Research Methods

Research Methods

1. Primary

   • observations
     • unstructured
     • structured
     • participant
       • where the researcher has a role within the group being studied and takes part in activities
     • non-participant
       • when the researcher has no involvement with the group being studied, and simply observes behaviours from the sidelines
     • overt observations
       • where the identity and purpose of the researcher is made clear to participants
     • covert observations
       • where the identity and purpose of the researcher is hidden, and participants are unaware that research is being conducted
   • interviews
     • structured
     • semi-structured
     • unstructured
     • group interviews
     • a method of gathering qualitative data, where a researcher asks questions face-to-face with a respondent
   • questionnaires
     • a list of open/closed questions that respondents answer without the help of the researcher
       • open questions
         • allows the respondent to answer in their own words, collects qualitative data
         • more valid
         • allows issues to be explored more deeply
         • difficult to quantify
         • less reliable
       • closed questions
         • allows the respondent to answer through pre-set options, collects quantitative data
         • more reliable
         • can easily be quantified and graphed
         • less valid
         • does not tell a researcher WHY a behaviour is occurring
     • postal questionnaires
       • a large sample can be reached at once
       • highly targeted
       • cheap (low cost of printing and distribution)
       • the respondent may have more time to think about their answers, so their answers may be more honest and valid
     • self-completion questionnaires
     • face-to-face questionnaires
     • internet/email questionnaires
     • telephone questionnaires
   • experiments
     • sociological studies carried our under test conditions in order to remove the influence of confounding variables
     • laboratory experiments
       • an experimental group is exposed to the independent variable, and a control group is not changed in any way, while all other variables are controlled
     • field experiments
       • where an experiment is carried out in a natural setting

2. Secondary

   • official statistics
     • statistics collected and published by agencies, eg. NGOs or governments
   • official documents
     • public documents, eg. media, school records, parish registers
     • personal documents, eg. letters, photos, autobiographies
     • historical documents
   • content analysis
     • analysing social life by investigating and interpreting the content of mass media

3. Qualitative

   • gives an idea of how participants view the world
   • more "real-life"
   • validity
   • preferred by Interpretivist sociologists
     • we construct society through the meanings we attach to things
     • bottom-up (micro) approach
     • subjectivity

4. Quantitative

   • information presented in a numerical form
   • trends/patterns can be analysed
   • preferred by Positivist sociologists
     • interested in social reality
     • objectivity
     • top-down (macro) approach
     • reliability and representativeness

5. PET

   • Practical
     • time
     • cost
     • access
     • research opportunity
     • utility in relation to the research issue
     • funding bodies
     • researcher's safety
     • researcher's personal skills/characteristics
   • Ethical
     • harm to participants
     • informed consent
     • confidentiality
     • privacy/anonymity
   • Theoretical
     • methodological preference
     • reliability
     • validity
     • representativeness
     • researcher bias
     • Hawthorne effect

6. Sampling techniques

   • random sampling
     • every member of the target population has an equal chance of being picked
     • equal opportunities
     • quick, can even be performed by a computer
     • reduces the potential for human bias
     • allows researchers to make generalisations within the target population
     • can only be used if the population is whole
     • may not be representative
   • quasi-random sampling
     • every 10th, 100th, etc. person on a list is selected
     • quick, can even be performed by a computer
     • reduces the potential for human bias
     • allows researchers to make generalisations within the target population
     • equal opportunities
     • it can be difficult to gain access to a list
     • the list must equally represent every member of the sampling frame
     • it can be difficult to contact chosen participants
   • stratified random sampling
     • dividing the sampling frame into smaller strata (eg. age, gender) and randomly selecting a number of people from each strata
     • very representative
     • allows large groups to be studied effectively
     • fewer sampling errors
     • greater precision than random sampling
     • smaller samples can be used
     • may be more time-consuming to organise
   • snowball sampling
     • sociologists ask a few volunteers to find other volunteers
     • makes it easier for researchers to get volunteers from groups that they may not have access to
     • very little effort is required on the part of the researcher
     • volunteers may only choose their friends
     • peer pressure may force some people to volunteer even if they are not comfortable with the research
     • not representative
   • volunteer sampling
     • sociologists ask for volunteer participants within their sampling frame
     • informed consent is easily obtained
     • only people from certain social groups may be willing to volunteer
   • opportunity sampling
     • selecting individuals nearby
     • quick, easy
     • individuals are not forced to participate
     • researcher bias, as they choose who to approach
     • may not be representative