Sage Research Methods

Constructing Research Questions: Doing Interesting

Research

For the most optimal reading experience we recommend using our website.
A free-to-view version of this content is available by clicking on this link, which
includes an easy-to-navigate-and-search-entry, and may also include videos,
embedded datasets, downloadable datasets, interactive questions, audio
content, and downloadable tables and resources.

Author: Mats Alvesson, Jörgen Sandberg

Pub. Date: 2013
Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

Product: Sage Research Methods

DOI: [Link]
Methods: Doing research, Research questions
Keywords: research questions, knowledge, knowledge, personal experience
Disciplines: Anthropology, Business and Management, Criminology and Criminal Justice, Communication
and Media Studies, Counseling and Psychotherapy, Economics, Education, Geography, Health, Marketing,
Nursing, Political Science and International Relations, Psychology, Social Policy and Public Policy, Social
Work, Sociology
Access Date: August 2, 2025
Publisher: SAGE Publications Ltd
City: London
Online ISBN: 9781446270035

© 2013 SAGE Publications Ltd All Rights Reserved.

Page 2 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

The Context of Constructing and Formulating Re-

search Questions

As we argued in Chapter 1, most standard textbooks on research methods in the social sciences do not dis-
cuss (Denzin and Lincoln, 2011; Freebody, 2003) or only barely discuss (Hesse-Biber and Leavy, 2011; Sil-
verman, 2001) the actual ways of constructing and formulating research questions (a notable exception is
White, 2009). Even less attention is given to how research questions are constructed from existing literature
and practically nothing is written about how to construct innovative research questions that are likely to lead
to the development of more interesting and influential theories.

In Chapters 3 and 4 we specifically investigate how researchers construct research questions from existing lit-
erature. This is followed up in Chapters 5 and 6 where we provide an extensive elaboration of how to facilitate
sharper and more imaginative research questions. As a background to those investigations, our aim in this
chapter is to situate our study more precisely in the larger context of constructing and formulating research
questions. We begin by exploring the role that questions play in knowledge development. Thereafter we look
more specifically at what makes a question a research question, the major types of research questions that
can be asked, where research questions originate from and the major factors that influence the framing of re-
search questions. Finally, we distinguish the main stages and steps involved in constructing and formulating
research questions from existing literature.

The Priority of the Question in Knowledge Development

It may seem odd and perhaps even rude to ask about the role that questions play in developing knowledge,
as most textbooks about research methodology rarely devote any space to it. Perhaps they regard it as too
obvious or not important enough. Or they may think there are endless ways of coming up with research ques-
tions and therefore find it difficult to say something meaningful about it. With the risk of stating the obvious,
therefore, we think it is critical to spell out an answer, one which is simple but fundamental: questions are the
core ingredient in all knowledge development.

As Gadamer states, ‘the path of all knowledge leads through the question’ (1994: 363). Questions provide
the necessary starting point and path for all forms of knowledge development. It is by asking questions that

Page 3 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

we are able to generate knowledge about things. Similarly, asking questions forms the basis of every kind of
research investigation. In order to find out how students learn, educational researchers need to pose specific
questions, such as whether class size matters in learning and how students will go about learning specific
ideas and topics. Likewise, in order to find out how humans think, act and behave in particular ways psychol-
ogy researchers need to ask specific questions about human cognition and emotion. Hence, without asking
questions it is not possible to develop any knowledge, or at least not valuable knowledge. Research questions
therefore provide the inevitable and necessary starting point for, and have the priority in, all forms of scientific
knowledge development. They provide the basic direction and path in our development of knowledge, point
at which research design and methodology we should use, and define the theoretical and practical contribu-
tions our research is likely to generate. It is important to note that research questions can be understood in
terms of different levels and with more or less precision and focus. In this book we see research questions
not necessarily as very detailed questions or as specific objectives (close to testing an hypothesis). Instead,
we regard research questions as setting the somewhat broader intellectual motive of a study, whether it is
empirical and/or theoretical, that is, the rationale and direction of a study.

What Makes a Question a Research Question?

Although research questions are the core ingredient in all theory development, not all questions are research
questions. A common criterion for what defines a research question is that it must be ‘researchable’ (White,
2009: 35) or, as Savin-Baden and Major (2012) have said, ‘investigable’. Being researchable often means
that research questions have to be formulated in such a way that they can be investigated scientifically and
answered empirically. As White states: ‘Not all questions are social science research questions … other ques-
tions might be interesting to social scientists but cannot be answered using empirical evidence and so are
not “researchable”’ (2009: 35). Another common criterion is that a research question should not be too wide-
ranging, but instead confined and focused. As Silverman suggests, research questions should be able to say
‘a lot about a little’ rather than ‘a little about a lot’ (2001: 5).

The above statements seem to assume that only empirical investigations require a research question. The
recommendations are also directed at rather narrow empirical problems and investigations. However, re-
search questions are equally central to projects that have a stronger theoretical and conceptual research fo-
cus. We can ask questions that do not necessarily call for a new, specific empirical inquiry to be conducted:

Page 4 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

many good research questions imply a thorough, critical, imaginative and synthesizing approach to the the-
ories and empirical studies that already exist in a specific subject area. One could for example ask: ‘What is

the point of economic growth in countries that became “affluent” 50 years ago?’1. Or ‘Why is leadership such

a big research topic when people are more well-educated and claim to be more individualistic than ever?’2
These kinds of questions call for an approach other than focused empirical inquiry. On the whole, this book
aims to give a broader relevance to the theme of constructing and formulating research questions, and there-
fore does not limit itself to focused empirical inquiries such as those proposed by Silverman and others above.

Although being researchable and precise (or at least not too imprecise) is an important qualifier for a research
question, at least in empirical research, this is not the only aspect to consider and doing so is typically not
enough. Research questions should also be able to generate knowledge that matters to society and/or the
larger professional practices under study (for example, education, health care: Adler and Hansen, 2012; Van
De Ven, 2007). The point of research is of course not just to make it possible for academics to self-actualize
or boost their narcissism and careers through seeing their names on publications, although one sometimes
suspects that these kinds of motives do take the upper hand: research questions also need to have the po-
tential to address the central problems in society in some way or another.

There is however often a tension between the criterion of mattering and the criteria of having researchable
and precisely defined research questions, as well as between self-indulgent researchers and the interests of
society. As Adler and Hansen note, many scholars tend to ask well-defined and neat (and as a consequence,
sometime trivial) research questions ‘rather than dedicating themselves to investigating large, messy, com-
plex, controversial, and important societal issues’ (2012: 5) that have the potential to make a difference to
society. Research is often caught between rigor and relevance, and such debates remain evergreen in many
fields. As we describe more fully in Chapter 7, a central reason for researchers' unwillingness to address com-
plex and messy questions is their fear of deviating from what traditionally are seen as acceptable research
questions and, thus, reducing their chances for publication, research funding and career progress.

While a research question should be considered in terms of its being researchable, precise and able to gen-
erate knowledge that matters, perhaps the main qualifier for a really good research question is that it can
produce knowledge that has the potential to make a significant theoretical contribution. What a theoretical
contribution consists of will be discussed more fully in Chapter 3. But broadly speaking it is possible to distin-
guish between two basic forms of theoretical contributions: to advance existing knowledge by incrementally
adding to it or extending it in some significant way; and/or producing something more original and novel, for

Page 5 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

example, providing an alternative explanation or understanding of a specific subject matter (see also Corley
and Gioia, 2011). As stated in Chapter 1, we are particularly interested in how to construct research ques-
tions from existing literature that will lead to theoretical contributions which will challenge and differ from that
existing literature in some significant way. This is because (as will be developed in Chapter 4) consensus-
challenging knowledge is generally considered to be more interesting and to become more influential than
consensus-confirming knowledge, which modifies and advances but rarely alters or adds significantly to ex-
isting knowledge within the scientific field.

But what characterizes research questions that may lead to consensus-challenging knowledge contributions?
Such research questions are likely to be characterized by genuine openness. Being open creates unsettled-
ness and indeterminacy, which are crucial for more original and revelatory knowledge to be developed. As
Gadamer put it: ‘To ask a question means to bring into the open. The openness of what is in question consists
in the fact that the answer is not settled … The significance of questioning consists in revealing the question-
ability of what is questioned’ (1994: 363). Note here the double meaning of an ‘open question’ in the quote,
namely, that it is characterized by both interrogating the subject matter and questioning what we already know
about it. To be more specific, a genuinely open question addresses the subject matter anew and, at least
temporarily, questions what we already know about it; in particular, the assumptions underlying our existing
knowledge about it. Hence, innovative research questions open up and unsettle what we already ‘know’ about
a subject. Such questions therefore have the potential to challenge our current understanding of the subject
matter in some significant way and, thus, lead to the development of clearly novel knowledge.

It is important to note that some questions, which appear to be genuinely open, are in fact ‘closed’ questions
in the sense that they do not open up the subject matter and instead preserve it. The prime example of
closed questions is found in educational settings where the teacher pretends to ask students a genuinely
open question but already has a ready-made answer. Another common example of closed questions is the
kind of rhetorical questions often asked by politicians. Again, in such circumstances, the questioners will of-
ten know the answer but will want to temporarily unsettle their audience in some way or other. Unfortunately,
similar kinds of more or less closed questions are frequently posed, deliberately or non-deliberately, within the
social sciences, which to a large extent will tend to confirm what we already know (or believe we know). For
example, the question ‘How is gender being constructed among Spanish upper-class bisexuals?’, implies that
there are bisexuals constructing gender, and the question ‘What are the attitudes to divorce in East Timor?’,
implies that there is something fairly uniform in East Timor and that people there have (fixed) attitudes about
this issue. Yet another type of question that falls in between open and closed questions is what Gadamer calls

Page 6 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

‘slanted questions’ (1994: 364). A question is slanted if it suggests openness but is posed in such a way that
it does not, and therefore is not likely to, generate original and revelatory knowledge.

However, a research question can never be completely open: it is always limited by the specific perspective or
horizon from which it is posed. In other words, a research question comes from somewhere; it is always con-
structed and formulated within a specific perspective that conceptualizes and determines the subject matter
in a particular way. For example, questions about identity constructed and formulated within a psychological
perspective are likely to differ from the questions about identity constructed and formulated within a sociolog-
ical perspective. Similarly, questions about identity formulated within a specific theoretical perspective, such
as trait psychology, are likely to differ from questions about identity formulated within a Foucauldian perspec-
tive. This means that questions open up the subject matter in particular ways while at the same time they limit
the knowledge we can gain about the subject matter.

The fact that all research questions are constructed and formulated within certain frameworks, such as dis-
ciplinary, theoretical or methodological perspectives, but also within culturally taken-for-granted understand-
ings, is important in two ways. First, it shows the importance of becoming aware of our own perspectives and
how they affect the construction and formulation of our questions. This is because a particular perspective
allows us to ask certain questions and thus develop particular knowledge and theories. Second, and most
central in this book, it shows that a crucial element for developing more interesting and influential theory is
to direct our attention not only towards constructing and formulating questions per se, but also towards the
specific perspectives in which they are formulated and, in particular, the assumptions underlying those per-
spectives. Furthermore, to be able to formulate genuinely open questions, we may also need to deliberate-
ly challenge our own perspectives and cultural frameworks, at least temporarily. In other words, we need to
be aware of the perspectives and their assumptions that govern the construction and formulations of our re-
search questions. This is because the identification and challenging of assumptions underlying the perspec-
tives and cultural ‘truths’ within which we are situated precede the construction and formulation of research
questions. This is what the problematization methodology developed in Chapter 5 and exemplified in Chapter
6 will help us to do.

What Types of Research Questions Can Be Asked?

Although asking genuinely open questions is crucial for generating consensus-challenging knowledge contri-

Page 7 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

butions, it is also crucial to be able to distinguish and formulate different types of research questions. Several
different types of research questions can be asked (Hesse-Biber and Leavy, 2011; Van de Ven, 2007; White,
2009). Dillon (1984) provides perhaps the most comprehensive review of different types of research ques-
tions. He identified 17 different types of research questions. It is, however, possible to group these 17 ques-
tions into four more overarching types, namely: descriptive; comparative; explanatory; and normative ques-
tions. A central observation made by Dillon is that these questions are to a large extent hierarchically related
to each other in the sense that descriptive questions are the most basic, followed by comparative questions,
and then explanatory and normative questions.

Descriptive questions, or what Dillon called first-order questions, are the most basic and aim to find out what
makes up some phenomenon. Descriptive questions generate knowledge about what characterizes a phe-
nomenon, such as its substance (for example, what it is), function (for example, what it does) and rationale
(for example, why it has certain qualities). Examples of descriptive questions are ‘What characterizes hyper-
active people?’ and ‘How does student learning take place in an online environment?’ These questions focus
on the specific qualities that characterize hyperactive people and student learning, respectively.

The main purpose of comparative, or second-order, questions is to produce knowledge about the relations
among phenomena, such as concomitance (for example, to what extent two phenomena are related to each
other), equivalence (for example, to what extent two phenomena are similar to each other) and difference
(for example, how two phenomena differ). An example of a comparative question is ‘What similarities and
differences, if any, exist between male and female receivers of social welfare?’ As we can see, comparative
questions entail and presuppose descriptive questions because they cannot be meaningfully answered be-
fore answering descriptive questions about what characterizes female and male welfare recipients.

Explanatory, or third-order, questions aim to generate knowledge about the contingent relations between phe-
nomena and their attributes. Explanatory questions pursue correlation (for example, whether there is a con-
tingent relation between specific attributes of two phenomena), conditionality (for example, if that correlation
is conditioned by additional attributes) and causality (for example, if X produces a change in Y). Asking mean-
ingful explanatory questions about contingent relations presupposes ‘knowledge about comparative attribut-
es’ (Dillon, 1984: 331).

Normative questions aim to produce knowledge about how something should be done. Often normative ques-
tions ask what should be done in order to improve something, such as student learning, unemployment
among young people or acceptance of homosexuality in groups known for their hostility towards such a sexual

Page 8 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

orientation. Again, as we can see, to be able to pose and answer normative questions requires answers from
the previous types of questions.

The above typology highlights two key issues in constructing research questions. First, it highlights the im-
portance of being aware that specific types of research questions generate specific kinds of knowledge. For
example, if we want to generate knowledge about what makes up a phenomenon, we need to ask descrip-
tive rather than explanatory questions. Second, and equally important, it shows that research questions are
intimately related to each other in the sense that higher order questions entail lower-order questions. For ex-
ample, to be able to ask comparative questions presupposes that we will first answer a set of descriptive
questions.

It is important to note that a research question typically needs to be supplemented with a detailed and precise
statement of purpose (Creswell, 1998; White, 2009), as a way to further sharpen the focus of a study. The
purpose of a study postulates the researcher's intention with the study, that is, what the researcher wants to
achieve with the study, such as describing, explaining or evaluating something (Ritchie, 2003; Savin-Baden
and Major, 2012; White, 2009). We regard research questions as being broader and more general than pur-
pose. This is because while purpose stipulates and clarifies what the researcher wants to achieve with a
study, research questions frame the study in more significant ways. As stated above, research questions pro-
vide the overall direction and decide the path for the development of knowledge, as well as indicate what to
look for, how to design the study, what methods should be used and what contributions the study can poten-
tially make.

Where Do Research Questions Originate From?

A fourth major facet making up the context of constructing and formulating research questions is their origin.
Research questions are not constructed and formulated in a vacuum but originate from the specific places
from which the researcher gets their inspiration and ideas. While research questions are likely to come from a
variety of places the four most common (and interrelated) are society (Silverman, 2001), personal experience
(Easterby-Smith et al., 2008), existing scientific literature (White, 2009) and empirical material (Alvesson and
Kärreman, 2011).

There are numerous issues in society that provide inspiration and ideas for research questions. For example,

Page 9 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

as reported in the media, managers and employees in our companies, policymakers and administrators in
our various government bodies, teachers in our schools and health professionals in our hospitals constantly
encounter problematic issues. Additionally, general trends in society, such as climate change, the global fi-
nancial crisis, terrorism, major health issues, high unemployment, racism, changed migration patterns and the
digitalization of the economy, can inspire questions. Therefore, watching news and documentaries on televi-
sion, reading newspapers and specialist magazines, surfing the Internet, and more generally trying to stay in
touch with what is going on in society and the world, can be important for generating an interesting research
question. Using society as a source for constructing research questions is particularly critical to being able to
generate research questions that matter.

Nonetheless, as Silverman (2001) and others have pointed out, it is crucial for several reasons to exercise
caution when using broader society as the main source for generating research questions. It is not uncommon
for managers, policymakers, trade unionists and politicians to drive their own specific agenda when pointing
out some problematic issues in society. In other words, the issues communicated may appear to be genuine
but can in reality be heavily fabricated or twisted as a way to push forward a certain agenda colored by various
sectional interests, ideologies and political motives. And even if the issues raised are genuine they may be
hard to turn into investigable research questions owing to their complexity and messiness. There are also
often unrealistic expectations about the delivery of neat and reliable knowledge offering quick fixes. Howev-
er, as we said before, this does not mean we should avoid broader social issues. Rather, it means we must
carefully scrutinize to what extent it is possible to turn them into research questions, in particular into research
questions that can make a real difference to society (Adler and Hansen, 2012).

Another common source of inspiration for generating research questions is personal experience. Through
our personal experiences we are constantly exposed to various events that may function as sources in the
construction and formulation of research questions. While there are several forms of personal experiences,
three forms appear to be particularly central for generating research questions, namely work, life and study
experience. Research interests may emerge from observations and the experience of being a teacher, having
HIV/AIDS or having been in jail. Questions can also originate from a specific study experience, such as an
inspiring course that triggered curiosity about a particular topic. However, our personal experiences may not
only enable us to formulate specific research questions but also allow us function as filters that let in some
aspects but leave out others and potential research questions. Personal experiences can also lead to fixed
convictions and a belief that one already knows the answer to the question to be answered in the study.

Page 10 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

Sometimes personal and research experiences will overlap, such as when people gradually develop new un-
derstandings, interests and questions during a project, which can provide input for the next research project.
The output of one study can thus be the next study's input. For example, the output of a Master's thesis can
provide input for a doctoral thesis. Similarly, a research project will often identify some unexamined issues
and in many cases it is only at the end that the researcher recognizes the interesting new idea for study.
Sometimes new research ideas are a matter of building directly on earlier reported results. But often there are
other forms of outcome, such as the identification of questions that were outside the focus of the project and/
or could not be answered, but provided a good starting point for later work. One could actually say that most
good research leads to some (preliminary) answers and some new questions.

Although society and personal experience are important sources for developing research questions, existing
published studies and influential theory are often the most common as well as the most important source for
generating research questions. This is so for a variety of reasons. First, even if a research question originates
elsewhere (either as a societal issue or from personal experience) a researcher needs to engage with existing
literature to further elaborate the tentatively formulated research question. The researcher needs to review
the existing literature to make sure that the formulated research question has not already been answered.
Equally important, is that even if the formulated research question has not been addressed in the existing
literature, this does not automatically mean that the researcher should pursue it. This is because there may
be good reasons why it has not been addressed before, such as it may be unlikely to generate scientific
knowledge. Existing studies and theory also help with finding the right framing and focus. How others have
structured the field and what they managed to accomplish will be helpful for finding the right scope, sample
and level of ambition in one's own study.

Second, the existing literature in itself provides a crucial source for generating research questions. Reviewing
the existing literature in a particular area, such as higher education, family conflict or human aggression, en-
ables us to tap into key debates and see which issues are central in the field, those that can function as
important sources for constructing research questions. These questions need then to be further specified
through a more targeted review of the existing literature as a way to see to what extent the question has
been addressed, and how. In particular, the researcher needs to identify potential deficiencies and shortcom-
ings in existing research that will enable him or her to develop questions that are likely to generate valuable,
new knowledge. The critical scrutiny should not just focus on domain-specific issues or knowledge areas, but
should also involve theoretical perspectives that the researcher is using or considering using, for example,
psychoanalysis, symbolic interactionism or Marxism. A familiarity with critiques and debates is vital. Scrutiny

Page 11 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

may include both relevance/usefulness as well as possible problems and shortcomings in the theory, in gen-
eral, or with a specific bearing on the subject matter of a study.

However, as White (2009) rightly points out, while the existing literature in an area can offer a rich source for
constructing innovative research questions, it can also stifle creativity and innovation, as it tends to confine
one's thinking. As the conclusion in Chapters 3 and 4 shows, generating research questions through build-
ing positively on existing studies leads mainly to the development of incremental knowledge contributions
rather than more revelatory knowledge contributions. Therefore, the major part of this book addresses how
to identify and challenge the assumptions underlying the existing literature and, based on doing so, formulate
research questions that can lead to the development of more interesting and influential theories (see in par-
ticular Chapters 5 and 6).

Empirical material being produced in the research process can also function as a central source for research
questions. In some cases, particularly in qualitative studies, being inductively oriented and open towards the
views and meanings of those being studied may encourage revisions of the initial intention behind a study.
Expectations may have been misleading. A researcher may want to study conflict and cooperation in team-
work and expect to find it, but then the actual research may show that the ‘team’ idea in this context is a
management construction, that they are unaware of what actually goes on, and in fact workers tend to do
the job fairly autonomously with rather limited contact with each other. Problems with access or unexpected
changes in the object of study may also force revisions of the project.

Sometimes unexpected empirical material may have positive consequences, such as the researcher being
able to generate more innovative research questions than those initially formulated. Often one is involved in
specific empirical investigations about a topic when one sometimes suddenly ‘sees’ a potentially more inter-
esting research question. (The case just mentioned about ‘sham team work’ could have started as a study
of how workers hide work practices from management or, more generally, the existence and structuration of
‘sham structures’ in organizations.) This source may be more common for experienced researchers who are
involved in various kind of empirical research. From the empirical material or the collection of empirical mate-
rial, researchers may encounter issues that surprise them, deviate from expectations and so on. We discuss
this further in Chapter 8.

There is of course always an initial research question but it is seldom set in stone, and even if it continues
to guide the research to the end of the project, it is often to some extent influenced by the empirical setting
and the gradual insights generated by the empirical results. In some cases empirical material may lead to

Page 12 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

more significant revisions. Although it is good in principle if the question precedes the search for answers, an
interaction between questions and answers coupled with the search for a good research question during an
exploratory research phase – and occasionally later – may be productive.

It is important to note that the formulation of research questions is often not a result from one specific source
but one from an interaction between multiple sources, although the existing theory appears as the center of
the interaction. This is because even if a research question originates from elsewhere, researchers still need
to engage with the existing literature to make sure that it has not already been addressed or partly addressed
and must then frame the research question in such a way that it will produce knowledge that makes a distinct
theoretical contribution. In the interaction, the different sources can also temper each other's weaknesses.
For example, we may be able to generate a promising research question from existing literature but then we
also need to ask to what extent it matters to society. Similarly, if we are able to generate a research question
based on some societal issues we need to consider to what extent it will add something of scientific value. If
it does not do this, we will need either to reformulate the question or drop it. In some cases, in particular in
qualitative research (often being more iterative), empirical material and findings may encourage re-readings
of the literature and this can give rise to different ways of engaging with the literature in constructing and for-
mulating a research question.

Factors That Influence the Construction and Formulation

of Research Questions

Regardless of their origin research questions are never constructed and formulated in isolation but always
within the broader scientific fields and cultural frameworks in which the researcher is situated. As we will ex-
plain more fully in Chapter 7, scientific fields consist of a range of contextual factors and broader societal cul-
tural traditions (for example, views on gender, authority, equality, community), which influence the construc-
tion and formulation of research questions in various ways.

One such contextual factor is fashion and fads in society and the social sciences. Popular topics often point
to some critical problems in society or in a discipline that appear worthwhile for investigation. For example,
fashionable topics in today's society are social media such as Twitter and Facebook, as well as the Internet
more broadly, and how they affect our lives. Similarly, emotion, identity and workplace diversity are topics that

Page 13 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

are currently receiving a lot of attention in many disciplines.

Fashionable topics, like the above, tend to get a lot of attention in the media and in public debates and there-
fore appear to be important and worth studying. They have a propensity to direct and encourage researchers
towards specific inquiry domains, and within them to construct and formulate specific research questions.
However, just because an issue is fashionable does not mean that it is worth studying. The mass media often
exaggerate trends and changes and like to emphasize drama. It is therefore vital not to automatically accept
what is trendy but to critically scrutinize to what extent such topics may be worth studying. Often less fashion-
able but socially very important topics are persistent and it may be better to address them than look at what
is considered to be new and hot. Another problem here is that what is viewed as a hot topic today may be
cold tomorrow. A research contribution should ideally have some value over time. In addition, as it takes time
to carry out research and to get it published, at least in scientific outlets, there is a risk that a trend-sensitive
topic guiding a project may have lost its appeal when it is finally ready for publication, perhaps several years
after the start of the project.

Another contextual factor that is likely to influence the formulation of research questions is the availability of
research funding. Often funding bodies will have certain priority areas (sometimes influenced by what is in
fashion), and if a research project falls inside such a priority area, it may increase the likelihood of getting
funding. It is therefore not unusual that these priority areas influence how researchers construct and formu-
late their research questions in the sense that researchers will try to frame the research question so it seems
to address a priority topic. Journals exercise a similar influence in the sense that they will have specific poli-
cies about what kind of research they publish and will often issue calls for papers on specific topics (to form
special issues), which may influence the questions a researcher might ask. Often researchers will believe it
is easier to get published in a special issue and then re-orient their work (or simply change some keywords
and do some polishing) to try to get into it. However, while being opportunistic when it comes to applying for
research funding and addressing a trendy topic may have some short-term benefits, such opportunism may
be less positive in the long run and is therefore always a bit risky.

Another important contextual factor that tends to influence the formulation of research question is other re-
searchers. It is common that more senior researchers within a school or department will influence the choice
and formulations of research questions among doctoral students and more junior academics. For example,
supervisors will tend to heavily influence their doctoral students' research questions. However peers who
review research proposals and submitted papers may also influence how we construct and formulate our

Page 14 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

research questions. A highly interrelated contextual factor is researchers' career prospects. As mentioned
above, many researchers tend to adjust their research questions to meet the criteria stipulated by funding
bodies and journals, as a way to maximize the possibility of receiving research funding, getting published in
good journals and, thus, climbing the ladder more quickly. Moreover, in order to increase the possibility of
rising academically, many researchers will appear to avoid messy and complex questions and instead stick
to doable – or even ‘quick and dirty’ – research tasks (Adler and Hansen, 2012). In particular, they seem to
construct incremental and narrow research questions, which, as we will show in Chapters 3 and 4, are likely
to lead to the development of less interesting and less influential knowledge contributions.

Another major influential factor is the researcher's specific knowledge interest. Habermas (1972) pointed out
three basic forms of knowledge interest, namely, a technical, a practical-hermeneutic and an emancipatory
knowledge interest, which guide scientific research and thus also the formulation of research questions. A
technical interest, which may be aligned with functionalist research, aims to develop knowledge of cause-
and-effect relations through which control over natural and social conditions can be achieved. This approach
is typically oriented towards discovering law-like patterns and offers instrumental knowledge that suggests
how to accomplish the desired outcomes. For example, how we can improve institutional management, how
to find out which teaching technologies are most effective and how to make interventions to prevent unwant-
ed and deviant behavior. The practical-hermeneutic knowledge interest is more associated with interpretive
scholarship. In contrast to a technical knowledge interest and its focus on prediction and control, a practical-
hermeneutic knowledge interest emphasizes enriched understanding of human cultural experiences: how we
communicate, generate and transform meaning. While interpretive scholars are fairly open with regard to how
such knowledge can be used, they typically refrain from normative statements. The emancipatory knowledge
interest comes from a more critical orientation often associated with various critical theory approaches. Here,
attention is focused on power relations and revealing them in ways that can liberate the subject from various
repressive relations that tend to constrain agency. This knowledge interest is about the critical examination
of institutions, ideologies, interests and identities with the aim of encouraging critical insight, rethinking and
liberation from, or resistance to, unnecessary social forms of domination. A closely related factor that also sig-
nificantly influences the framing of research questions is the specific disciplinary, paradigmatic and method-
ological perspectives in which the researcher is situated, as discussed earlier.

Page 15 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

The Process of Formulating Research Questions

Most research methodology textbooks conceptualize the process of formulating research questions as pre-
dominantly linear (Creswell, 1998; Savin-Baden and Major, 2012; White, 2009; Van de Ven, 2007). Although
different authors distinguish the steps involved in the formulation of research questions in various ways, the
formulation process is usually conceptualized as consisting of the following sequential steps. First, is to iden-
tify the domain or subject area with which we are interested, such as learning in higher education, domestic
violence, youth unemployment and motivation. Then, we are advised to identify a specific topic within the do-
main or subject area, such as online learning in higher education or female violence among elderly couples.
Thereafter, we should formulate a research problem in relationship to the chosen topic, with both a theoret-
ical (for example, it is an area that has been largely overlooked) and practical (it is something that matters
in society) motivation, that is, the rationale of the study. This is then often followed by the formulation of the
purpose of the study, which clearly states what the study is about. Finally, we should specify the more precise
direction of our research with a set of research questions.

While we partly agree with this description, we also think it is problematic as it leaves out several critical
stages in the process of constructing and formulating research questions. First, we contend that the process
is far more iterative and freely evolving than a linear progression allows. There appears to be, or at least ought
to be, much more interaction between the sources from which research questions originate. Second, the ex-
isting literature seems to play a considerably more central role than the other sources in the formulation of
innovative research questions. This is because even if a research question originates from another source the
researcher still has to engage with the existing literature as a way to (1) generate a knowledge contribution,
(2) achieve a firm base in what has been done previously, as well as drawing upon theoretical perspectives,
and (3) make sure the construction and formulation of research questions do not only happen at the begin-
ning of the research project but sometimes after an exploratory phase and occasionally even throughout the
research. Sometimes the collection of empirical material and the initial interpretation of it can generate new
insights that may require a reformulation of the research questions (Alvesson and Kärreman, 2011). As we
highlight in Chapter 3, another crucial stage is the actual publication phase in which the research questions
are finally formulated. It is, for example, not unusual for editors' and reviewers' comments to require a refor-
mulation of research questions, as a way to further specify what kind of theoretical contribution the answers
to the research questions have generated.

Page 16 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

Conclusion

The aim of this chapter has been to situate our investigation of how researchers can construct research ques-
tions from the existing literature within the larger context of constructing and formulating research questions.
We identified five major facets making up the context of generating research questions through the existing
literature, namely: (1) the priority of questions/questioning in all knowledge development; (2) what makes a
question a research question; (3) the major types of research questions that can be asked; (4) where do re-
search questions originate from; and (5) the major contextual factors that influence the framing of research
questions. Finally, we pointed to the evolving and iterative nature of generating research questions through
existing literature. In summary, a consideration of the following issues seems particularly important when we
are constructing and formulating research questions:

• Engagement and curiosity are important between strong commitment and openness, but be wary of
conflict (as an engagement may lead to premature convictions and lock the project).
• Formulate questions that are open but also provide direction.
• Make certain of working with research questions, not just socially relevant or journalistic questions.
• Think through the knowledge interest, and what is the ultimate goal: instrumental value, enriched un-
derstanding, liberation through critical insight?
• Relate carefully and critically to the literature – do not just ‘blindly’ build positively on it.
• Funnel the research question – avoid too broad and vague questions.
• Beware of fashions and trends – research is a long-term oriented field.
• Be open to and reconsider what is to be done in the light of empirical findings.
• Be reflective about how broader contextual factors (for example, fashion, funding opportunities, jour-
nal policies and career prospects) may influence the construction and formulation of research ques-
tions.

In Chapters 3 and 4 that follow, we narrow our focus and specifically investigate the ways social science re-
searchers construct and formulate research questions from the existing literature, and to what extent these
ways are likely to result in the development of more interesting and influential theories.

1The term ‘affluent’ was coined by Galbraith in 1958.

2For an effort to illuminate these research questions see Alvesson (2013a).

Page 17 of 18 Constructing Research Questions: Doing Interesting Research

Sage Sage Research Methods
© Mats Alvesson and Jörgen Sandberg 2013

[Link]

Page 18 of 18 Constructing Research Questions: Doing Interesting Research