research methods Flashcards
primary sources
first hand data generated by the researcher themselves
- observations
- questionnaires
- interviews
strengths of primary data
- fit for purpose
- researcher is aware of any problems
weaknesses of primary data
- more costly
- more time consuming
secondary sources
involves researcher taking advantage of data which already exists
- official statistics
- public/personal documents
- media
strengths of secondary data
- cost and time efficient
- may be the only source of information for historical topics
weaknesses of secondary data
- not fit for purpose
- researcher unaware of problems with the data
quantitative data
data that revolves around counting or measuring things
- official statistics
- closed ended questionnaires
- experiments
strengths of quantitative data
- fairly easy to analyse with objective procedures
- easy to make comparisons
- usually collected in a systematic, objective way
- can be more representative
- tends to be reliable
weaknesses of quantitative data
- lacks detail
- may not be totally objective if sociologist displays bias
- tends to lack validity
qualitative data
data which takes the form of words
- unstructured interviews
- open ended questionnaires
- participant observation
strengths of qualitative data
- more in depth and detailed insight into meanings and motives
- tends to be more valid
- allows new insights
weaknesses of qualitative data
- lacks reliability
- more difficult to analyse so costly and time consuming
- may be unrepresentative
- it is subjective
factors affecting the choice of a topic
- personal reasons
- availability of data
- funding/policy issues
- interests of society
- theoretical perspective of the researcher
longitudinal research
where individuals are researched at intervals over a long period, sometimes even a lifetime
example of longitudinal study
Parker - studied drug use in a group of 1125 teenagers over 4 years from 14-18 and found that the types of drugs used changed over time and that w/c people experimented from an earlier age but by 18 the m/c had caught up
strengths of longitudinal studies
- can observe changes in a group over time
- eliminates the problem of individual differences
weaknesses of longitudinal studies
- panel attrition (people dropping out)
- keeping in touch with participants
- difficulties of analysing large amounts of data
cross sectional studies
where different groups of individuals are compared at different stages of their lives
- eg compare a group of 15 year olds with a group of 20 year olds with a group of 25 year olds
practical factors of research methods
- time
- cost
- subject matter
- personal characteristics
- hardship/danger
- recording data
ethical factors of research methods
- informed consent
- deception
- privacy
- confidentiality
- harm
- guilty knowledge
theoretical factors of research methods
- validity
- reliability
- representativeness
- positivism v interpretivism
validity
truthful, meaningful and accurate reflection of reality and what it set out to measure
reliability
data is reliable if a piece of research is reproduced with the same methods and findings are consistent and replicable
representativeness
would like research to be representative of wider society, a larger or more diverse group will be more representative
positivism
believe sociology should be like the natural sciences and human behaviour should be studied in a non-biased and objective way
interpretivism
believe sociology is nothing like the natural sciences because what it studies is very different and therefore cannot be non-biased and instead subjective
example of positivist research
Durkheim suicide statistics - used official statistics to find out precise links between suicide rates in different social groups
example of interpretivist research
Atkinson suicide interviews - interviewed coroners and found that they have a picture of a typical suicide and suicide victim (eg road deaths are rarely classed as suicide but hanging, drowning or drug overdose are)
overt observations
the researcher is honest and the group knows they are observing
strengths of overt observations
- direct questions
- more ethical
- easier note taking
- less danger of hostility if cover is blown
- easier to get in and out
weaknesses of overt observations
- may still remain an outsider to secrets
- hawthorne effect
- some groups may not allow entry
covert observations
researcher observes the group without them knowing
strengths of covert observations
- forbidden fruit (accepted and trusted into all secrets)
- less risk of hawthorne effect
- sometimes no other way
weaknesses of covert observations
- less ethical
- guilty knowledge
- harder to get in and out
- harder note taking
- danger of hostility if uncovered
participant observations
involves the researcher joining a group and their daily activities, covertly or overtly
non-participant observation
researcher observes the group from the outside and doesn’t join in
ethnography
an in depth study of a group or culture which tries to gain insight into the way of life of a group of people
example of covert participant observation
Humphrey’s tea room trade:
- adopted the role of ‘watch queen’ for homosexual men in public toilets to get to know the gay community then pretended to conduct a survey on health based on this information and found that over 50% were married in tried to live a respectable life style that didn’t reveal their homosexuality (this was the 1970s when it was not acceptable to come out)
example of overt participant observation
Venkatesh gang leader for a day:
- spent nearly 10 years conducting overt observation on a crack gang in Chicago, giving him guilty knowledge of the crimes they carried out (first day he approached the gang leader with a questionnaire and was held at knife point but managed to befriend them despite his research being overt)
strengths of participant observation
- only way to study difficult areas
- confidant/rapport
- ethical considerations depend whether overt or covert
- verstehen (empathetic understanding of seeing the world)
- getting to the truth
weaknesses of participant observation
- time consuming, expensive and dangerous
- hawthorne effect if overt
- ethical issues if done covertly
- going native (judgement becomes clouded within group perspective)
- lack reliability and representativeness
- can be subjective
- observer characteristics can influence behaviour
strengths of non-participant observation
- more cost/time efficient
- possibility of using quantitative data and more reliable
- more ethical (not invading privacy)
- objectivity as not participating
- reduces hawthorne effect
weaknesses of non-participant observation
- may not be practically possible on some groups
- causes ethical issues
- lacks validity as outsider perspective
- observer characteristics can influence participants
case studies
a detailed examination of a single sample of something such as an institution, social group or even an individual
strengths of case studies
- really in depth and provide rich and valid data
- can generate new hypotheses to be further tested later
- good for providing new insights into a subject
- can prove or disprove a general statement
- may be the only way to study a historical topic
weaknesses of case studies
- unrepresentative
- unreliable
- time consuming and expensive
experiments
the main idea is to prove or disprove a hypothesis made at the start of an experiment
dependent variable
the factor measured in an experiment
independent variable
the factor changed or manipulated in an experiment
control variable
the factor kept the same in an experiment
confounding/extraneous variable
the factor that is failed to be control and influences an experiment
lab experiment
conducted in an artificial, controlled setting and preferred by positivists
field experiment
conducted in a real life setting where individuals are often unaware they are participating and preferred by interpretivists
example of a lab experiment
Zimbardo Stanford Prison experiment:
- involved 24 male volunteers in an artificial prison setting in Stanford university split into guards and prisoners, but they found each group took their roles too seriously, with prisoners experiencing punishment and degradation while the guards took their roles too seriously
example of a field experiment
Rosenthal and Jacobson Pygmalion in the Classroom:
- labelled random 20% of children as ‘spurters’ to teachers and found that this group experienced higher growth in IQ due to labelling and teacher expectations
strengths of lab experiments
- high control
- can be ethical due to consent
- positivists say they are objective and scientific
- reliable
weaknesses of lab experiments
- society cannot be replicated in the lab
- unrepresentative sample
- can be unethical if harmful
- lack validity as artificial
- hawthorne effect
- may bring experimenter bias due to characteristics or mannerisms
strengths of field experiments
- may be the only way to study complex situations
- less artificial
- reduced hawthorne effect
weaknesses of field experiments
- less control
- can be less ethical
- less reliable
pilot studies
small scale preliminary study conducted before the main research to check feasibility of a study and improve its design
key stages in carrying out a social survey/questionnaire
- aim and hypothesis formation
- operationalising concepts
- questionnaire design
- administering questionnaires
- questionnaires and validity
examples of questionnaire based research
Cumberbatch - horror films:
- gave young people questionnaires based on what horror films they had seen and many reported seeing ones which did not exist due to social desirability bias
S. Hite - sex survey:
- sent out 100,000 sex survey questionnaires and only received a 4.5% response rate
strengths of questionnaires
- quick and cheap way of gathering a lot of data
- no need to train interviewers for self-competition questionnaires
- quick and easy to analyse sets of data
- few ethical problems
- tend to be reliable
- objective
weaknesses of questionnaires
- may be incomplete, illegible or incomprehensible
- data tends to be limited and superficial
- don’t know if intended person filled it in
- low response rate
- volunteer bias
- inflexible
- may lack validity if answers are fixed or participants aren’t honest
- subjective as researcher interprets data themselves
group interviews/focus groups
some interviews take place with one interviewer and a group of interviewees to observe the group context
example of a group interview
Willis ‘the lads’:
- interviewed the w/c boys in a group to understand normal interactions in their school activities
strengths of group interviews
- rapport due to informal environment
- open up and bounce off ideas
weaknesses of group interviews
- may find conformity to others
- may find dominator of group
- lack anonymity/confidentiality
- exaggeration/embarrassment
structured interviews
orally administered questionnaire with pre-defined questions
example of structure interview
Wilmott and Young - Bethnal Green:
- used structured interviews on their study of family life in Bethnal Green and found only 54/933 did not participate
strengths of structured interviews
- higher response rate
- use with low literacy
- more representative
- informed consent
- more reliable
- clarify questions and answers
weaknesses of structured interviews
- more expensive as have to train interviewers
- more pressured to respond
- structure limits validity
- interviewer bias
- social desirability, ‘screw you’ and ‘right answerism’
unstructured interviews
the interview is much more open ended, maybe with a list of topics they want to explore but no list of questions
strengths of unstructured interviews
- rapport
- informed consent
- freedom to express themselves
- clarify meanings
- flexible
weaknesses of unstructured interviews
- time consuming to conduct and analyse
- may produce irrelevant content
- over confide leaving guilty knowledge
- unreliable
- interviewer bias from relationship
- social desirability, ‘screw you’ and ‘right answerism’
example of unstructured interviews
Dobash and Dobash - domestic violence:
- found it a useful method when researching such a sensitive subject as the rapport that developed conveyed the true horror of domestic assaults which quantitative data diminished
sampling frame
a list of all member of the target population you are interested in studying
random sampling
- manual selection - members are given a number and numbers are drawn out like the lottery
- computer selection - members’ names are stored electronically and computer software generates a random sample
advantage of random sampling
should produce a bias free cross section of a population
disadvantage of random sampling
does not guarantee a representative sample
stratified sampling
target population is sampled according to known criteria of different groups within that population (if 60% of population are female then 60% of sample is female)
advantage of stratified sampling
sample is representative of other factors within the population
disadvantage of stratified sampling
problem of deciding what categories to us and measurement of these categories
quota sampling
similar to stratified sampling in that they decide how many individuals from a social group are required and stops drawing them once a particular quota is met
advantage of quota sampling
if they are aware of the makeup of a general population they can use this as a guide to produce a representative sample
disadvantage of quota sampling
bias could creep into sample and there is possibility of members of quota not being representative of general population
self-selecting/volunteer sampling
individuals will directly volunteer to take part in research
advantage of self-selecting/volunteer sampling
easy to gain willing participants
disadvantage of self-selecting/volunteer sampling
can account volunteer bias in the type of people who volunteer
snowball sampling
sample created by contact with one member of a population who suggest other individuals who may be willing to participate
advantage of snowball sampling
good for gathering participants when studying a sensitive subject matter or an issue with few people involved
disadvantage of snowball sampling
high likelihood of unrepresentative or biased sample
cluster sampling
where a cluster of individuals within a population are targeted, such as one area of a town or even one street in a w/c area
advantage of cluster sampling
sometimes the only suitable type because a population spread over such a large area
disadvantage of cluster sampling
highly likely to be unrepresentative
theoretical sampling
some argue it is best to have a non-typical sample to illustrate particular theoretical insights (feminists have studied societies where women have not traditional gender roles to demonstrate sex roles are socially constructed)
advantage of theoretical sampling
can be the only way to illustrate particular theoretical standpoints
disadvantage of theoretical sampling
almost impossible to generalise from them to typical groups
opportunity sample
researcher just uses individuals that are ready to hand to take part
advantage of opportunity sampling
easy to gain willing participants
disadvantage of opportunity sampling
taking advantage of available participants is unlikely to be a representative sample
pro-large samples
some researchers believe that the larger a sample size is the better it is as bias is less likely to enter as well as the sample being more representative
anti-large samples
large samples can be expensive and time consuming and disguise important trends within the sample
public documents
official publications:
- official statistics
- government reports
- media
- historical documents
personal documents
not official publications:
- diaries
- letters
- text messages
- emails
- blogs
- suicide notes
- photos
- oral/family histories
strengths of historical/government documents
- only source of historical information
- cost and time efficient
- few ethical issues
- verstehen (insider perspective)
- government research is on a larger scale so representative
weaknesses of historical/government documents
- no data available on some subjects
- problems with accessing data
- problems analysing lengthy documents
- problems of interpreting different cultures or eras
- may have been manipulated
- authenticity problems
- source may not be credible
- may be unrepresentative and not give full picture
example of document research
Laslett - marriage and families in the past:
- used church records from the 17th/18th centuries to find that family size and age of marriage were roughly the same as in the 1960s
strengths of personal documents
- only way to gain insight if involved cannot be interviewed
- verstehen (insight into how people felt)
- can be valid as personal and not written for a particular audience
weaknesses of personal documents
- difficulty accessing
- ethical issues consent and harm with private information
- could lack authenticity
- personal bias
- misinterpretation
- subjective
quantitative content analysis
researcher develops a number of categories and counts how frequently material fits in with these categories
example of content analysis
Lobban - primary school reading:
- looked at primary school reading schemes and the roles taken my males and females and found females were more often found in domestic tasks
qualitative content analysis
semiotics or thematic analysis where the researcher describes what they think the media is showing
strengths of content analysis
- cost/time efficient as media available
- fewer ethical issues
- objective if quantitative
- reliable if strict coding scheme
weaknesses of content analysis
- problems getting representative sample
- can be time consuming
- problems of interpretation
- may be problems with quantification or using coding schemes
- subjective if qualitative
- interpretivists say it tells us nothing of the media’s meaning
official statistics
quantitative data gathered by the government or other agenices
example of official statistics in research
Durkheim’s study of suicide
strengths of official statistics
- cost and time efficient
- easy comparison of trends between groups and time
- fewer ethical issues
- representative sample
- believed to be credible and valid
- reliable
- positivists see statistics as social facts
- objective
weaknesses of official statistics
- not fit for purpose
- problems with definitions
- may have been manipulated to make government look better
- may contain inaccuracies or missing information
- ‘soft’ statistics lack validity
- socially constructed and subject to a variety of biases
- subjective when interpreted by sociologist
