Problem Selection in the Social Sciences: Methodology
Wout Ultee, Radboud University Nijmegen, Nijmegen, The Netherlands
Ó 2015 Elsevier Ltd. All rights reserved.
Abstract
Sociological methodology has rules for testing hypotheses more severely, and for the formulation of theories rich in content.
This article presents rules for problem selection by surveying cases from sociology’s past and present. Among these rules are
treating a seemingly small question as a subproblem of a larger one, expanding existing problem structures by adding new
subquestions, revising a problem structure by a new overarching question, picking out contradictions between theories and
research findings, and replacing false assumptions behind questions.
Social methodology textbooks state techniques of proper
empirical analysis. For instance, they tell how partial coefficients
of association are to be computed, adding that these
coefficients provide a more severe test of multifactor explanations
than bivariate associations. Books on theory building
in sociology set standards for determining whether certain
hypotheses are better than others. Logically unfalsifiable ones
are out; future propositions should provide more information
than current ones. Any science starts with problems. Are there
precepts for selecting problems too? Yes, and this article draws
upon principles intimated by sociologists (Lenski, 1966: p. 20;
Blau and Duncan, 1967: pp. 9–10, 194–199, 402), proposed
by philosophers of science (Bunge, 1967: pp. 165–221), and
applied by sociologists to sociology (Merton, 1959; Ultee,
1980). These standards are elucidated by cases from
sociology’s past and present.
Overarching Problems and Subproblems
Durkheim ‘was not interested in suicide at all,’ so says Collins’
(1994) textbook on sociological traditions. But then, why did
Durkheim dub his 1897 study Le Suicide? Weber in 1910
grumbled that his critics understood him to explain the rise
of capitalism fully from the ethos of Protestantism, and
retorted that he only dealt with the question of whether this
ethos fitted the spirit of entrepreneurial capitalism. Yet in the
1920 preface to his collected papers on world religions,
Weber’s interest was rationalization processes, including the
rise of capitalism. These changes had gone furthest in the
West, and Weber wondered why.
If these two stories about sociology’s questions carry
a moral, the lesson is not that persons nowadays regarded as
sociology’s founders, did not know what they were up to. The
message amounts to a methodological command on meritorious
problems. Durkheim grasped that seemingly small
questions become pertinent when taken as subproblems of
larger ones, and Weber understood that it is wise to divide
mammoth problems into more modest ones, and those in their
turn into even more tractable questions. Durkheim and Weber
entered sociology’s hall of fame because they pointed toward
a bunch of smaller questions falling under one and the same
heading. The subtitle of Le Suicide runs Étude de sociologie, lifting
a phenomenon from psychology into that science which
studies societies – and the extreme case of leaving them.
Problem Structures and Their Dynamics
For Durkheim, a society’s suicide rate tells about its cohesion;
the higher this rate, the less cohesive a society. Weber observed
that a society’s inhabitants do not always use their economic
freedoms, for instance, those of enterprise and labor, and
supplemented the problem of how these rights got included
in a society’s code of law, with the problem of the strength
of a person’s motivation to use these liberties. Both questions
were part of the question of why entrepreneurial capitalism
arose in the West. In its turn, the question of the rise of capitalism,
including bankers’ as well as entrepreneurial capitalism,
was a subquestion of the even larger question of the speed
of rationalization processes in the West during the last couple
of centuries.
In Durkheim’s oeuvre, the question of a society’s degree of
cohesion also comprises long-term trends in the prevalence of
economic exchanges when comparing Ancient Israel, the Roman
Empire, and medieval Burgundy, and the degree to which
assemblies of hunters and gatherers are religious in nature.
Weber’s question of rationalization led to less grand ones on the
rise of science and the bureaucratization of states. Any methodology
of problem selection should contain the rule that questions
should not be judged on their own, but according to their
place in a structure of overarching problems and subproblems.
However small a problem seems, it has more appeal if it is part
of an overarching problem. And however towering a question
appears, it grows less scary if it is broken down into subquestions.
The Durkheimian Problem Structure
Figure 1 depicts the Durkheimian problem structure. The
novelty of Le Suicide was not simply that Durkheim turned the
question of suicide into a subquestion of the larger one of
societal cohesion. Hobbes (1651) took order and its opposite
violence as the master problem of those studying the body
politic. For him, civil wars and wars between sovereigns were
part of the problem of order and violence. Next to it,
Durkheim placed the problem of the presence and absence of
ties between a society’s members. The counterpart of the
question ‘why do people not always live in peace?’ is not
only ‘why do they sometimes use violence?’ but also ‘why do
they sometimes live relatively isolated from each other?’ For
Durkheim, suicide was the supreme example of severing the
ties between persons and the society in which they live.
International Encyclopedia of the Social & Behavioral Sciences, 2nd edition, Volume 19 http://dx.doi.org/10.1016/B978-0-08-097086-8.44043-2 49
Apart from Le Suicide’s question, Figure 1 encompasses the
questions guiding Durkheim’s two other main studies. De la
division du travail social from 1893 focuses the frequency of
economic transactions between a society’s members at various
points in history. It narrows this question down to that of the
degree to which cooperative law (civil law regulating
economic contracts by restitution) is more widespread than
criminal law (penal law involving repressive sanctions). Les
formes élémentaires de la vie religieuse from 1912 notes that
Australian aboriginals disperse when hunting and gathering
their food, and asks to what extent their leisurely meetings
celebrate a totem and turn them into members of one society.
Durkheim’s questions involve a shift from questions on
economic ties, by way of questions on coliving, to questions
on religious ties, yet they always pertain to societal cohesion.
Given the Durkheimian problem structure, later scholars
can raise questions on types of attachment to which Durkheim
paid limited or no attention. Examples are questions on
changes in the rates of marriage, birth, and divorce in industrial
societies. A wedding creates a sexual and domestic bond
between two members of the opposite sex, divorce dissolves
it, and a higher birth rate amounts to more parent–child relations
in a society. focuses on the making and undoing of sexual
and domestic ties. These monographs amount to a new branch
of the Durkheimian problem structure.
The Weberian Problem Structure
The Weberian problem structure and its dynamics are given in
Figure 2. The overarching problem is that of rationalization,
and Merton (1938) pursued the subproblem of the rise of
science in the West, and Blau (1974) that of bureaucratization.
The art historian Gombrich (1960) studied questions on the
skills of painters to furnish a faithful representation.
Societal cohesion
Order and violence The frequency of
economic
transactions
(Durkheim, 1893)
The frequency of
religious
gatherings
(Durkheim, 1912)
The rarity
of suicide
(Durkheim, 1897)
The frequency of
sexual and
domestic bonds
Marriage rates
(Goode, 1963)
Civil wars Wars between
sovereigns
(Hobbes, 1651)(Hobbes, 1651)
Divorce
rates
(Goode, 1993)
Birth
rates
Figure 1 The Durkheimian problem structure and its dynamics.
Why are all kinds of
rationalization processes more advanced in the West around 1900?
(Weber, 1920)
The bureaucratization
of the state
(Blau, 1974)
The development of
representational skills
in the arts
(Gombrich, 1960)
The emergence
of science
(Merton, 1938)
The rise of capitalism
The rise of
bankers
capitalism
The rise of
entrereneurial
capitalism
The emergence of
free enterprise
and free labor
The motivation
to use the freedoms
of enterprise
and labor
(Weber, 1906)
Figure 2 The Weberian problem structure and its dynamics.
50 Problem Selection in the Social Sciences: Methodology
The relation between problems as depicted in Figures 1 and
2 are logical in nature. An overarching problem, usually by way
of uncontroversial auxiliary assumptions, implies subproblems.
It is another, and less important, matter whether a person
raising a subproblem of an overarching problem of an earlier
scholar studied under that master or read a book by her or
him. Merton and Blau took the lead from Weber; Gombrich
was not influenced by him. Weber invokes Heinrich Wölfflin
once when discussing the question of representational skills;
Gombrich quotes Wölfflin frequently.
As stated, the step from an overarching question to a subquestion
involves largely undisputed assumptions. Few would
argue that capitalism is less efficient than slavery. Yet this
proposition was belied by Fogel and Engerman’s (1974) test
for the American South shortly before slavery was abolished.
A scholar working on a question can make progress by
replacing these auxiliary assumptions by better approximations.
Weber did not do so, but Durkheim did. In countries
prescribing suicide under certain circumstances, suicide is not
an instance of low cohesion, but of strong solidarity. In that
case, Durkheim spoke of altruistic suicide.
Behind Weber’s distinction between bankers’ and entrepreneurial
capitalism lurks evidence speaking against the
hypothesis that the rise of capitalism is always accompanied by
the growth of banking. Protestant businessmen did not take
out loans. Their religion put a premium on frugality, and
contributed to reinvestment of profits in their enterprise. The
background of Weber’s distinction between laws granting
people economic freedoms and the use of these freedoms by
a society’s members forms the observation that people sometimes
do not use their rights, as well as the hypothesis that they
exercise these rights only if they are motivated to do so. Classical
economics assumed that every person had this drive;
Weber held that Protestantism strengthened it. A wise rule is to
state the background of a question as fully as seems necessary
for bringing out a problem structure.
The Weberian Problem Structure Radically Revised
The division of an overarching problem into subproblems may
drastically alter. Studies by Smith (1776), Elias (1939),
Samuelson (1954), Van den Doel (1979), and De Swaan
(1988) imply the revision of the Weberian problem in
Figure 3. Weber never made clear what state bureaucracies
were doing. States possess the legitimate monopoly on the
means of violence; Elias added that states wield the
legitimate monopoly on taxation. Were all taxes spent on
armies, police, and prisons? This obviously is not the case.
Here, the hypothesis is useful that markets make for the
optimal production of individual goods (Smith), but not for
the optimal production of collective goods (Samuelson). The
production of collective goods approaches optimality in
a state with universal suffrage and a bureaucracy that executes
its laws (Van den Doel).
De Swaan listed collective bads and laws contributing to
their elimination and to the production of collective goods.
First, poverty is a collective bad. Poor people are more likely to
trespass the law, and crimes form a general threat to life and
property. Following De Swaan, this danger is alleviated by state
insurance against disability and unemployment. Contagious
diseases amount to a second collective bad. If local authorities
provide each household with piped water and an outlet on the
sewers, public health improves. Third, compulsory schooling
lowers the percentage of children who cannot read and write,
and makes for a more optimal functioning of markets and
bureaucracies. Weber’s conception of the state is saved by the
assumption that armies, police, and prisons avert collective
bads. Armies lessen a threat to a state’s independence, police
patrols contribute to safety on the streets, and prisons deter.
Problem Structures and Theories
As the saying goes, a question well put is a question half
answered. If it is clear that some problem is part of a larger
problem, and if a solution for the overarching problem is at
hand, a possible solution to the smaller problem is available
too. But then, if an answer to a particular smaller question of
a larger one turns out to be false, the opportunities to disestablish
a solution to another subproblem of the larger one
increase. Additionally, the notion that problem structures exist
and have a dynamic makes the project of sociology have more
continuity than suggested by Boudon (1979) and Gove (1995).
Conditions for rationalization
(Weber, 1920)
Bureaucratic democratic states
and the optimal production of
collective goods
(Samuelson, 1954; Van den Doel, 1979)
Markets and the optimal
production of
individual goods
(Smith, 1776)
Poverty and legislation
for unemployment
and disability
benefits
(De Swaan, 1988)
Public health
and piped
water plus
sewers
(De Swaan, 1988)
Compulsory
schooling
(De Swaan, 1988)
Armies plus
police and
general
safety
Figure 3 The Weberian problem structure radically revised.
Problem Selection in the Social Sciences: Methodology 51
They accepted that sociology’s questions to outsiders show
a bewildering variety, and bypassed the possibility that one
and the same larger question may harbor a host of seemingly
disparate questions.
Contradictions and Issues
One way of formulating a problem in a more useful way is to
ascertain its place in a problem structure. Are there much
injunctions for problem choice? Sociologists debating sociology
have argued about the primacy of theory versus the
supremacy of research. In contrast, Popper (1962) defended
that every science, including sociology, most fruitfully starts
with contradictions between general hypotheses and specific
research findings.
Contradictions – or, to use (Kuhn’s 1962) terms, puzzles
and anomalies – contribute to sociology’s growth. For
Merton’s (1973) sociology of science, the prevalence of priority
disputes in a field was surprising in several respects. Its
explanation was not the egotism of the scientists having
made discoveries, since their part in the disputes was small.
Nor was it to be accounted for by the increasing competition
between scientists, since the frequency and intensity of
priority disputes in the West decreased over time. A rule of
method runs: single out contradictions.
Incompatibilities between hypotheses also invite new
research and theorizing. They sometimes are called issues, but
that term also refers to false alternatives. An example is the
question of whether intelligence is determined by social or by
genetic factors. Intermediate positions are possible, the list of
social factors may be incomplete, and the location of intelligence
on genes may be unknown. Popular science states
issues in a metaphorical way; are we born with a brain
hardwired with rules for language, or is the brain a big
sponge, capable of creating rules for language by mopping up
what we hear?
The phrasing of a contradiction problem sometimes
contains seeds for its solution. For instance, Huizinga (1941)
asked how it was possible that painting and other forms of
culture flourished in the Netherlands during the seventeenth
century, so shortly after this society became independent
from Spain. His answer was that Dutch culture blossomed
because the Netherlands was a new state. In states that have
recently become independent, the upper strata have risen
from the lower and the more open a country’s elite, the
more its culture will flower. Trevor-Roper (1976) wondered
how it was possible that an emperor as preoccupied with
expanding his territory as Charles V (1500–58) furthered
the art of painting so much. In the end, Trevor-Roper
surmised that Charles V commissioned paintings to the
extent that, and precisely because, they legitimated his political
power.
Apart from the rule for theory building to mobilize the
information in a contradiction problem, there is the precept
to turn the problem around. Rather than asking why the
number of unruly passengers on airlines is increasing, it
might be worthwhile to ponder why this number until now
did not increase much more strongly. A more academic
example is as follows. Marxism noted that a lot of workers
refrain from joining unions and strikes, and raised the question
of how workers come to have a false consciousness.
Olson (1965) held that unions and strikes produce collective
rather than private goods, applied the prisoners’ dilemma
from game theory in mathematics, predicted that rational
workers would not strike, and raised the question of why
people participate in social movements. However, it should
be clear that any theory with some substance to it should
explain that some people join labor unions and certain
others do not.
Strings of Questions
The terms problem and question sometimes are used interchangeably,
referring to queries of any specificity. However,
everyday usage of expressions like object of investigation, task,
question, and problem suggests that some inquiries are more
focused than others. What does the articulation of problems
involve?
Problem Articulation
If it is maintained that sociology is about societies (and not
about human beings taken on their own), it sometimes is said
that the object of research has been stipulated. Yet no single
feature of this unit has been singled out for explanation. If this
property is taken to be, say, the ‘functioning’ of a society, the
object of inquiry turns into what has been called a task. Yet the
circumscription of this aspect remains limited, since the term
does not focus any particular way of functioning. If a distinction
is made between optimal and suboptimal functioning, the
study involves a question, with that term being taken in
a narrow sense. Of course, other types of functioning may be
distinguished too. Questions in their turn do not allude to
possible causes. If research on some sample of societies determines
the extent to which less stability in a society’s environment
goes together with a less optimal functioning of this
society, the matter at hand is articulated so fully that the
problem now is whether some logically falsifiable hypothesis
is true. Putting a question mark behind a falsifiable proposition
amounts to posing a problem in some narrow sense of that
word.
The example of a string of increasingly articulated queries
just presented neglects any difficulty in obtaining measurements
for the degree of stability in a society’s environment, and
the extent to which the functioning of a society approaches
optimality. It has been held that questions should refer to
directly observable phenomena only, and in the human
sciences therefore to outward behavior and not to mental
states. This rule is a leftover from a wrongheaded empiricist
philosophy of science. All terms to some extent are theoretical,
and any proposition whatsoever, whether about acts or
thoughts of persons or processes and situations within of
societies, never can be definitely proved.
From Descriptive to Explanatory Questions
There are strings of questions of another kind. A researcher’s
early questions often are descriptive: how many of the people
52 Problem Selection in the Social Sciences: Methodology
living in 1972 Britain and born to working-class parents had
moved up to the higher classes? A follow-up question might be
whether this probability to climb equals the chance to descend
from the upper to the lower classes. A yet later question
presupposes the answers to both earlier questions, and seeks to
ascertain whether upward and downward mobilities were as
prevalent in 1949 as in 1972. By raising additional descriptive
questions, narratives become richer.
At some point, researchers leave descriptive questions
behind them, to raise explanatory questions. The explanatory
question following up on the previous string of descriptive
questions ran as why was during the 1970s upward mobility
less widespread in Britain than in Sweden, and more widespread
than in France? The obvious answer to that question was
that left-wing governments make for more upward mobility
(Heath, 1981; Goldthorpe, 1987). And after a comparison of
upward mobility in several industrial societies at the same
point in time, the follow-up question may be raised how
these societies compare over longer periods (Erikson and
Goldthorpe, 1992; Breen, 2004).
From Superficial to Deeper Explanatory Questions
Some explanations ‘scratch the surface only,’ others delve
deeper. Why would particular ideologies make for more
upward mobility? Just like individuals, governments cannot
always attain their goals. By which means then do socialdemocrats
attain the goal of more upward mobility? Does
Sweden have more upward mobility than Britain, because
social-democrats make children branch out at a later age to
schools that demand more or less of pupils? Or is this the case
because social-democrats pursued a policy of full employment?
And then, if the answer to the first question is ‘no’ and
that to the second one ‘yes,’ why would there be more
upwardly mobile persons in a society without unemployed
persons?
There is no obvious point at which the hunt for deeper
explanations should stop. Some theoreticians in effect defend
that ‘the buck stops here.’ No one nowadays seriously defends
that propositions about societies are logically independent
from propositions about individuals. However, whether rockbottom
explanations of the features of societies always
should refer to the motives and aversions of a society’s
members, as well as the opportunities they have and the
barriers they face, is controversial. In the discussion on
‘methodological individualism’ (O’Neill, 1973), the borderline
between methodological rules and substantive requirements
was violated. Perhaps it is wise to no longer state this debate
in terms of theories explaining societal phenomena, but to
settle it by recognizing that short-hand statements of sociology’s
questions refer to societal features only, but that these
questions, if they are stated fully, always refer to the persons
making up a society’s population, that is, to individuals and
societies (Ultee, 1998).
From Meager to Thick Descriptive Questions
Some descriptive questions – and their ensuing explanatory
questions – can be meager, whereas others aim for
lavishness. Thus, it often has been ascertained that no society
ever has been and ever will be completely egalitarian. As Lenski
(1966: p. 20) maintained, in the debate between the
conservative and the radical tradition in sociology, it is
possible to make headway by transforming problems stated
in a categorical way into problems phrased in variable terms.
It may indeed be the case that full equality never has been
observed in any society; however, is the lowest degree
of inequality observed until now the lowest that is
possible? And do long-standing differences between highly
industrialized countries in their ruling ideology explain
differences in their degree of income inequality? Questions
about equality that demand ‘yes’ or ‘no’ are less precise than
questions about degree of inequality.
Following Lenski, progress also can be made by breaking
down the compound concepts contained in problems into
their constituent elements. Thus, the question of whether
power – in the singular – determines privilege – in the singular
– will not do. Rather, the question is to what extent differences
in the consumption of a certain good in a particular society
depend on the control of various resources. When answering
this question, Lenski distinguished the skills of hunters and
gatherers, the size of gardens in horticultural and of fields in
agrarian societies, the possession of metal weapons, the private
ownership of machines, the technical skills of the people
designing machines, the range of legitimate command accruing
to a political position, and the right to vote in elections. Lenski
also suggested that the change from agriculture to industry was
accompanied by a shift in the right to vote. Once elections were
infrequent with only the propertied classes having the right to
vote, whereas nowadays there is universal suffrage for periodic
elections. By breaking down compound problems into their
constituents, thick descriptions and detailed explanations are
obtained.
Erroneous and Well-Stated Questions
If descriptive questions are logically prior to explanatory
questions, explanatory questions may be wrong. Answers to
questions can turn out false, questions themselves may be
mistaken. The question of why in the 1970s social mobility
was more widespread in France than in Sweden is erroneous,
since in that decade mobility occurred more often in Sweden.
The Olson question of exactly which selective incentive makes
persons protest against the hunting of seals assumes that some
selective incentive is always at work, which may not be the
case. This example suggests that if a rock-bottom explanation
slips into a question, that problem turns into leading
question.
Methodological rules that stipulate conditions banishing
erroneous questions and guaranteeing well-stated ones seem
doomed. It is wise to review the presupposed description, but
checking has to stop somewhere. In this respect, it is interesting
that the answer to an explanatory question always implies that
the original question may not have been fully right. When
Durkheim asked why Protestantism goes together with more
suicide than Catholicism, he devised the hypothesis that more
tightly knit religions show a lower suicide rate. To derive what
he wished to explain, Durkheim assumed that Catholics are
more integrated than Protestants. He knew that this assumption
was off the mark for England, since the Anglican Church
Problem Selection in the Social Sciences: Methodology 53
was a state church. Yet in this way, he could account for
England as the deviant case to the regularity he wished to
explain. Malewski (1964) showed with examples from social
psychology that every explanation of a regularity predicts
conditions under which exceptions occur to the regularity to
be explained.
Duncan held that the question of factors influencing the
chances that the members of a society move up or down poses
the issue poorly. The prime factor is father’s occupation.
However, the proposition that the lower the occupational
prestige of a person’s father, the higher this person’s chances of
upward mobility forms a truism. After all, the hypothesis
thrives on bottom and ceiling effects. To test more interesting
hypotheses, according to Duncan, questions on change in
a person’s life should not be stated in terms of some difference
score, but as questions on the strength of the relation between
a person’s earlier and later score, and on the factors influencing
that relation.
Undertheorized Questions
Schütz (1932) raised the question of why human beings
understand one another, and did so with a certain urgency by
phrasing his question as ‘how is it possible that people
understand each other?’ This query sounds like a contradiction,
although casual observation suggests that some people at some
places and in some times indeed understand one another.
However, no particular hypothesis has been derived from
a specific theory holding that it is impossible for human
beings to understand each other. Thus, the theory rendering
the question compelling is lacking. The Parsonian problem of
how order is possible (Parsons, 1937) is undertheorized in
a similar way. Any derivation of the hypothesis that society is
violent, from the principle that persons follow their interests,
involves additional assumptions. These can be wrong, which
makes the question of how order is possible less pressing than
it sounds. Indeed, a full statement of the question ‘how is
order possible’ implies the conditions under which order
occurs, and those under which it does not.
Research of the past decades not infrequently addressed the
question of whether the association observed in a crosssectional
sample of a society’s inhabitants between their year
of birth and a particular characteristic is to be accounted for as
an age, cohort, or period effect. As methodologists point out,
this question is undertheorized too. The available hypotheses
do not predict some age effect, some cohort effect, and some
period effect. Those hypotheses are so vague that their
conjunction is almost if not fully logically unfalsifiable. The
hypotheses at hand have more substance; if people marry, grow
up in lean times, and live through a war, then their values will
be less postmaterialist (Inglehart, 1977). Those variables can be
independently ascertained, making statistical models
identifiable.
Conclusion
What are the prospects for more attention to problem selection
in scholarly studies on human societies and in social methodology
books? Since 1998, there is an English language
textbook aiming to state (and answer) sociology’s questions:
Sociology, The Central Questions by Kornblum. One of its questions
runs as “Stratification and social class, are they inevitable,
or can there be a classless society?” This question may be taken
as erroneous, yet the text following it makes clear that students
are taught to think in gradual instead of categorical terms about
inequalities. Nevertheless, this type of textbook does not seem
to have caught on. As to review articles, some now distinguish
generations of research. They do so not only by pointing
toward more advanced techniques for data analysis and new
theories, but also by invoking problem shifts. Ultee (2006) did
so in research on the theme of welfare states and inequalities,
and Knutsen (2007) in research on class voting. But this type
of review articles has not become common yet.
Lieberson (1985) subtitled one of his social methodology
books ‘the improvement of social research and theory,’ but
pointed toward the prevalence of ‘unanswerable questions,’
and he pleaded for social researchers to avoid them. Abbott
(2004) sought to complement rules for testing hypotheses by
stating rules for thinking them up. For this, he took his
inspiration from Pólya’s (1957) How to Solve It, with it being
a problem. But in the final chapter, Abbott, fortunately, was
concerned with ‘how to pose it,’ that is, how to arrive at
interesting problems, pointing toward puzzles. Firebaugh
(2008) presented seven rules for good social research. At first
sight, none of them refers to problems. But upon studying
the argument, it turns out that Firebaugh repeatedly points
toward erroneous questions. His prime example is the
question of why income disparities on a worldwide scale
have been growing for the past few decades, whereas they
have been declining (although they have been rising in most
countries of the world taken on their own). Upon closer
inspection, this example also illustrates the rule that
a question well put is a question half answered (the rise in
average income has been strongest in the poorest and most
populous countries).
In everyday life, researchers consider some problems
difficult, others significant, and yet others exciting. These
notions are difficult to put into formal rules. However, a more
difficult problem might be taken as an explanatory question
to which several solutions have been proposed, none of which
worked or led to closer approximations. Thus, difficult problems
are specific kinds of Popperian contradictions. Also,
a structure consisting of overarching problems and subquestions
goes some way toward formalizing the intuition
that a problem is significant. Perhaps a problem area is called
exciting, if, after many a failed attempt, the solution of an
overarching problem seems near, because progress is being
made on subproblems.
Of course, it may be argued, with Ziman (1987), that the
question of how scientists choose their problems is
erroneous, since most researchers do not have a choice. Their
problems are pregiven by the organization that employs
them. This observation does not do away with the question
of problem choice, but makes the question of the formal
rules for problem choice by the higher authorities of these
organizations more pressing. The methods outlined above
should be taken not only as principles for researchers who
have a choice, but also as precepts for institutes run on
a budget, and for foundations that award subsidies.
54 Problem Selection in the Social Sciences: Methodology
See also: Explanation: Conceptions in the Social Sciences;
Generalization: Conceptions in the Social Sciences; Grounded
Theory: Methodology and Theory Construction; Historical
Explanation, Theories of: Philosophical Aspects; Kuhn, Thomas
S. (1922–96); Logical Positivism and Logical Empiricism;
Merton, Robert K (1910–2003); Objectivity: Philosophical
Aspects; Popper, Karl Raimund (1902–94); Social Properties
(Facts and Entities): Philosophical Aspects; Weber, Max
(1864–1920); Weberian Social Thought, History of.
Bibliography
Abbott, A., 2004. Methods of Discovery, Heuristics for the Social Sciences. Norton,
New York.
Agassi, J., 1975. Science in Flux. Reidel, Dordrecht.
Blau, P.M., 1974. On the Nature of Organizations. Wiley, New York.
Blau, P.M., Duncan, O.D., 1967. The American Occupational Structure. Wiley, New York.
Boudon, R., 1979. La Logique du Social. Hachette, Paris (1981. The Logic of Social
Action. Routledge, London).
Breen, R. (Ed.), 2004. Social Mobility in Europe. Oxford University Press, Oxford.
Bunge, M., 1967. Scientific Research I. The Search for System. Springer, Berlin.
Collins, R., 1994. Four Sociological Traditions. Oxford University Press, Oxford.
De Swaan, A., 1988. In Care of the State. Polity, Cambridge.
Durkheim, É., 1893. De la Division du Travail Social. Alcan, Paris (1933. The Division
of Labour in Society. MacMillan, New York).
Durkheim, É., 1897. Le Suicide, Étude de Sociologie. Alcan, Paris (1951. Suicide.
A Study in Sociology. Free Press, Glencoe).
Durkheim, É., 1912. Les Formes Élémentaires de la Vie Religieuse. Alcan, Paris (1995.
The Elementary Forms of Religious Life. Free Press, New York).
Elias, N., 1939. Über den Prozess der Civilization. Franke, Bern (1994. The Civilizing
Process. Basil Blackwell, Oxford).
Erikson, R., Goldthorpe, J.H., 1992. The Constant Flux, a Study of Class Mobility in
Industrial Societies. Clarendon Press, Oxford.
Firebaugh, G., 2008. Seven Rules for Social Research. Princeton University Press,
Princeton.
Fogel, R.W., Engerman, S.L., 1974. Time on the Cross. The Economics of American
Negro Slavery. Wildwood, London.
Goldthorpe, J.H., 1987. Social Mobility and Class Structure in Modern Britain, second
ed. Clarendon, Oxford.
Gombrich, E.H., 1960. Art and Illusion: A Study in the Psychology of Pictorial
Representation. Phaidon, London.
Goode, W.J., 1963. World Revolution and Family Patterns. Free Press, New York.
Goode, W.J., 1993. World Changes in Divorce Patterns. Yale UP, New Haven, CT.
Gove, W.R., 1995. Is sociology the integrative discipline in the study of human
behavior? Social Forces 72, 1197–1206.
Heath, A., 1981. Social Mobility. Fontana, Glasgow.
Hobbes, T., 1651. Leviathan (1968 Penguin, Armondsworth).
Huizinga, J., 1941. Nederland’s Beschaving in de 17de Eeuw. Tjeenk Willink,
Groningen (1969. Dutch Civilization in the 17th Century. Harper & Row, New York).
Inglehart, R., 1977. The Silent Revolution. Princeton UP, Princeton.
Knutsen, O., 2007. The decline of social class. In: Dalton, J.R., Klingemann, H. (Eds.), The
Oxford Handbook of Political Behavior. Oxford University Press, Oxford, pp. 457–480.
Kornblum, W., 1998. Sociology, the Central Questions. Harcourt Brace, Fort Worth.
Kuhn, T., 1962. The Structure of Scientific Revolutions. University of Chicago Press,
Chicago.
Lieberson, S., 1985. Making it Count, the Improvement of Social Research and Theory.
University of Californai Press, Berkeley.
Lenski, G.E., 1966. Power and Privilege. McGraw-Hill, New York.
Malewski, A., 1964. Generality levels of theories concerning human behavior. In:
Zetterberg, H.L., Lorenz, G. (Eds.), A Symposium on Theoretical Sociology. Bedminster,
Totowa.
Merton, R.K., 1938. Science, Technology and Society in Seventeenth-Century England.
Saint Catherine Press, Bruges.
Merton, R.K., 1959. Notes on problem finding in sociology. In: Merton, R.K., Bloom, L.,
Cottrell Jr, L.S. (Eds.), Sociology Today: Problems and Prospects. Basic Books,
New York.
Merton, R.K., 1973. The Sociology of Science. University of Chicago Press, Chicago.
Olson, M., 1965. The Logic of Collective Action. Harvard UP, Cambridge, MA.
O’Neill, J. (Ed.), 1973. Modes of Individualism and Collectivism. Heinemann, London.
Parsons, T., 1937. The Structure of Action. McGraw-Hill, New York.
Pólya, G., 1957. How to Solve It. Double Day, Garden City.
Popper, K.R., 1962. Die logik der sozialwissenschaften. Kölner Zeitschrift für Soziologie
und Sozialpsychologie 14, 233–248 (1976. The logic of the social sciences. In:
Adorno, T.W., Albert, H., Dahrendorf, R., Habermas, J., Pilot, H., Popper, K.R.
(Ed.), The Positivist Dispute in German Sociology. Harper & Row, New York).
Samuelson, P., 1954. The pure theory of public expenditure. Review of Economics and
Statistics 36, 387–389.
Schütz, A., 1932. sinnhafte Aufbau der sozialen Welt. Springer, Wien.
Smith, A., 1776. The Wealth of Nations (1970 Penguin, Harmondsworth).
Trevor-Roper, H., 1976. Princes and Artists. Thames and Hudson, London.
Ultee, W., 1980. Fortschritt und Stagnation in der Soziologie. Luchterhand, Darmstadt.
Ultee, W., 1996. Do rational choice approaches have problems? European Sociological
Review 12, 167–179.
Ultee, W., 1998. Bringing individuals back into sociology. In: Blossfeld, H.P., Prein, G.
(Eds.), Rational Choice Theory and Large-Scale Data Analysis. Westview, Boulder.
Ultee, W., 2006. Problem shifts in the study of welfare states and societal inequalities.
Research in Social Stratification and Mobility 24, 377–386.
Van den Doel, H., 1979. Democracy and Welfare Economics. Cambridge UP,
Cambridge.
Weber, M., 1920. Gesammelte Aufsätze zur Religionssoziologie. Mohr, Tübingen
(1930. The Protestant Ethic and the Spirit of Capitalism. Allen & Unwin, London).
Winckelmann, J. (Ed.), 1978. Maw Weber. Die Protestantische Ethik II. Kritiken und
Antikritiken. Gütersloh, Mohn.
Ziman, J.M., 1987. The problem of “problem choice”. Minerva 25, 92–106.
Zuckerman, H., 1978. Theory choice and problem choice in science. In: Gaston, J.
(Ed.), Sociology of Science. Jossey-Bass, San-Francisco.
Problem Selection in the Social Sciences: Methodology 55