The research process

Kenneth D. Bailey

K.Bailey in his work focuses on the notions of social research defining it as being concerned
with gathering data that can be used for answering questions about society and helps us
understand it better. Alternatively, social research may also be used to provide answers to
questions of theoretical interests to a particular social science discipline which may not have
any apparent application in the current scenario.
The conventional usage of social science data collection methods, like surveys, made the
implicit assumption that the study method was a means to an end. Research methods have an
impact upon society through their findings and the methods themselves are increasingly used
by sectors of the society outside of the social sciences. An example is the increasing use of
surveys by newspapers, television networks and political parties and their candidates for office
which has led to several irregularities in election procedures.
The lesson for the 1990s and 21st century is that research methods are not an indisputable part
of mass media operations and thus of society. This is true not just in United States but also in
other countries such as Israel where the desire to forecast elections has been labelled an
obsession.
One can conclude that social research methods not only impact social scientists but other
members of the larger society. Methods of social scientists such as surveys when used by
media may come dangerously close to affecting important processes such as elections and thus
become a form of social engineering rather than purely a method of research. Thus, it seems
more important for the social scientist to understand social research methods both as one who
conducts research and as one who is affected by such methods.

Social Science as Science

A crucial question in social science is what constitutes the proper understanding of society and,
how this understanding can be achieved. Krathwohl deals with those issues and emphasises on
the ways in which one’s orientation to knowledge affects the research methods used and the
different criteria utilised by researchers with different orientations.
The social science practitioners have sought the proper position of their discipline with respect
to physical science and to humanities. The essential problem concerns the nature of social
phenomena and how they can be understood. Wilhelm Dilthey believed that humans had free
will, and thus no one can predict their actions and generalise about them.
Durkheim opposed this view stating that social phenomena are orderly and can be generalised.
In his view there was little difference between physical and natural science and social science
except for subject matters. The logic of inquiry was essentially the same. This scientific view is
labelled positivism. According to Durkheim, sociologists could use the methods of natural
science, such as experimentation to study and explain social phenomena.
Weber on the other hand, took an intermediate position, as per which social phenomena were
not merely determined by social laws but were product of human volitional action. According to
Weber, the use of the methods of natural science plays a role in social research, but not an
exclusive role. According to Weber, the use of the scientific methods followed in the physical
and natural science is legitimate but inadequate for the study of all social phenomena. Direct
understanding- Verstehen, Weber discusses is not possible in physical science but is possible in
social science because of the different relationship between the researcher and his or her data.
There is disagreement about how social phenomena should be studied and even about which
phenomena to study. As discussed by Bierstedt, the scientific view (positivism) embodies the
notion that sociology is seen as a pure rather than an applied science; its task is thus to gather
knowledge rather than to use it. Moreover, Sociology is seen as an abstract science rather than
a concrete one, and thus should be more concerned with generalising rather than with
particularising. Ethnomethodology is the author's most cutting-edge theory.
Ethnomethodologists rarely value broad social laws or concepts. They stress social phenomena'
situational meaning.
Stages of Social Research
All research projects share a common goal of furthering our understanding of society and thus
all share certain basic stages. These stages are as follows:
1. Choosing the research problem and stating the hypothesis
2. Formulating the research design
3. Gathering the data
4. Coding and analysing the data
5. Interpreting the results so as to test the hypothesis
Each of these stages is dependent upon the others. The researcher needs to have an adequate
knowledge of the later stages before he or she can perform the earlier prerequisites. Research
then, is a system of interdependent related stages.
Circularity -The research process is best conceived as a circle. One enters a particular project at
Stage 1. However, the researcher is able to draw on past studies in formulating his or her
hypothesis. After the research completes stage 5, he/she may stop but the research process
itself is not completed at this stage. If the study was unsuccessful or only partially successful,
the research must return to the early stages of investigation.
Replication- Even if the research is successful and the findings of stage 5 confirm the
hypothesis of stage 1, it is advisable to repeat the study so as to demonstrate that the findings
are not an accident or a coincidence. If the study is repeated exactly, especially with a different
sample, a second confirmation of the findings will lend further support to the contention that the
hypothesis cannot be rejected. This exact repetition of a study is called replication.
Density Research- While all social research projects share the five basic stages, they exhibit
much diversity in the way in which these stages are carried out. Research projects in social
science range from highly controlled experimental laboratory studies to uncontrolled
observational studies.
The author provides examples of research that one can use to illustrate the five stages of
research and the range of permissible activities by comparing two studies that address the
same broad subject—the effects of population density on individuals. Researchers have
speculated about the possible negative effects of population density theorising that high density
levels cause such phenomena as crime and mental illness and anxiety or alienation.
The father of population theory, Thomas Malthus who theorised that overpopulation was a major
cause of poverty that if unchecked would lead to mass starvation. Calhoun tested Malthus’s
hypothesis that overpopulation causes social pathologies. However his subjects were rats rather
than humans. In studies on both wild and tame rats, Calhoun found that a certain density that he
determined was about twice the comfortable level caused such negative effects as high infant
mortality rates, hyperactivity, cannibalism etc. Finding that the population density has negative
effects on rats does not mean that this relationship holds to the same degree or even at all, for
humans.
The problem the author identifies is that how can we design a study adequate to test the
hypothesis that overcrowding causes harmful effects on humans without actually harming the
population of study? Quite different approaches are evident in two of the best-known recent
studies on the effects of density on human populations. One of these studies is "Population
Density and Pathology: What Are the Relations for Man?" by Galle, Gove, and McPherson
(1972). The second study, by Griffit and Veitch (1971), is entitled "Hot and Crowded: Influences
of Population Density and Temperature on Interpersonal Affective Behavior." We can compare
these studies in terms of our five stages of research.
Stage 1: Choosing the Problem and Stating the Hypothesis - Choosing the problem was
relatively easy in both studies, for both were based on past research. Both research teams were
aware of the evidence showing negative effects of density that had been found by Calhoun. In
addition, the Griffit team was aware of findings by US Riot Commission that most riots occurred
under crowded conditions and on hot days. Thus, the general research problem and the
hypothesis were about the same for both studies.
Stage 2: Research Design- In this stage the researcher must decide how to measure the two
main variables in his or her hypothesis (density and negative social effects) and on what group
of people to test the hypothesis. This involves deciding not only how many people will be used
as subjects but also what their particular characteristics should be and under what
circumstances the data will be gathered.
One of the major problems in social research is control of extraneous factors that might interfere
with the study. Griffit and Veitch decided to conduct a laboratory study involving a decision to
take people out of their natural everyday environment and put them in an artificial setting - a
Sherer-Gillet environmental chamber, seven feet wide, nine feet long and high. The researcher
can easily vary density from high to low simply by controlling the number of persons admitted to
the chamber at one time. Griffit and Veitch defined low density as three to five persons in the
chamber and high density as twelve to sixteen persons in the chamber. While it facilitates the
measurement of density, the use of such chamber makes the measurement of harmful effects
more difficult. For example, Calhoun found extreme aggressive behaviour such as cannibalism
among rats. It would be unethical for Griffit and Veitch to attempt to create such effects among
humans.
The Galle team also studied the effects of density of humans. However, their research design
for testing their hypothesis was different. Rather than being on the scene, Galle team simply
studied the things that had already happened which allowed a larger sample size but this led to
a lack of flexibility in design and limitation in data. Furthermore, the data was old and the
researchers didn’t have direct control over the research either. They did not have data recorded
for each individual but only summary data.
Stage 3: Gathering the Data- The persons in the sample in the Griffit study were randomly
assigned to one of eight experimental conditions formed by varying density and heat. All
subjects were dressed alike and were told that the experiment was intended to investigate
judgemental processes under altered environmental conditions. While in the chamber, each
group was given a number of pencil and paper tests in addition to the Interpersonal Judgement
Scale. The subjects were asked to rate their attraction to a hypothetical stranger. In Galle’s
study data had already been collected from the Fact Book
Stage 4: Coding and Analysing the Data- Griffit and Veitch used answers to two items on the
Interpersonal Judgement Scale to obtain a measure of attraction toward the stranger. This
measure yielded scores ranging from 2 to 14 (high level of attraction). Important findings from
the study included – subjects showed more attraction towards strangers similar to themselves,
irrespective of density. Most important finding was subjects in higher density showed lower
attraction level and the opposite held true for low density. Therefore, their hypothesis was
supported by the data that high density leads to people disliking on another
Data analysis were more complicated in the Galle study, not only because more variables were
involved but also because there were many confounding factors that might affect the
relationship between density and pathology thus making it difficult to ascertain whether density
really caused pathology. While rat societies do not exhibit much social structure, human
societies do which can affect the relationship between density and pathology. Thus, if Galle
found differences among the various community areas in rates of juvenile delinquency and
admission to mental hospitals, it is not clear that these differences are due to different levels of
density. They might be due to differences in social class or ethnicity. Without including social
class and ethnicity in the analysis, the Galle study found an apparent relationship between
density measured as number of persons per acre and the pathology measures. But when social
class and ethnicity were included in the analysis, this relationship disappeared. Thus, the
researchers concluded that density does not cause pathology but that differences in both
density and pathologies are caused by differences in social class.
Stage 5: Interpreting the Results and Testing the Hypothesis- Griffit and Veitch found evidence
to support their hypothesis that increasing level of population density increases the level of
aggression. The next step is to replicate the study with a larger varied sample to make sure
finding isn’t a fluke. Evidence in the Galle study seems to indicate that density does not cause
pathology but that both density and pathology level vary with social class. If the researchers still
feel that density cause pathology, they need to revise the study. They left the hypotheses
unchanged but changed the measure of density at stage 2 and redid the study.
Ultimately, discussion of these two studies has shown how interesting and challenging social
research can be and what wide difference exists in social research. Although each research
project is unique in some ways, all projects involved the same five basic stages. Each of these
stages is dependent upon the others and the total research process is quite circular. If the
hypothesis is rejected, the researcher must revise the study and begin again. Even if the
hypothesis is not rejected, the researcher wishes to replicate to make sure the findings are
accurately.