Please enable JavaScript.
Coggle requires JavaScript to display documents.
Choosing a research method - Coggle Diagram
Choosing a research method
Primary and secondary sources of data
Primary data:
Info collected by sociologists themselves for their own purposes (obtain first-hand 'picture' of a group or society, test a hypothesis)
Methods:
Social surveys asking people questions in a written questionnaire or an interview
Participant observation: sociologist joins in with activities of the group being studied
Experiments: rarely use lab experiments, sometimes use field and comparative method
Advantage - gather precisely info they need to test hypothesis, can often be costly and time consuming
Secondary data:
Info collected or created by someone else for their own purposes, but sociologists can then use
Methods:
Official statistics: produced by gov on range of issues - education, crime, divorce, unemployment. Other stats produced by charities, businesses, churches, other organisations
Documents: letters, emails, diaries, photos, official reports, novels, newspapers, internet, television broadcasts
Can be quick and cheap, who produced it may not be interested in same questions so sources may not provide info needed
Quantitative and qualitative data
Quantitative:
Info in a numerical form - official stats, percentages
Info collected by opinion polls and market research surveys
Qualitative:
Gives a 'feel' for what something is like
Evidence gathered using participant observation aims to give sense of feels like to be member of particular group
In-depth interviews probe deeply into person's view give insight into what it's like to be in their 'shoes;
Provide rich descriptions of people's feelings and experiences
Factors influencing choice of methods - practical, ethical, theoretical
Practical issues
Time and money:
Different methods require different amounts of time and money which may influence choice of method
Requirements of funding body:
Research institutes, businesses, organisations that provide research funding may require results in a particular form - use method capable of producing such data
Personal skills and characteristics:
May affect ability to use methods
Subject matter:
May be harder to study particular group or subject by one method
Research opportunity:
Sometimes opportunity is unexpected so structured methods that take longer to prepare may not be possible
Glasgow gang leader offered Patrick (1973) chance 'out of the blue' to spend time with his gang - little time to prepare = participant observation
Others may have more time to set up research opportunity carefully and have time to select method
Ethical issues
Informed consent:
Ps should be offered right to refuse to be involved (right to withdraw)
Should be told about relevant aspects of the research to make an informed decision - may need debrief at the end
Consent should be obtained before research begins
Confidentiality and privacy:
Keep identity of Ps secret to prevent possible negative effects
Respect their privacy
Personal info should be kept confidential
Harm to research Ps:
Researchers should be aware of possible effects of their work on those studied
Could include: police intervention, harm to employment prospects, social exclusion, psychological damage
Physical, psychological harm
Should anticipate and prevent harm when possible
Vulnerable groups:
Special care taken - age, disability, physical/mental health
Covert research:
Researcher's identity and purpose are hidden from those studied - ethical problems of deceiving, lying to gain trust or obtain info, cannot gain informed consent
May be justified to circumstances like gaining access to secretive dangerous or powerful groups
Theoretical issues
Validity:
Valid method produces true or genuine picture of what something is really like - get closer to the truth
Many argue qualitative methods give more valid/truthful account - insight into first hand experience
Reliability:
Similar to replicability - reliable if another researcher repeats the same method and get the same results
Representativeness:
Whether the people studied are a typical cross-section of the group we're interested in
If sample is representative or typical of the wider population then findings can make generalisations without having to study all
Large scale quantitative surveys using sophisticated sampling techniques more likely to produce representative data
Methodological perspective:
Positivists: prefer quantitative data, seek to discover patterns of behaviour, see sociology as a science - functionalists and marxists: see society as a large-scale (macro-level) structure that shapes out behaviour
Interpretivists: prefer qualitative data, seek to understand social actors' meanings, reject sociology can model itself on natural sciences - interactionists: micro-level view, focus on small-scale, face-to-face interactions
Conclusion:
Theoretical perspective usually main factor when choosing method - obtain this type of data perspective views most appropriate
Practical and ethical factors limit choice - time, resources, access, consent, privacy all constraints on choice
Sheer chance may determine method - Tuckett (2001) - one postgraduate sociology student found himself taken ill with tuberculosis and confined to hospital, used opportunity to conduct P observation study
Factors affecting choice of topic
Perspective:
Theoretical perspective influence choice - NR study welfare benefits effects, Feminists study DV
Society's values:
As values change so does research - rise feminism led to focus on gender inequality
Practical factors:
May restrict topic choice - inaccessibility to certain situations
Funding bodies:
Gov agencies, Economic and Social Research Council, charities, businesses - pay for research so determine topic investigated
The process of research
Formulating an aim/hypothesis:
Most studies have general aim or specific hypothesis.
Hyp is a possible explanation that can be tested by collecting evidence to prove it true or false
Aim is more general - identifies what's intended to be studied and hope to achieve through research
More open-ended = not tied to trying to prove particular hypothesis - gather data on anything that appears interesting about the situation
Interpretivists favour broader aim as they're interested in understanding actors' meanings so don't impose possible explanations
Operationalising concepts:
Before testing need a working/operational definition of the key idea - define concepts in a way that allows them to collect data
Helps make research more clear and precise, allows researchers to compare findings with other studies, helps make sociological research more objective
Problems may arise when things are operationalised differently - hard to compare findings
Positivists concerned to operationalise because of importance place on creating and testing hypothesis
Interpretivists less emphasis on operationalising - more interested in actors' own definitions and understandings than imposing their own
Pilot study:
Draft version of the method on a small sample
Aim to iron out any problems, refine or clarify questions and wording, interviewer practice
Young and Willmott (1962) - over 100 pilot study interviews to decide design of study, questions and how to word them
Samples and sampling:
Smaller sub-group drawn from wider group we're interested in
Purpose to ensure people chosen to be in the study are representative/typical of the research population
Can then make generalisations to whole research population - positivists like as wish to make general, law-like statements about wider social structure
Sampling frame:
List of all members of the population interested in studying
Young and Willmott used electoral register
Should be as complete and accurate as possible, up to date without and duplications
Can choose sample from it to ensure it's representative
Sampling techniques:
Random: selected by chance - equal chance of selection, large enough random sample should reflect characteristics of whole pop
Quasi-random/systematic: every nth person in frame is selected
Young and Willmott used every 36th name
Stratified random: break down pop by age, class gender etc, sample taken of same proportions
Quota: pop stratified then each interviewer given quota needed to fill with respondents who fit characteristics
Non-representative sampling:
Practical and theoretical reasons means not all studies use representative sampling techniques
Practical:
Social characteristics of research population may not be known - impossible to create sample of exact cross-section
May be impossible to find/create sampling frame for particular population
Potential respondents may refuse to participate
May use alternatives:
Snowball sampling: collect sample by contacting number key individuals whoa re asked to suggest others who being be researched - may not be representative but useful to contact those who may not otherwise be persuaded to take part
Opportunity sampling: choosing individuals who are easiest to access
Theoretical:
May not choose to create representative sample due to theoretical perspective
Interpretivists - more important to obtain valid data and authentic understanding of meanings then discover general laws or behaviour - less need for representative sample