Research Methods - Types Flashcards
Structured Interview
Pre-written questions (closed)
Unstructured Interview
Questions not set just topic
Semi-structured Interview
Mix of open and closed questions
Group Interview
Unstructured Interview but with groups
Non-participant Observation
Observer plays no active role in the group
Structured- behaviour observed, coded and quantified (positivist)
Hawthorn effect - change behaviour cus someone’s watching
Participant Observation - Covert
Driven from the inside - full immersion and undercover
Theoretical: \+ verstehen - empathise with group \+ high validity \+ truth is more likely to be observed \+ can form rapport with group \+ only/best method to use with some groups - loss of objectivity - no reliability
Practical: \+Flexible in approach and behaviour towards group - Time consuming - costly (money and personal) - difficult to record data
Ethical:
- no consent
- no privacy and confidentiality
- no protection from harm
- possibility for illegal/immoral activity
- deception
Participant Observation - Overt
Driven from inside - identity known
Theoretical: \+ high reliability (schedule used) \+ produces quantitative data (positivist) - Hawthorne effect - reduce validity - lack of qualitative data
Practical: \+ able to record data - lack of flexibility (if schedule used) - time consuming - costly
Ethical:
+ infomed consent
+ no deception - able to brief/debrief participant
+ lower risk of researcher harm
+ lower risk of witnessing/ participating in illegal/immoral behaviour
- no privacy/confidentiality
Questionnaires
Quantitative data
Non-sensitive topics
Theoretical:
:😭💁🏼♀️🙈😫😩🙈💁🏽♀️💁🏼♀️😍🤭
Official statistics
Secondary
Produced and published by government
Surveys - census, British crime survey
Records - crime/unemployment/health/exam stats
Theoretical:
+ large samples thus generalisable (postivist)
+ produced regularly allows trends over time
+ reliable can be checked anytime(positivist)
- no reasons for trends. Low validity
- agrue not facts just results
Practical:
+ readily available (no time)
+ cost little to use
+ can be reused to test different variables
Ethical:
- deception - created by government, can be manipulated to show them in good light
Personal documents
Secondary
Letters, diaries, journals and photographs
Theoretical:
+ personal - a true picture of their life
+ detailed therefore valid and interpretivist
- biased to the person’s opinion
- subjective
- qualitative data therefore not reliable
- may not be the full truth
Practical:
+ not costly
- time consuming
Ethical:
+/- informed consent if permission is received
+ there is no manipulation
- how it is obtained (privacy and confidentiality)
- possession of illegal or immoral knowledge
Content Analysis
Used to analyse content of any text, a way of quantifying data.
Set predetermined categories and count the times a section of the text fits into each one of these categories.
Theoretical:
+ follows systematic process, therefore reliable
+ can be used in a variety of secondary sources
+ can uncover and quantify hidden messages and ideas
- doesn’t look in depth and meaning of the text - validity
- words and phrases can be analysed out of context
- relies on good initial coding, bad categories result in little usefulness
- relies on some subjective interpretation to fit the categories
Practical:
+ cheap and easy to conduct
+ quick
Ethical:
+ already available to public - confidentiality, deception and informed consent all met
- they manipulate results.
Triangulation/ methodological pluralism
Combine methods that result in quantitative and qualitative data in order to check and verify the reliability and validity of the data.
More representative and builds a fuller picture of what is being studied.
Theoretical
+ produce quantitative and qualitative data
+ fuller picture of what is being studied
+ increases validity and reliability
+ allows researcher to produce representative data which can be generalised
- can produce contradictory findings
Practical:
- can be expensive
- produces lots of qualitative data which is time consuming to analyse
Ethical:
- priority is given to one method, allowing researcher bias
Random Sampling
Probability - every member of the population has an equal chance of being selected
eg pulling names out of a hat
+ large samples; provides the best chance of an unbiased, representative sample
- large populations, time consuming to create a list of every individual
Stratified Sampling
Probability - divide target population into important subcategories, selecting members in the proportion they appear in the population.
+ deliberate effort is made to make the sample representative of the target population
- time consuming as subcategories have to be identified and proportion is calculated
Quota Sampling
Non-probability - like stratified sampling but goes out looking for the right number of people instead of using a sampling frame
+ often used for opinion poles and market research, it is simpler, quicker and cheaper than stratified sampling
- is less likely to produce a sample which is representative of the target population
- the researcher chooses who to sample which leads to bias
Systematic Sampling
Probability - choosing your participants from a sample frame using a system, eg every 1/3
+ a bit like random sampling therefore can give unbiased samples from large populations
- in some situations, using a system can make the sample biased
Purposive Sampling
Non-probability - researcher chooses individuals that fit the nature of the research. Researcher chooses a particular group of place to study because it is known to be the type that is wanted.
+ quicker and cheaper
- unrepresentative as it leads to bias
Opportunity Sampling
Selecting those people who are available at the time.
+ quick, convenient and economical - the most common type of sampling
- very unrepresentative and often biased by the researcher who will likely choose people who are helpful
Snowball sampling
Participants are selected from an initial contact, who puts the researcher in touch with other possible participants.
+ its a useful way of finding participants with certain attributes who might not wish to be found otherwise such as cult members or drug users
- can be very biased, people with more friends are more likely to be selected as participants
- some people might not want to be found
Volunteer Sampling
Sociologists advertise for research volunteers in magazines, newspapers, university noticeboards or via the internet.
+ faster method to obtain sample and one which is ethical as the respondents choose to take part
- could lack representativeness as people taking part may not be representative of the target population
