Research Methods (YR12+13) Flashcards
define primary data
- primary data = data collected first hand from sociologists for their own use
define secondary data
- secondary data = data collected by other sociologists, Gov departments or official bodies (e.g. charities) or individuals
give examples of primary data
- interviews, questionaries, experiments
give examples of secondary data
- official statistics (crime rates, league tables), documents (newspapers, diaries)
outline the advantages + disadvantages of primary data
- A: more accurate/ specific to the topic being studied. more trustworthy
- D: time consuming, expensive, cant be compared over time
outline the advantages + disadvantages of secondary data
- A: cheaper, quicker
- D: less accurate/ specific to topic being studied
define + give examples of quantitative data
- quantitative data = data in the form of numbers
- e.g. percentages, statistics
define + give examples of qualitative data
- qualitative data = data in the form of everything other than numbers
- e.g. interviews, documents, pictures
what are the advantages + disadvantages of quantitative data
- A: can spot trends and compare data throughout time
- D: doesn’t provide thorough insight into peoples views
what are positivists 5 main viewpoints
1) society can be measured objectively
2) prefer quantitative data + methods
3) society/ institutions exerts influence over its members + shapes their behaviour (structural ideology)
4) P seek cause and effect
5) P prefer questionnaires, structured interviews, experiments, and official stats as they produce reliable and representative data
what are interpretivists 5 main viewpoints
1) society cannot be measured objectively
2) prefer qualitative data + methods
3) individuals construct their own reality (social action)
4) I seek to gain a subjective understanding of individuals meanings
5) I prefer participant observation, unstructured interviews + personal documents as they produce valid data
what are the advantages + disadvantages of qualitative data
- A: provides rich insight into peoples feelings + views
- D: hard to compare and analysis can be time consuming
define triangulation
- triangulation = the combined use of diffferent types of methods (e.g. both quant. and qual. methods) to counteract the limitations of each method
outline Validity
Validity
- how close the results are to the truth
what 3 factors should be taken into consideration when researching teachers
- power relationships; teachers have more power; may be translated on researchers. they should pose as TAs or supplies to equalise the power imbalance
- teachers are over worked; may be less cooperative
- teachers are used to being scrutinised (e.g. OFSTED)/ HAWTHORNE EFFECT (people behave differently when they know they’re being watched); may not give valid answers if it reflects badly on school
what 3 factors should be taken into consideration when researching schools
- is a formal organisation and has a formal hierarchy; headteachers may disapprove of possible interruptions, or in same sex skls, an opposite gendered researcher may cause implications
- headteachers hold the power; they are the gatekeepers who have the power to refuse researcher’s access to their school
- legal framework creates certain requirements; there is a strict legal duty of care that schools have for their pupils
what does PERVRT stand for
Practical
Ethical
Reliable
Valid
Representative
Theoretical perspective
outline Practical - MEAT
Practical
- the MEAT (Money, Effort, Access, Time) of a method that could affect the research
outline Ethical
Ethical - Cerys Is Very Crazy
- Confidentiality; identity of Ps kept private
- Informed consent; Participants made aware of all elements. can back out of/ refuse to take part in research
- Vulnerable groups; special care for vulnerable groups (children, disability, mental health)
- Covert research; hiding the researchers identity and topic of research can cause ethical problems
outline Theoretical perspectives
Theoretical perspectives
- Interpretivists + Positivists perspectives
outline Reliable
Reliable
- can it be replicated to get the same results
- Positivists care about this
outline Representativeness
Representativeness
- can the data be generalised to a wider population
- Positivists care about this
what 3 factors should be taken into consideration when researching in classrooms
- controlled social setting; highly controlled environment - behaviour observed may not be accurate
- may disguise real thoughts; teachers + pupils are used to disguising true feelings; may conceal true feelings from researcher
- peer pressure as influence; young people are more sensitive to peer pressure - reduces validity
what 3 factors should be taken into consideration when researching young pupils
- power and status; they have less power, may not feel comfortable
- ability; comprehension skills + vocab are developing - researchers have to take more care when forming Qs (OPERATIONALISING)
- vulnerability; are more vulnerable to physical + psychological harm
outline the 4 steps of the research process
1) deciding on hypothesis/ aim; a statement outlining what you intend to study -e.g. ‘does class affect achievement’
2) operationalise concepts; turning sociological concepts into measurable data - e.g. to find someone’s social class ‘what is your job?’
3) pilot study; a small scale trial run done to find any flaws to save money + time for the real study
4) choose your sampling frame (e.g. electoral roll) and technique to find a sample
what 3 factors should be taken into consideration when researching parents
- interactions; parents raise their kids individually + differently - not all kids react the same way
- are mostly outside of school and so harder to research; parents play a vital role in child’s education but most of their interactions happen at home which is closed off to research
- some may be more/ less willing to participate than others; class, gender, ethnicity play in apart in how willing they are- e.g. M/C parents more likely than W/C parents - creates unrepresentative data
outline simple random sampling + its As and Ds
- every member of the population has an equal chance of being selected; e.g. pulling names out of a hat
- A: practical, easy
- D: might not be representative
outline systematic random sampling + its As and Ds
- systematically selecting people from the sampling frame by choosing every 5th or 15th person, for example
- A: replicable
- D: may be unrepresentative
outline stratified random sampling + its As and Ds
- selecting members in proportion to what % of the population they make up. e.g. 2.5% of Brits are of Indian origin, so 2.5% of the sample will be of Indian origin
- A: representative - it reflects society accurately
- D: is time consuming
outline volunteer sampling + its As and Ds
- made up of individuals who have decided to be involved in a study; e.g. through a newspaper advert
- A: practical
- D: not representative
outline snowball sampling + its As and Ds
- finding participants based on contacts. participants find other participants through word of mouth
- A: practical; easy
- D: not representative
outline quota sampling + its As and Ds
- separating a sample into catergories (e.g. male and female) and then looking for specific qualities of that sample (a quota) (e.g. 100 males, 100 females)
- A: representative
- D: lacks practicality
outline opportunity sampling + its As and Ds
- selecting those who are easiest to access at the time; e.g. interviewing people on the street
- A: practical; quick, cheap, easy
- D: may not be representative of population
outline laboratory experiments
- experiments in a controlled, artificial environment that use scientific methods to test a hypothesis
- can control the variables
- there is an experimental (exposed to independent variable) + a control group (not exposed to indep. variable)
- can discover cause + effect and patterns
outline field experiments
- takes place in the subjects natural surroundings
- those involved arent aware that they are involved inn an experiment - eliminates Hawthorne effect
- the researcher isolates / manipulates one/ more variables of the situation
outline the comparative method
- the comparative method is carried out only in the mind of the sociologist - e.g. Durkheim’s study of suicide
- its a ‘thought experiment’
- it re-analyses secondary data and discovers cause + effect relationships
- the researcher identifies 2 groups that are alike apart from the variable being studied. then compare the 2 groups to see if the variable has any effect
- this method avoids artificiality, can study past events, avoids deceit
- BUT, it gives researcher less control, unreliable
outline lab experiments in terms of PERVRT
- P: lacks, is more difficult to conduct, costly, time consuming
- E: lacks, lacks informed consent, deception - e.g. Milgram, Zimbardo
- R: has, standardised procedure + detached researcher, controllable variables, produces quant data, a detached + objective method
- V: lacks, Pos; is a controlled environment - isnt the real world, Hawthorne effect, lacks internal validity (findings arent true for Ps due to artificial environment)
- R: lacks, smaller sample size, lacks external validity (aren’t generalisable to wider pop)
- T: Pos favour it due to reliable nature, Int dont due to lab experiments search for causes - Ints disagree w/ cause and effect
outline field experiments in terms of PERVRT
- P: lacks, gatekeepers can prevent access to an environment (e.g. school). can be time consuming
- E: lacks, potential for harm, lack of informed consent, e.g. Jacobsen + Rosenthal
- R: lacks, procedures may differ in different settings and so cant control all variables
- V: has, is a less artificial setting. people are unaware of them being studied - lack of Hawthorne Effect
- R: lacks, hard to obtain a perfectly representative sample
- T: favoured by Int, not by Pos
outline questionnaires
- close ended: Ps choose from a limited range of possible answers selected in advance
- open ended: Ps are free to answer however they wish without pre selected answers by the researcher
outline questionnaires in terms of PERVR
- P: has, easy, cheap + quick to send off, accessible to all, data is easily quantifiable to identify trends, no need to train interviewers, BUT; low response rate on postal, are inflexible, Q are snapshots (fails to capture changes in attitudes)
- E: has, you can choose to not answer the questionnaire
- R: has, easy to repeat due to standardised nature (Qs are a standardised measuring instrument), easily comparable
- V: lacks, close-ended Qs with pre set answers restrict the interviewee’s responses. pre set answers may not reflect the respondent’s view - doesn’t give true picture. ‘right-answerism’
- R: lacks, certain people are more likely to respond - e.g. unemployed, BUT; large scale, Qs more likely to use R sampling methods
outline questionnaires in terms of Theoretical perspectives
Positivists:
- like: Questionnaires produce representative findings that are generalisable + reliable + objective + tests hypothesis + scientific/ objective
Interpretivists:
- dislike: Qs are invalid due to objectivity of researchers/ detached nature which disallows valid data + fails to provide a true picture
- lying/ forgetting from respondents skews validity
outline positivists hypothesis testing with questionaires
- Ps seek to discover cause + effect
- Questionnaires allow Ps to identify c+e and test hypothesis
- for Ps to test hypothesis, they must identify a link between the 2 variables
- Qs establish correlations from which Ps can construct a hypothesis
define standardised measuring stick
- standardised measuring stick = refers to a research method (e.g. questionnaires) that can be used by any researcher - allowing research to be easily replicated + checked by another
- also means that any differences in answers is due to the real differences of respondents - not the researcher
outline interviews
- structured: every interview is conducted in the same standardised procedure - same questions, order, one of voice.
- unstructured: like a guided conversation, interviewer is free to vary the qs, wording, order etc.
- group: usually unstructured, includes focus groups
- semi-structured: same set of questions, but the interviewer can probe for more info + additional qs
outline unstructured interviews in terms of PERVR
- P: lacks, time-consuming, must thoroughly train interviewers (costly), unable to quantify data easily
- E: has, is good for researching sensitive issues as researchers can build a rapport + a relationship. e.g. Dobash and Dobash
- R: lacks, due to natural flow of conversation - its not replicable and wont give you the same results
- V: has, no set Qs and so interviewee can speak at length about topic, gives rich qual. data, informality allows a rapport (trusting relationship), allows sensitive topics to be discussed
- R: lacks, there are fewer interviews due to lengthy nature of them and so a less generalisable sample
outline unstructured interviews in terms of Theoretical perspectives
Interpretivists:
- like: rich, authentic qualitative data - understanding comes through involvement
- grounded theory: we build up + modify our hypothesis throughout the research, based on facts we learn - for this, UI are best
- open-ended Qs allow for in-depth/ unrestricted answers/ the true meaning
Positivists:
- dislike: UIs lack objectivity + reliability due to lack of standardised measuring instrument
- answers cant be easily quantified + categorised and so cant correlate variables, test hypothesis, or establish cause + effect links
- unrepresentative data due to small sample size + subjective
outline the 3 values of feminist approaches to research
- value committed: takes the w’s side + aims to voice their experiences + to free them from patriarchal oppression
- involvement: researchers should be involved rather than detached from the research
- equality + collaboration: there should be this between the researcher + subject, rather than being hierarchal
outline structured interviews in terms of PERVR
- P: has, is quick, cheap to conduct, can cover large numbers of people, data is easily quantifiable, suitable for hypothesis testing, small cost for training interviewers, high response rate
- E: may lack if researching vulnerable groups, e.g. children, pressure to answer sensitive Qs, anonymous
- R: has, pre set Qs allow procedure to be repeated with similar results, easily comparable
- V: lacks, close-ended / pre set Qs create limited choice for answers - leads to invalid data
- R: has, quick nature of interviews allows for a large number of people to be studied
outline structured interviews in terms of Theoretical perspectives (+ Feminists)
Positivists:
- like; SI are representative (large numbers can be surveyed, high response rate), reliable (easy to replicate), objective, can test hypothesis (can est correlations)
Interpretivists:
- dislike; provides a false picture; invalid due to close ended Qs, people can lie/ exaggerate
Feminists:
- dislike: relationship between the researcher + researched reflects the exploitative nature of patriarchal society
- strict division of labour (researcher as active role, interviewees as passive role)
outline observations
- non-participant observation = the researcher observes the group without taking part
- participant observation = the researcher takes part in the group whilst observing
- structured non-participant observation = uses a structured observational schedule (pre-determined list of behaviours the interviewer is interested in). each time a behaviour is observed, it’s noted - produces quant data
- overt observation = researcher’s role + purpose is known to those being studied
- covert observation = researchers role + purpose is unknown, they pose as a genuine member of the group
outline ‘getting into’ observation groups
- making initial contact with a group will depend on personal factors such as having the right connections - may be gatekeepers to access - e.g. headteachers
- the researcher will have to win the group’s trust to overcome their suspicions - their age, race, gender etc may be an obstacle if it differs from group
- once entered, the R must adopt a specific role that is non-disruptive + a good vantage point for observations
outline ‘staying in’ observation groups
- when accepted, the R must be involved enough to understand the group fully whilst staying detached to remain unbiased
outline ‘getting out’ of observation groups
- leaving a group the R has become acclimatised to and then re-entering reality can be difficult
- loyalty to the group may prevent the R from fully disclosing all details to protect them
outline overt observation in terms of PERVRT
- P: lacks, may not be bale to gain access (e.g. to schools), have to work around timetables/ holidays, and schools may be less likely to allow your access if they know you’re a researcher
- E: has, has informed consent, no deceit
- R: lacks, no standardised, repeatable procedure
- V: lacks, doesn’t as provide in depth info + Hawthorne Effect/ impression management
- R: has, schools are representative of population
- T: pos: like structured NPO; can easily quantify data - don’t like PO. Int: like PO - gives rich data
outline covert observation in terms of PERVRT
- P: lacks, issue of access - headteacher acts as gatekeeper
- E: lacks informed consent - issue of deceit
- R: gives qual data
- V: has, more valid data - no Hawthorne effect/ no impression management. lacks, cant openly take notes - take them from memory
- R: has, schools are representative of population
- T: Positivists dislike, Int like
outline participant observation in terms of PERVR
- P: lacks, time-consuming, must be trained, issue of access, hard to quantify large amounts of qual data. BUT, PO may be the only option for access - e.g. deviant people suspicious of RM like interviews, first-hand experience provides great insight (verstehen), its flexible (can enter the research open-mindedly + research Qs when new situations arise)
- E: lacks, if covert, there is issue of deceit
- R: lacks, doesn’t have a standardised procedure, so not easily replicable. also, produces qualitative data so cant easily compare data / identify trends
- V: has, provides rich qualitative research, ‘verstehen’, can build a rapport, flexibility, BUT, Hawthorne effect (in overt O)
- R: lacks, PO study groups are small + the sample wont perfectly reflect population in the way a stratified random sample would
outline participant observation in terms of Theoretical perspectives (+ social action)
Interpretivists:
- like: gains insight into actor’s meanings
- the R can gain a subjective understanding of the subject’s life through their high level of involvement (verstehen)
Positivists:
- dislike: lacks representativeness due to small sample size and so isnt generalisable
- lacks reliability as PO isnt a standardised, scientific measuring instrument
- lacks objectivity - R can get attached/ loyal to subjects
social action:
- like: PO allows for observations of micro-level interactions first-hand
define verstehen
- verstehen = understanding someones lived experience (putting yourself in someone else’s shoes)
outline non-participant observation in terms of PERVRT
- P: lacks, time-consuming, issue of access, must be trained
- E: lacks, if covert - issue of lack of informed consent
- R: lacks, no standardised procedure, so not easily replicable. also, produces qualitative data - cant easily compare data / identify trends
- V: lacks, wont be as much of a true picture
- R: lacks, PO study groups are small + the sample wont perfectly reflect population in the way a stratified random sample would
- T: Pos like structured NPO
outline documents
- is a secondary source
- documents include: written texts (diaries, letters, texts, state records etc), other texts (paintings, music, photos etc)
- 3 types of documents: public (produced by organisations - e.g. media output, Ofsted), personal (e.g. letters, photo albums, social media pages), and historical (personal or public docs made in the past)
outline documents in terms of PERVR
- P: has, cheap + easy access for most Ds - not all, may be only available source for historical info, saves time. lacks, is it authentic, not always possible to access
- E: has, may lack consent of the person who owns that D (esp. if historical)
- R: lacks, Ds are unique + not standardised
- V: has, verstehen. lacks, Ds can be personal + not objective - can create a false view of someone - e.g. diaries - was it written to be published?
- R: lacks, with Hist. Ds, not representative + typical of all views of that period, aren’t many Hist Ds - cant generalise, some groups may not be represented in docs (e.g. the illiterate), not all Ds are accessible, not all Ds survive
outline documents in terms of Theoretical perspectives
Interpretivists
- like: provides a valid picture of actors’ meanings, lack of Hawthorne effect as they aren’t written for sociological use
Positivists
- dislike: theyre unreliable - they arent standardised, unrepresentative as some may not be represented in docs (e.g. the illiterate),
outline Scott’s 4 criteria for assessing document’s validity - CRAM
CRAM:
1) Credibility: is it believable + sincere, diaries of politicians may be intended for publication, does it historically match up to the events of the time?
2) Representativeness: are other Ds of the period typical to that one, not all survived Ds are available for public use,
3) Authenticity: is the D what it claims to be, are there missing pages, who wrote it
4) Meaning: the researcher may need special skills to understand a D - e.g. translate from another language. also, meanings change over time
outline official statistics
- OS are produced by the gov/ official bodies
- they are a secondary source
- the yearly UK Census is a major source of OS
- OS are collected for policy-making
- there are also non-official statistics made by non-state organisations - e.g. by charities
outline official statistics in terms of PERVR
- P: has; is a source of lots of data, allows comparison of trends/ patterns between groups over time, state holds power to get data that sociologists cannot, high response rate. lacks; the stats are made for Govt use - not sociologists’ - may not offer vital sociological data, definitions used may be different (e.g. truancy) + change over time,
- E: has, compared to other methods
- R: has, standardised procedure, uses a standardised measuring instrument. lacks, occasional random errors in filling out the forms/ in coding
- V: there can be a ‘dark figure’ of unrecorded data which skews the V of OS. e.g. some births, marriages go unrecorded
- R: has, OS like the Census are taken by the whole population, due to big budget it has a large sample size
outline official statistic’s Theoretical perspectives from Positivists + Interpretivists
Positivists:
- like: ‘OS are social facts’, are objective, reliable, representative, quantitative, can identify patterns, test hypotheses
Interpretivists:
- dislike: invalid as OS are socially constructed - they represent labels attached to people - e.g. OS on mentally ill people reflect those who get a diagnosis, which isnt everyone - OS don’t reflect a ‘real rate’
outline official statistic’s Theoretical perspectives from Marxists + Feminists
Marxists:
- OS arent objective facts, they reflect the interests of the ruling class
- OS are a part of Althusser’s ISA - they distort reality to maintain the capitalist’s power - e.g. evidence of exploitation isn’t published
- the definitions change; e.g. the definition of unemployment is changed to reduce the official numbers of unemployed
Feminists:
- state creation of OS legitimises gender inequality - e.g. there are few stats on women’s unpaid domestic labour - underestimating w’s economic contribution
define soft statistics
- soft statistics = data that is difficult to measure as its people’s opinions or feelings
- reflects a picture of administrative agencies, not a real world picture
- SS neglect a dark figure of unrecorded cases - e.g. unreported incidents of racism in school
define hard statistics
- hard statistics = reliable + valid data taken from official or organisational bodies
- HS are created from registration data
- there is little dispute over the categories used to collect data - e.g. death, divorce
outline ‘going native’
- ‘going native’ refers to a danger of participation observation where one over-identifies with the group and forms a bias
- they have stopped being an objective observer and have become a member of the group
outline relationship to education for RM analysis
- the research group may have a more positive or negative relationship with education
- how will this impact the findings
- how will the method interact with this issue
- e.g. ethnic minorities/ WC may have a negative relationship with education
define rapport
- rapport = a relationship built between the researcher and the participant(s)
- allows research to be more valid and in depth
what is a longitudinal study
- a study conducted throughout a long period of time
outline content analysis
- content analysis is a method for quantifying content - esp from mass media
- 2 types: formal content analysis + thematic analysis
outline content analysis
- 2 types: formal content analysis + thematic content analysis
- formal CA: produces quant data from qual by categorising and counting. positivists (objectivity + representativeness) and feminists like this
- thematic CA: a qual analysis of media content used by Ints + feminists. aims to reveal underlying meanings to uncover the author’s bias
