Introduction
5
Part IV focuses on theory construction as applied to measurement. Measurement
and observation are, of course, central to science. When we formulate measures, we
invoke theory to link measures to the underlying construct the measures are assumed
to reflect. As such, measurement theory is a core part of science. Measurement is typically viewed as the province of methodology, but we argue here that key principles of
measurement derive from theory and that measurement-­oriented theory construction
is thus essential. We seek to develop your theory construction skills in this important
domain. Chapter 13 emphasizes the concepts of metrics, reliability, validity, and measurement facets. Chapter 14 focuses on self-­reports, observer reports, and “objective”
measures, strategies that form the backbone of social science research. Both chapters are
central to qualitative as well as quantitative research.
In the final section, we first consider theory revision in light of disconfirming data.
When we collect data designed to test or evaluate a theory, sometimes results emerge
that lead us to revise the theory or abandon it all together. When faced with disconfirming or only partially supportive data, one uses critical thinking processes to make decisions about whether and how to revise a theory. Chapter 15 highlights these processes.
Theory revision in light of disconfirming data is as relevant to qualitative researchers as
it is to quantitative researchers, so this chapter should be of interest to all. Chapter 16 in
this section discusses strategies for reading journal articles and scientific reports so as
to make explicit the theories that the authors describe and subject to empirical evaluation. We also discuss strategies for presenting theories in different kinds of reports. We
close with an Epilogue that comments on the theory construction process in light of the
material covered in previous chapters and that addresses some odds and ends that did
not fit well into the other chapters.
We recognize that some chapters will appeal to those with a more quantitative bent
to science, whereas other chapters will appeal to those with a more qualitative orientation. We have written the chapters so that each stands on its own, allowing you to skip
around as you see fit. Having said that, we strongly urge you to read all the chapters to
truly broaden and enrich your toolbox for generating theory. Both qualitative and quantitative approaches are powerful tools for generating ideas and theories. At some point,
you will want to master both.
THEORIES AND SETTINGS
This book is written primarily for students and professionals interested in pursuing a
career as a researcher in the social sciences. It is intended to provide you with concrete
strategies for building upon existing theories and constructing your own theories. Theorizing does not occur in a vacuum. It occurs in the context of individuals pursuing a
career in some professional setting, usually an academic setting. At times, we describe
how the setting in which you work impacts the way in which you theorize and the kinds
of questions you ask. We also discuss strategies for dealing with the constraints you face
as a result of these settings.
6
Basic Concepts
For us, constructing theory is one of the most rewarding aspects of doing science; it
is on a par with the excitement associated with empirically testing and finding support
for theory. In all honesty, we probably are more captivated by research that questions
the theories we have posited because of the ensuing call to “put on our detective hats”
to figure out why we were wrong. This invariably demands that we approach the problem from a new conceptual angle. This book describes some (but not all) of the types of
detective hats that we have put on over the years. We hope that we can help to start you
down the path of a richer and more productive approach to the construction of theories
as you fashion your own strategies and set of detective hats for thinking about phenomena and solving problems.
2
The Nature of Understanding
Reality is merely an illusion, albeit a very persistent one.
—Albert Einstein
The whole of science is nothing more than an extension of
everyday thinking.
—Albert Einstein
Despite the fact that science has been practiced for thousands of years and countless
books have been written on the subject, many people still consider it mysterious and
forbidding. Perhaps the reason for this reaction is their view of science as something
fundamentally different from anything they normally do. Actually, this is not the case.
The essence of science is something we all do, which is to try to understand ourselves
and the world around us. Scientific research is a process that is designed to extend
our understandings and to determine if they are correct or useful. The basic difference
between everyday thinking, on the one hand, and science and scientific research, on
the other, is that the latter strives to operate according to a more rigorous set of rules.
Because science and the process of scientific research can be viewed as extensions of
everyday thinking, we find them easiest to explain if we begin by considering how an
individual tries to make sense of, and cope with, his or her world.
The present chapter explores the nature of understanding, relying on informal
and everyday examples of human thought to draw parallels to scientific conceptions of
understanding. In doing so, we build on Albert Einstein’s assertion that “the whole of
science is nothing more than an extension of everyday thinking.” We begin by describing the different ways in which social scientists think about reality, considering the perspectives of realism, social constructionism, critical realism, and hypothetical realism.
Next, we address the building blocks of human understanding, namely, concepts and
conceptual systems that relate one concept to another. Given that a scientist has evolved
a conceptual system to address an issue, he or she then must communicate that system
to other scientists. We conclude the chapter by briefly considering the nature of communication so as to set the stage for future chapters on how to derive precise conceptual
definitions in theory construction.
7
8
Basic Concepts
THE NATURE OF REALITY
The process of understanding our world and making sense of reality is central to our waking lives. Accordingly, let us be more specific about what we mean by reality and understanding. Much philosophical thought has been devoted to the question of the nature of
reality, and there is controversy among scientists about whether a single objective reality
could ever be shown to exist.1 According to the traditional perspective, termed realism,
reality exists independent of any human presence. There is an external world composed
of objects that follow myriad natural facts and laws. It is up to us to discover these facts
and laws. Using this perspective, science has evolved and prospered as an approach for
gaining knowledge that mirrors the presumed actualities of the real world.
In contrast to realism, the social constructionist perspective holds that reality is a
construction of the human mind; that this construction is tied to a particular time and
social context; and that what is considered reality changes as the social context changes.
In its most extreme form, constructionism maintains that there is no reality and there
are no facts until these are conceptualized and shared by some number of people. A
more moderate position holds that, though there is an external reality independent of
humankind, we can never know its units and true laws—or even if it has units and true
laws. All we can know is our interpretation or construction of these experiences. Since
the same experiences are open to many interpretations, any or all of which may be correct, the correctness of an interpretation depends on the purposes of those doing the
interpreting. Thus, as Scarr (1985) notes:
We do not discover scientific facts; we invent them. Their usefulness to us depends both on
shared perceptions of the facts (consensual validation) and whether they work for various
purposes, some practical and some theoretical. (p. 499)
As a simple example that drives the point home, we frequently draw a set of parallel lines on a blackboard and ask our students to describe what they see. Some reply
“a road,” others say “two lines.” When we attempt to focus their thinking by saying “Hint: It’s a number,” three principal interpretations emerge: “Arabic number 11,
Roman numeral II, or binary number three.” Each of these responses represents a
different, but potentially accurate, reconstruction of the same objective reality that
reflects that individual’s mental perspective. This point is fundamental. Even if the
existence of a single objective, external reality could be assumed, the way in which
this reality is interpreted can vary within the individual over time and across individuals, and can be heavily influenced by context (e.g., someone immersed in the study of
Roman antiquity is more likely to interpret two parallel lines as representing Roman
numeral II). Every individual develops his or her own reality, so that a number of
different realities may be constructed out of the same set of “objective” facts. As is
1
The discussion that follows is a simplified and not necessarily universally shared perspective on realism,
social constructionism, and hypothetical realism. Philosophers and social scientists use these terms in
different ways.
The Nature of Understanding
9
increasingly being recognized, it is possible for more than one of these different realities to be correct and useful.
The social constructionist perspective has implications for the way in which science is viewed (see, e.g., Gergen, 1985; Gergen & Gergen, 2003). The principal implications of the social constructionist perspective do not affect so much the way in which
scientific empiricism is practiced, but rather the way in which the conceptualizations
and outcomes of the assessment process are interpreted. According to the realism perspective, conceptual systems and theories are created so as to mirror an existing reality.
The outcomes of the assessment process can be taken as direct representations of that
reality, and it is possible to make claims regarding ultimate truths. By contrast, although
the social constructionist perspective usually involves the researcher doing virtually
the same things as are done in a realism perspective, the recognition that there exist
multiple possible realities orients the researcher toward interpretations that reject more
absolute perspectives on mapping out a single, existing reality:
The admission that reality is a construction of the human mind does not deny the . . . value
of the construction. Indeed, we get around in the world and invent knowledge that is admirably useful. But the claim that science and reality are human constructions denies that
there is any one set of facts that is absolute and real. Instead, it asserts that there are many
sets of “facts” that arise from different theory-­g uided perceptions. (Scarr, 1985, p. 501)
On the one hand, the social constructionist perspective can be discomforting
because it makes us less certain of what we do and what we think we know. “How can
we know what is right if there is no right?” (Scarr, 1985, pp. 511–512). On the other
hand, this perspective enables us to more clearly recognize that any given conceptualization, and the facts that are given meaning by that conceptualization, is a function of
the sociocultural time and space in which they occur.
A middle ground relative to these somewhat conflicting perspectives has been
articulated by Blumer (1969). According to this view, reality is indeed seen through
human conceptions of it. But the empirical world also “talks back” to our conceptions
in the sense of challenging, resisting, and failing to bend to them. If a knife is plunged
into someone’s heart, certain ramifications follow (e.g., the heart will cease to function).
The ways in which these ramifications are construed and interpreted may vary from
one conceptual scheme to another. But the environment has spoken. It is this inflexible
character of the world about us that calls for, and justifies, empirical science. Science
seeks to develop conceptions that can successfully accommodate the obdurate character
of the empirical world. Blumer’s view roughly maps onto a philosophy of science known
as critical realism, though there are many alternative formulations of it (e.g., Manicas,
2006; Sayer, 1992).
Another influential perspective on the debate is that even if it cannot be proven that
reality exists, it is useful to assume that it does. This approach has been termed hypothetical realism. Here the concept of reality is a heuristic device—­something that helps
us organize our thoughts and think about matters so as to accomplish certain goals and
objectives. Strictly speaking, reality may or may not exist, but we approach the world
10
Basic Concepts
and our attempts to understand the world as if it does. In doing so, we may be able to
accomplish a wide range of goals, but accordingly, we also may be constrained in our
thinking. Hypothetical realism derives from a broader approach to epistemology—pragmatism. The approach is reflected in the work of philosopher C. I. Lewis (1929), who
argued that science does not provide a copy of reality but must work with conceptual
systems that are chosen for pragmatic reasons so as to aid scientific inquiry. Assuming
a hypothetical reality is one such aid.
In sum, whereas realism embraces a view that external reality exists and the goal
of science is to discover the laws that govern that reality, constructionism emphasizes
that reality is a construction of the human mind that is tied to a particular time and
social context. There are many gradations of these viewpoints, such as the position
advocated by Blumer (1969), which emphasizes that reality is seen through human
conceptions of it but that there is an empirical world that “talks back” to our conceptions; and hypothetical realism, which recognizes that one may not be able to prove
that reality exists, but nevertheless approaches science with a working assumption that
it does.
The broader literature on the philosophy of science explores myriad perspectives on
how scientists (and laypeople) think about reality. Because this literature presents more
nuanced perspectives than what we present here, interested readers are encouraged to
pursue it (see Suggested Readings at the end of the chapter).
How Reality Is Experienced
Assuming for the moment that reality exists, how is it experienced by the individual?
Most would agree that we experience the world around us as a complex, dynamic flow
of unique and unrepeatable phenomena and events. Furthermore, most of these phenomena and events—­ranging from those occurring deep in intergalactic space to those
occurring in the micromolecular structure of this book—are not directly observable. No
wonder, then, that attempting to understand our world can be a difficult process.
• Reality appears complex. Whatever else it is or may be, the world—­especially the
external world—that we experience is complex. Consider a lecture hall filled with 200
students. Forget about the world beyond our immediate view; to describe, in precise
detail, the sizes, shapes, colors (of clothing, objects, etc.), relationships, psychological
components, and sociological components of that lecture hall at one instant in time
could take months, years, or perhaps even lifetimes.
• Reality appears dynamic. Moreover, things never stay the same. The world at any
given instant is different from the world at the very next instant. From the tiniest particles that constitute physical matter to the largest galaxies, things are always in motion.
The cells of living organisms are always growing or decaying, and the impulses in the
neuronal system are always at work. So even if we were able to describe in infinite detail
our hypothetical lecture hall filled with students, once one or more students moved, we
would have a set of different relationships and, hence, a different reality.
The Nature of Understanding
11
• Reality appears unique. Because of this dynamic quality, the universe at any given
instant—­and everything in it—is never the same as the universe at any other instant,
either previous or subsequent. The water that flows at one particular instant or during
any given day down the rivers of New Hampshire, the raindrops that fall on a particular
evening in Houston, the expense account dinner that was eaten in Paris—all are unique
and can never be repeated precisely. The planet contains more than 7 billion human
inhabitants, yet no two people are precisely identical in all respects—­from their mundane external features (e.g., fingerprints) to their more complex internal features (how
and what they think and feel). Even the inanimate rock lying on the ground is unique.
In theory, no two rocks are identical in terms of all their distinguishing characteristics.
• Reality appears mostly obscured. Probably the major share of reality remains hidden from direct detection by any of our senses. To be sure, scientific instruments are
being developed that enable us to probe more deeply into space, see ever tinier particles
of matter and, through functional magnetic resonance imaging, observe how regions
of our brains are activated as we think, but the vast majority of nature’s secrets still
remains mysteriously hidden from direct view. These secrets cannot be seen, heard,
tasted, smelled, or touched. With specific respect to human phenomena, whereas many
are openly visible (e.g., we can see a person walking, eating a sandwich, purchasing a
newspaper), a vast number of others are not. A person’s psyche—­the inner thoughts
and feelings that presumably guide much of our behavior—­is one of the most obscure
realms of all. We have yet to be able to see what we think is a motive or to point to the
resting place of jealousy or pride.
The four characteristics just described—­that reality is experienced as complex,
dynamic, unique, and mostly obscured—­refer to what is often termed the external environment. More than a century ago, famed philosopher and social scientist William James
(1890) referred to this external environment as “a bloomin’, buzzin’ world of confusion.”
These four characteristics apply to individuals’ “internal environments” as well.
CONCEPTS: THE BUILDING BLOCKS OF UNDERSTANDING
The Nature of Concepts
Confronted by this array of complex, dynamic, unique, and mostly obscured phenomena, how do individuals manage to make sense out of this world? They do so, almost
automatically and often unconsciously, by conceptualizing—­that is, by using their mental processes to consider and sort their experiences in terms of the concepts they have
acquired and stored in memory. They also develop new concepts to describe things they
had never previously experienced. Just as concepts are the fundamental building blocks
of everyday thinking, they also are the fundamental building blocks of scientific thinking.
According to Webster’s Dictionary, the word concept refers to something that is conceived of in the mind. It is a generic idea or thought, usually developed from experienc-
12
Basic Concepts
ing one or more particular instances. Examples of concepts include shirt, book, dripping,
chair, mother, ice cream, advertising, smashed, home, vacation, memory, love, prejudice,
attitude, and expectations. As you can see, concepts refer to things that are tangible and
denotable (e.g., shirts) as well as to things that are not as concrete and directly seen (e.g.,
memory).
Concepts are the building blocks for all thinking, regardless of whether that thinking occurs in the context of everyday living, art, politics, sports, religion, or science. In
fact, without concepts, thought as we know it would be impossible. It is our concepts
that enable us to achieve some basic understanding of the world.
The most basic level of understanding can be termed identification. We understand
something, in part, when we can identify it. When experiencing the world about us, we
use the concepts we have in mind to identify and classify our experiences: this is an ice
cream cone; that is a shirt. Social scientists identify and classify people using concepts
such as race, gender, intelligence, and attitudes. Because concepts are so central to all
thinking, we examine their nature in somewhat greater detail.
• Concepts are generalized abstractions. When an individual has a concept, it means
that he or she has a general idea that can be applied across a number of specific instances.
Consider the concept shirt, for example. Shirts differ in a great number of ways—in
terms of their fabric, color or number of colors, sleeve length, number of buttons (or
whether they have buttons at all), size and shape of the collar, whether there are pockets
and the number of pockets, whether the shirt is squared off at the bottom or has tails,
and so on. Yet, having the concept shirt in mind is sufficient to enable the individual to
sort things into two (or possibly three) categories: shirts, items that have some of the
characteristics of shirts but are not shirts, and everything else. When we say that concepts are generalized abstractions, we mean that the general idea subsumes a universe
of possible instances. Note, also, that concepts can be “fuzzy” at the margins. Does a
woman’s blouse qualify as a shirt? What about a woman’s halter top? Such fuzziness
can lead to disagreements among individuals and scientists alike. For example, a recent
controversy in astrophysics involved how to define the concept of a planet.
• Concepts encompass universes of possibilities. An important feature of concepts—­
one that has fundamental implications for scientific theory and research—­is that each
concept consists of a universe of content. As just discussed, the concept shirt encompasses a universe of many specific possibilities. The concept ice cream cone encompasses
a universe of possibilities. The concepts attitude toward abortion, romantic love, and so
on, all encompass universes of possibilities.
• Concepts are hypothetical. Concepts are not reality, just ideas regarding reality.
This point is easy to appreciate when concepts or constructs apply to nebulous, amorphous, abstract things, such as wanderlust, attitude, or sustainable development. But this
point also applies to items that are denotable and concrete. For example, although the
concept of a shirt exists in our minds, we do not walk around with little shirts in our
minds. Neither the word shirt nor the thought that this word evokes is a shirt. Until
The Nature of Understanding
13
and unless neurological science tells us differently, concepts possess no tangible reality, in and of themselves. In this sense, all concepts are necessarily hypothetical, and,
although concepts are themselves hypothetical, the things to which they refer include
both observable entities (e.g., shirts, tables, dogs), which form part of the external environment, and nontangible phenomena such as love, happiness, and hunger. Although we
cannot see a person’s hunger directly, we can see the effects of this assumed state and,
from these effects, infer its existence. Many of the concepts that populate our minds are
of this indirectly observable variety.
• (Most) concepts are learned. Most concepts are acquired creations. The infant does
not come into the world already possessing the concept shirt. Rather, he or she must
acquire this concept before being able to use it to understand reality and communicate with others. When individuals experience something completely new and different,
they must either acquire or create a concept to be able to identify this experience and
distinguish it from all other aspects they perceive. Similarly, the scientist who observes
something different under the microscope or in intergalactic space will need to first
conceptualize it and then give it a unique label (e.g., chromosome, quasar) with which
to identify this particular phenomenon and others like it. Although most concepts are
learned, there is evidence that certain concepts may be “hardwired,” such as the face of
a mother as perceived by a newborn (Bednar & Miikkulainen, 2003).
• Concepts are socially shared. In order for communication to occur, the set of concepts possessed by one individual generally needs to be similar to the sets possessed by
others. Consider trying to discuss the notions of balks, punts, and love–15 with someone who does not understand baseball, football, or tennis, respectively. Or consider a
researcher trying to discuss factor analysis with a nonresearcher who has never heard
of the subject. Until both parties utilize shared concepts, communication cannot take
place. That said, it is important to note that concepts in the social and behavioral sciences often have contested meanings. As examples, after reviewing the scholarly literature, Fishbein and Ajzen (1975) found more than 500 definitions of attitude, and Jacoby
and Chestnut (1978) found more than 50 definitions of brand loyalty.
• Concepts are reality oriented (or functional). Although not physical reality themselves, most of our concepts presumably are tied to the external world and used as a
guide for interpreting and reacting to this world. Concepts are thus functional. If a person’s interpretation and labeling of experiences do not mirror the world, then his or her
reactions could be dysfunctional, even fatal. Consider the implications of conceptualizing a lethal cobra as a nonlethal garter snake. We develop and share concepts because
they seem useful for helping us understand the reality we experience.
• Concepts are selective constructions. The world we experience can be conceptualized in almost countless ways. For example, looking at a woman’s white blouse, we can
think of it as something that provides a socially expected degree of modesty, as something that offers protection from the wind and sun, as something decorati…