Choosing a Research Method

Choosing a Research Method

1. Two key sources of data

   • Primary
     • This is data that the researcher has found themselves e.g. lab experiments
     • Can gather this data by questionnaires, interviews and experiments
     • This data allows hypotheses to be accurately tested
     • However it is time consuming and costly
   • Secondary
     • This is data that has been gained from elsewhere e.g. the internet
     • Can gather this data by documents and official statistics
     • This data is quick and cheap as someone else has already done the research
     • However those who produce the data may not be interested in the same questions as the sociologist so it may not provide the needed info.

2. Two broad types of data

   • Qualitative
     • Rich, descriptive data, such as unstructured interviews, open ended questions and medical documents
   • Quantitative
     • Numerical data, such as statistics, closed ended questions and structured interviews

3. Practical issues

   • Requirements of funding bodies, e.g. type of data
     • May require results to be in a certain form
   • Personal skills/characteristics e.g. methods
     • Some sociologists may better suited to some research methods than others e.g. shy -> use questionnaires
   • Subject matter, e.g. questionnaire when studying illiterate people
     • Who are we trying to study? e.g. It may be difficult for a male sociologist to study an all female group
   • Research opportunity may be out of the blue, can't structure it as you wish

4. Ethical issues

   • Informed Consent
     • Ps must be offered the right to refuse, researcher must explain all aspects of research so ps can give fully informed consent
   • Confidentiality and privacy
     • Researchers should keep ps identities secret to stop negative effects. Researcher should respect privacy. Personal info should be kept confidential

5. Theoretical issues

   • Validity
     • A valid method is one that produces a genuine picture of what something is really like, allows researcher to be close to the truth
   • Reliability
     • Where a study or test can be replicated many times by different sociologists and still get the same results
   • Representativeness
     • Enter text here