Methods in Sociology

Methods in Sociology

1. Positivists

   • Experiments
     • Conducted in a controlled environment. In a laboratory experiment, the researcher is interested in the relationship between an independent variable (a possible cause) and a dependent variable (a possible effect). Involves setting up two identical groups - an experimental group and a control group - which are treated differently. The experimental group is and a control group which are treated differently. The experimental group are is exposed to the independent variable and its behaviour is often compared to the control group (which is not exposed) to monitor any differences between the two. Any differences are seen as dependant variables or effects.
       • Reliability
         • Positivists regard the laboratory experiment as reliable for three reasons: 1.The original researcher can control the conditions so others can easily repeat the steps to re-run it. 2.It produces quantative data, so the results of the re-run experiment can be compared to the original easily. 3. It is a detached and objective method - the researcher merely manipulates the variables and records the results
       • Evaluation
         • There are practical, ethical and theoretical reasons why sociologists are not keen on using laboratory experiments. Humans are complex beings, so it is impossible to construct identical and control groups. No two humans are exactly alike, which means we experience and interpret all social situations in different ways. Laboratory experiments can only focus on small samples and so are not that useful in studying large-scale social phenomena. A laboratory experiments is an artificial environment and any behaviour that occurs in it may be a product of the environment. The subjects may act differently; they may feel self important, anxious or resentful, or they may work out the way the researcger wants them to. This is called the Hawthorne effect, named after experiments conducted by Mayo in the 1920s at the Hawthorne factory in the USA where it was first observed.
         • Experiments may result in harming subjects. For example, exposing children to violent films for long periods may do psychological or emotional damage. Interpretivists argue that humans are fundamentally different from other natural phenomena. We have free will and choice. Our behaviour isn't 'caused' by external forces, so it cannot be explained in terms of cause and effect relationships
   • Social survey
     • A social survey involves collecting the same type of data from many people, usually via questionnaires or, less often, structured interviews. Some surveys consist of longitudinal research, which involves studying the same group of people over a long period of time. These provide a clear image of changes in attitudes and behaviour over time, but can be problematic, as respondents may drop out or researchers may lose track of them. The views of those that remain in the sample may also be significantly different from the views of those who drop out, so over time the sample may become increasingly unrepresentative.
       • Questionnaires
         • A list of questions that are written down in advance. They are the main method for gathering data in social surveys. They are handed out or posted to the respondent - the person chosen by the researcher to answer the questions. This type of questionnaire is known as a self-completion questionnaire. Other questionnaires become interview schedules, in that they are read out and filled in on behalf of the respondent by trained interviewers. This type of questionnaire is known as a formal or structured interview.
           • When constructing a questionnaire, the sociologist must ensure that the right questions are asked to unearth the exact information that is needed. So the questions must focus on the hypothesis into s series of questions is called 'operationalization'. Ideally, questions should be objective and it is essential that they contain neutral wording. However, questions can be biased, in that they can 'lead' respondents to the answers the researcher requires. They can sometimes be 'loaded', or written in such a way that the respondent is provoked into an emotional response that seeks to evade the truth. Questionnaires tend to use 'closed' questions with a choice of pre-set answers with accompanying tick boxes, which produce quantitative data.
         • Strengths
           • If postal, can be used to reach large numbers of people around the country, which may improve the representativeness of the sample. They are less time-consuming for respondents than interviews. Reasonably cheap compared to other method. Useful for research that includes sensitive or embarrassing questions, as these can be answered in the privacy of the home rather than a face-to-face, which might undermine validity. Ensure that the sociologists has minimum contact with the respondent, so reducing the possibility of the respondent feeling suspicious or threatened
         • Weaknesses
           • Especially if postal, suffer from low response or even non-response. Those returned may not be representative of the research population, as the replies may be from people with strong unrepresentative views. May not be suitable for finding out why people behave the way that they do, as real life is often too complex to categorise in closed questions and responses. Respondents may interpret questions in a different way from that intended by the researcher. Artificial devices that are not a normal part of daily reality, so people may respond with suspicious and may not tell the truth, or may be partial, as they feel threatened by the researchers. By choosing the questions and responses, the researcher has already mapped out the experiences and interpretations of respondents (e.g. may be forced to tick boxes that only approximate to their experiences, views and opinions, thus undermining the validity of the data, which may frustrate respondents and result in non-response)
     • Self-report questionnaires
       • This lists various petty criminal acts and asks respondents to tick those that they have committed without being caught. Sociologists attempt to improve validity by stressing confidentiality and anonymity for the respondents.
         • Marsh: Validity is undermined by under-reporting and over-reporting. People may under-report because self-report studies are retrospective and depend on being able to remember crimes committed 12 months before. Some people exaggerate so that they create a 'tough' impression. Others keep quiet as they fear that they may be reported to the police. It is impossible to include all criminal acts in a questionnaire, this means it is difficult to suggest which crimes to include. Self-reports are usually distributed to young people - it would be difficult to get businessmen to cooperate and admit to various types of white-collar and blue-collar crime.
   • Structured Interviews
     • Viewed as a scientific method by positivists. These interviews usually involve the researcher