Education: Research methods Flashcards
features of laboratory experiments
experimental groups: have something done to them
control group: placebo group
operationalised variables: everything has to be numerised
reliability: if it was repeated, would you get the same results?
issues of experiments
practical issues: conducting experiments about society isn’t realistic because you cannot replace society in a lab
ethical issues: consent, debrief, protect from harm, right to withdraw at any point
Hawthorne effect: when you know your being watched, people change their behaviour
what is a field experiment
take place in natural surroundings and usually, PPs are unaware they are being studied
the comparitive method:
carried out in the mind of the sociologist
step 1: identify 2 groups of people who are alike in all major aspects except one
step 2: compare the groups to see if the difference has any effect
types of interviews
structural/ formal interviews: interviewer asks predetermined questions
unstructured: less limited answer formats. more like a conversation
semi-structured: has a question frame but interviewer can ask to elaborate
group interviews: multiple people are interviewed together. the researcher asks a group to discuss certain topics.
interviews evaluation
structured:
practical issues: training interviewers is inexpensive
response rates: structured interviews generally have increased chances of a representative population
unstructured:
rapport and sensitivity: informality allows the interviewer to gain a rapport, so PP is likely to be honest
interviewee’s view: allows interviewees to speak about what they see as important.
social interaction:
interviewer bias: the interviewer may ask leading questions or may become too close or attached to PPs.
artificiality: even relaxed unstructured are still artificial situations.
status and power: inequality for interviewer and interviewee can affect how honest and open PPs are. gender differences in power can also shape an interview.
cultural differences: there may be a misunderstanding as a result different meanings will be given to the same words.
social desirability: people often seek approval, but also people may seem ignorant if they do not want to answer a question.
sampling
opportunistic: taking the sample from people who are available at the time the study is carried out and fit the criteria you are looking for
snowball: researchers may find a few PPs then ask that PP to bring other PPs
quota: the sample frame will be divided into smaller groups e.g. class, age. a set number will be taken from each group.
random: like putting names in a hat, everyone in the research population has a chance to be involved, as names are randomly selected.
stratified: the sample frame gets divided into a number of social groups e.g. class and gender. the researcher would then randomly select proportional amounts of people from each group. represents society.
systematic: selecting the nth person until you reach until you reach the desired sample size.
types of observations
non-participant: where the researcher simply observes behaviour.
participant: when the researcher joins the group they are researching
covert: undercover
overt: obvious
participant observations evaluation
insight: a strength of this method as it allows researchers to truly experience other people’s lives and put themselves in their position. observation allows a large number of qualitative, rich data.
access: this method of research is also good for accessing new certain groups.
flexibility: this is a flexible method as it allows new hypotheses to be formed, and it allows the research to go off in any suitable direction.
limitations: time-consuming, large amounts of data that are hard to quantify, dangerous, demanding, requires specific skills, some groups are resistant to be studied.
issues of observations
interpretivists prefer this method as it allows rich, detailed, and authentic data
validity through involvement: valid method - people are joining in research
flexible and grounded theory: change/manipulate the hypothesis.
positivism rejects PO as a scientific method as it lacks objectivity, not reliable and representative.
representativeness: small sample size
reliability: cannot be repeated
bias: may feel sympathetic or make friends - skewed data
lack of validity: internal - are you testing what you wanted to test/investigate? external - can findings be applied to other areas?
hawthorne effect: if you’re being watched, you change your behaviour.
secondary sources: official statistics
government data e.g. birth rate, death rate, crime etc
advantages:
free sources
large-scale studies that have cost a huge amount
high response rate
allow us to make comparisons
collected at regular intervals
disadvantages:
government creates stats for their agenda
may not be available on the topic you want to investigate
can have a mismatch of data and therefore is difficult to look at a correlation
terminology may be used differently
official statistics: theoretical issues
positivism: quantitative data = numbers and percentages.
representativeness: large sample size: representativeness age, gender, occupation = generalise data
reliability: repeated = reliable
interpretivism:
hard and soft stats: large and small sample
believe some stats are misleading
marxism and stats: stats collected and analysed by the rich = class differences.
feminism and stats: data is collected and interpreted by males: gender bias
documentation
includes written and other texts: diaries, letters, emails, texts, paintings, drawings, and photographs.
sociologists use public documentation (produced by gov, schools, businesses and charities), personal documents (letters, diaries, autobiographies), and historical documentation.
practical issues:
might be the only source of information available
free, cheap source of information
saves time
not always possible to access
individuals create documents for their purposes.
documentation: theoretical issues
validity: is it testing what its supposed to be testing?
reliability: cant always be repeated
representativeness: small sample sizes= not representative
big sample sizes= representative
content analysis
a method used for dealing with large amounts of quantitative data, usually mass media. there are 2 types: formal content analysis and thematic analysis.
formal content analysis: this allows a quantitative analysis. this would mean the information is coded/categorised and quantified. this is favoured by positivists, however, interpretivists say there is a lack of validity as counting does not produce meaning.
thematic analysis: this is a qualitative analysis used for media texts by interpretivists and feminists this involves a small number of in-depth cases. the aim is to encode meanings. however, this can be criticised:
small samples
selective evidence, therefore not valid
no proof that the meanings allocated are the truth.
