OrientationObservationIn-depth interviewsDocument analysis and semiologyConversation and discourse analysisSecondary Data
SurveysExperimentsEthicsResearch outcomes
Conclusion
8.3 Doing survey research The following exploration will focus on conventional approaches to undertaking survey research, which usually quantifies data with an attempt to providing explanations of the social world. This approach, which involves measuring social phenomena and drawing some conclusions about relationships between them, is often called the 'quantitative approach'. However, this does not mean that the researcher needs to be a mathematical or statistical expert to use the approach, as the examples in this section of the Guide will show.
The quantitative approach usually requires significant amounts of data. Given the complexity of social phenomena and the number of factors involved, large data sets are required in order to identify the likely causal connections.
Furthermore, if the research aims to provide generalised results, then the data has to be representative of the population.
If you are measuring social phenomena then you need a measuring instrument that is reliable (see Section 1.9) and valid (see Section 1.8).
Sociological research using the survey method involves several stages.
1. Identify the research area and formulate the aims and purpose of the research.
2. Explore existing literature and sociological theory relating to the proposed research.
3. Assess the feasibility of doing the research and amend scope if beyond the available resources and time.
4. Clarify the objectives of the research and specify the hypotheses to be tested.
5. Identify the main concepts and determine how these will be operationalised and measured as variables.
6. Consider whether a survey is the most appropriate tool and if so specify how data will be collected (interviews, questionnaire, face-to-face or remotely) and analysed (what relationships are being assessed based on the hypotheses being tested).
7. Construct a research instrument (interview schedule of quetionnaire).
8. Pilot the survey to assess the instrument and the potential analysis
9. Identify the target population and determine how a sample will be selected.
10. Contact sample members and collect the data via interviews or questionnaire.
11. Where necessary, code the data and then compile error free numerical databases.
12. Analyse the data, focusing on the main relationships initially and then checking that any relationships found are not the result of other factors.
13. Reject or do not reject the hypotheses and any supplementary hypotheses that have emerged as a result of the research process.
14. Write a report that outlines the research aims, methodology and findings along with conclusions about the probable relationships and their implications for sociological theory and social action.
This is, to some extent, an idealised view of the quantitative approach that is often difficult to keep to in practice. However, it does provide a model guide for quantitative research.
The following Sections explore this model outline and explains how to carry out each stage. Some are much more resource demanding than others but all are necessary, even if, in practice. some of these stages are addressed cursorily due to time and resource constraints.
The model shows that undertaking a sociological survey is not just about asking a few questions. It is part of an integrated approach to sociological enquiry that begins and ends with sociological theory.
It is important that researchers avoid doing surveys the other way round. Inexperienced social surveyors often start with a few 'interesting' questions and then struggle to fit the data they get to a meaningful sociological theory. In other words, do not start with the social survey, start with the substantive problem.
The survey is just the tool by which the data is collected, it is part of a larger methodology. At the core of this methodology is the idea that it is possible to answer questions about why social phenomena occur. This is done by defining and measuring social phenomena and showing relationships between them.
