Sociology. Anthony Giddens
work from other types of inquiry, though these criteria are not fixed but change over time, alongside ongoing research programmes and studies. We can also agree with historians that science takes place within communities and broad theoretical frameworks or paradigms. Sociology has moved forward through competitive struggles between rival perspectives and, over time, the number of perspectives and theoretical syntheses has increased. Yet, in spite of this variety and competition, and against the anarchist position taken by Feyerabend, there remains a logic to the research process that is common across the majority of sociological studies, and this is outlined in the next section.
THINKING CRITICALLY
Determining which academic disciplines are ‘scientific’ has concerned many philosophers and historians, but does it matter? For instance, some sociologists say that ‘social studies’ better characterizes their work. List three consequences that may follow for the practice of ‘doing sociology’ within universities if sociology was not seen as a scientific discipline.
Carrying out research in sociology involves a number of steps, leading from identifying a research problem and devising a methodology right through to publishing the findings and responding to criticisms from colleagues (see figure 2.1). However, all research begins with the desire to know or better understand some aspect of the social world.
Defining the problem
All research starts from a problem or question. This is sometimes an area of factual ignorance: we may simply wish to improve our knowledge about certain institutions, social processes or cultures. A researcher might set out to answer questions such as ‘What proportion of the population today holds strong religious beliefs?’ or ‘How far does the economic position of women still lag behind that of men?’ Such questions are necessary and useful.
However, the best sociological research begins with problems that are also puzzles. A puzzle is not just a lack of information but a gap in our understanding. Much of the skill in producing worthwhile sociological research consists in correctly identifying puzzles. Rather than simply answering the question ‘What is going on here?’, puzzle-solving research tries to contribute to our understanding of why events happen as they do. Thus we might ask: ‘What accounts for the decline in the proportion of the population voting in elections in recent years?’ ‘Why are women poorly represented in high-status jobs?’ These questions are not simple factual questions but require us to go a stage further to provide explanations for the evidence we find.
It is important to remember that no piece of research stands alone. Research problems arise as part of ongoing work, and one research project may easily lead to another because it raises issues the researcher had not previously considered. A sociologist may also discover puzzles by reading the work of other researchers in books and professional journals or by being aware of trends in society.
Reviewing existing evidence
Once the problem is identified, the next step is usually to review the available evidence in a particular field through a review of the existing literature. It could be that previous research has already satisfactorily answered our question and there is no need to repeat the process. But, if not, the sociologist will need to sift through whatever research does exist to see how useful it is for their purpose. Have previous researchers spotted the same puzzle? How have they tried to solve it? What aspects of the problem have their studies left unanalysed? Drawing upon others’ ideas helps the sociologist to clarify the issues that might be raised and the methods that might be used in their own research. Reviewing the literature is an essential step that helps to avoid unnecessary duplication and repetition, and it can also point out where gaps in our knowledge still exist.
Making the problem precise
A third stage involves working out a clear formulation of the problem. If relevant literature already exists, the researcher might return from the library with a good idea of how the problem should be approached. Hunches about the nature of the problem can sometimes be turned into research questions, which, though rooted in an educated guess about what is going on, clearly state this in precise language. If the study is to be effective, research questions must be formulated in such a way that the empirical material gathered will provide evidence that either supports or challenges them. Studies involving the collection and analysis of numerical data, such as social surveys, tend to favour statistical testing as a method of verifying or falsifying clearly stated hypotheses, while qualitative research will often be exploratory in character and allow research questions to emerge during the research process.
Working out a design
The researcher must then decide just how the research materials are to be collected. A range of different research methods exists, and which one is chosen depends on the overall objective of the study, as well as on which aspects of behaviour are to be analysed. For some purposes, a social survey (in which questionnaires are normally used) might be suitable, especially where we need to gather a large quantity of data. In other circumstances, if we want to study small social groups in great detail, interviews or an observational study might be more appropriate. We shall learn more about these and other research methods later in this chapter.
Conducting the research
At the point of proceeding with the research, unforeseen practical difficulties can crop up, and very often do. For example, it might prove impossible to contact some of those to whom questionnaires are to be sent or those people the researcher wishes to interview. A business firm or school may be unwilling to let the researcher carry out the work they had planned due to concerns about sensitive information being leaked. Difficulties such as this could result in bias, as the researcher may be able to gain access only to a partial sample, which subsequently leads to a false overall result. For example, if the researcher is studying how business corporations have complied with equal opportunities programmes for disabled people, companies that have not complied may not want to be studied, but omitting them will result in a systematic bias in the study’s findings.
Figure 2.1 Steps in the research process
Bias can enter the research process in other ways too. For example, if a study is based on a survey of participants’ views, the researcher may, even unwittingly, push the discussion in a particular direction, asking leading questions that follow their own viewpoint (as the Doonesbury cartoon shows). Alternatively, interviewees may evade a question that they just do not want to answer. The use of questionnaires with fixed wording can help to reduce interview bias, but it will not eliminate it entirely. Another source of bias occurs when potential participants in a survey, such as a distributed voluntary questionnaire, decide that they do not want to take part. This is known as non-response bias, and, as a general rule, the higher the proportion of non-responses in the sample, the more likely it is that the survey of those who do take part will be skewed. Even if every attempt is made to reduce bias in surveys, the observations that sociologists make in carrying out a piece of research are likely to reflect their own cultural assumptions. This observer bias can be difficult and perhaps even impossible to eliminate, as sociologists – believe it or not – are human beings and members of societies as well as sociologists! Later in this chapter we look at some of the other pitfalls and difficulties of sociological research and discuss