Methods of research (Week 9) Primary sources of data

Question 1

Primary Quantitative Data

Answer 1

Methods:
* Questionnaires
* Surveys
* Experiments
* Content analysis

Question 2

Primary Sources

Questionnaires

Answer 2

Questionnaires consist of written questions. eg:
* Postal questionnaires
* Researcher-administered questionnaires (completed in the presence of researcher, with respondents answering questions verbally).

Question 3

Primary Sources

Postal Questionnaires

Answer 3

• Questionnaires involve closed ended questions/pre-coded. Which involve the researcher to provide a set of answers that the respondent can choose.
• The researcher limits the responses that can be given by offering multiple choice
  Eg. Do you own a sociology textbook?
  Yes - No - Don’t know
• Questionnaires also can have open-ended questions and for these the researcher doesn’t provide possible answers.
• The respondent answers in their own words which allows the researcher to gain deeper insight.
  eg. What was your day like today?

Question 4

Questionnaires - Strengths

Answer 4

• Pre-coded questions make it easier to quantify data because the options are already known, they are limited in number and easy to count.
• Closed questions are quick and easy to code; (a simple count of the number of responses).
• Closed questions are useful when researcher needs to contact large numbers of people quickly and efficiently.
• Can result in highly reliable data because everyone answers the same questions.
• Easy to replicate the research.
• Because they are anonymous, the validity of the research is improved especially when involves personal questions.
• Less risk the respondent will give biased answers or try to anticipate what the researcher wants to hear (Hawthorne effect).

Question 5

Questionnaires - Weaknesses

Answer 5

• low response rate - Only a few respondents will return it = unrepresentative sample
• Nothing the researcher can do if respondents ignore certain questions or respond incorrectly.
• The questionnaire format makes it difficult to examine complex issues and opinions.
• The researcher has to decide what is and what is not significant.
• Anonymity may encourage honesty
• Do not know if the correct respondent has answered the questions - affects the validity/representativeness of the research.
• unintentional biased questions eg. If a question has more than one meaning or
  Leading questions.
• If an option is not precisely defined, it will mean different things to different people.

Question 6

Structured Interviews

Answer 6

A structured interview is where the researcher asks questions to respondents in person.
To achieve consistent and comparable results, the same questions are asked in the same order each time – there is no flexibility to change the order of the questions.

Question 7

Structured Interviews - Strengths

Answer 7

• little chance of the questions not being understood or of the respondents answering the questions incorrectly.
• Will have your representative sample – response rate will be 100%

Question 8

Structured Interviews - Weaknesses

Answer 8

• involve assumptions about people’s behaviour.
• Can contain biased questions.
• The lack of anonymity in an interview can cause Hawthorne effects, respondents give answers, they think the researcher wants to hear.
• The researcher effect is when the relationship between researcher and respondent may bias responses:
  Aggressive interviewers may introduce bias by forcing or intimidating respondents into giving answers that they do not really believe.
  Status considerations, based on factors such as gender, age, class and ethnicity may also bias the data.

Question 9

Experiments

Answer 9

• Involve testing the relationship between different variables.
• Researcher manipulates IV to see if they change DV.
  Relationship could be:
  1. Correlation - Variables change at the same time, suggesting a relationship.
  2. Causation - when one action occurs, another always follows. Allow reseacher to predict future behaviour.

Question 10

Ways to separate correlation from causality

Answer 10

1. Test and retest a relationship. More times a test is replicated with the same result, the greater the chances that the relationship is casual.
2. Use different groups with exactly same characteristics.
3. Have a experimental group whose behaviour is manipulated
4. Have a control group whose behaviour is not manipulated.

Question 11

Lab experiments

Answer 11

• Take place in a controlled environment.
• Conditions are precisely monitored and controlled.
• Ensures no EVs affect the relationship between the DV and IVs.
• Unusual in sociology because they involve an artificially created situation so it’s unlikely the findings will apply to the real world.
• Ppts will be aware research is taking placr and this effects how they behave.
• Also raise ethical issues about consent

Question 12

Field experiments

Answer 12

• More appropriate for sociological research
• Not conducted in a closed, controlled environment.
• Try to establish correlations as it’s difficult to control all variables.
• Researchers use DVs and IVs to test a hypothesis or answer a research question.

Question 13

Case Study - Rosenthal & Jacobson (1968)

Answer 13

Aim - To test the hypothesis that teacher’s expectations influence how well their students do in school.
DV - was students level of achievement
IV - the expectations that teachers had about the ability of their students.
* R & J manipulated the IV by pretending to be psychologists who could, identify children who would display ‘dramatic intellectual growth’ based on a IQ test.
* They tested the students and then randomly classed some students as ‘later developers’.
* The researchers informed the teachers of their findings.
* They retested the students and discovered that the IQ scores of those students whose teachers believed were ‘late-developing high flyers’ had significantly improved.

Question 14

Experiments - Strengths

Answer 14

• Lab experiments are easier to replicate than field because researcher has more control.
• Standardised conditions give high reliability
• Can create highly valid statements about behaviour based on cause-and-effect relationships.
• Field experiments can be used to manipulate situations in the real world to understand reasons for everyday behaviour.

Question 15

Experiments - Limitations

Answer 15

• Difficult to control all possible influences on behaviour
• Awareness of being studied can cause the Hawthorne effect (observer effect)

Question 16

Formal content analysis

Answer 16

• This method attempts to quantify the contents of a document in an objective manner.
• Content Analysis has both quantitative and qualitative forms.
• Quantitative analysis uses statistical techniques to categorise and count the frequency of people’s behaviour using a content analysis table.

Question 17

Thematic analysis

Answer 17

• This approach looks for the motives and idelogies
• Job of the researcher is to uncover the underlying ideology.

Question 18

Textual analysis

Answer 18

• This method involves a close examination of the ‘text’ of a document to see how it encourages a particular reading and creates a particular impression.
• eg. examining if headlines are misleading to attract readers.

Question 19

Content analysis - strengths

Answer 19

• Can identify themes and patterns of behaviour that may not be immediately apparent.
• Used for ‘concept mapping’ Page (2005) tracked how media professionals portrayed global warming in order to show how far global warming was reported in terms of ‘natural’ or ‘social’ causes.
• Quantification of behaviour allows researchers to draw complex conclusions
• Use of standardised framework means data can be checked and replicated.

Question 20

Content analysis - limitations

Answer 20

• In some types, reliability may be limited because researchers must make subjective judgements about what they are counting.
• Concerns whether all observed behaviour can be neatly categorised.
• Doesn’t tell us much why audiences receive, understand, accept or ignore patterns or themes discovered by the research.

Question 21

Primary Sources - Qualitative data

Answer 21

• Unstructured interviews
• Semi-structured interviews
• Over/covert participant and non-participant observation

Question 22

Unstructured interviews

Answer 22

• Enable researchers to acquire an understanding of how people think and feel
• Respondents are encouraged to talk freely
• Free form method
  Eg. Ann Okaley’s classic study on housework. She used unstructured interviews to allow respondents to talk freely.

Question 23

Unstructured interviews - Strengths

Answer 23

• Data reflects the intrests of the respondent.
• More accurate and detailed expression of their beliefs.
• Data is highly valid
• Avoids researcher pre-judging what makes data important
• Researcher establishes a strong understanding with respondents.
• Sensitive issues can be explored.
• Can take place in more confortable environments.

Question 24

Unstructured interviews - limitations

Answer 24

• Requires skill
• Researcher must resist influencing, encouraging or interrupting.
• Researcher has little control over the direction of the conversation > may lead to irrelevant topics.
• Interviews are time-consuming
• Analysing and interpreting the data is also time-consuming
• Reliability is low due to no standardisation
• naturally biased interview effects
• Respondents may try to please the researcher by telling them what they think they want to hear.

Question 25

Semi-structured interviews

Answer 25

• Combines strengths of unstructured and structured interviews
• Researchers freely ask questions in any order.
• Allow respondent to talk at length
• Interview has a structure/schedule - an area the interviewer wants to focus on.
• No list of specific questions
• Different respondents may be asked different questions on the same topic
• Open ended questions frequently used
• Some questions created before

Question 26

Semi-structured interviews - strengths

Answer 26

• Less risk of researcher deciding what will be discussed
• possible to pick up ideas and information from respondent and can be used for further questions
• Rearcher can discover what the respondent really thinks
• Focuses on issues important to the respondent
• Increase in validity as it’s more likely the research will achieve the real aims.

Question 27

Semi-Structured interviews - limitations

Answer 27

• Requires skills(thinking quickly, asking right questions)
• More time consuming
• Data has to be analysed and interpreted
• Data may end up be irrelvant to main topic
• Respondents are relied on to remember past events can create problems
• Research doesn’t know if the respondent is telling the truth
• Memories can be selective
• Lack standardisation so not reliable

Question 28

Group interviews

Answer 28

• Respondents gathering to discuss a topic decided in advance
• groups may be selected as representative samples or a group the researcher wants to explore.

Question 29

Success of group interviews depends on…

Answer 29

1. An interview structure with clear guidelines for ppts - avoid arguments
2. Advance questions
3. Interaction within the group, Gibbs argues gives ‘unique insights into people’s shared understandings of everyday life.
4. Researcher must ensure that discussion is not dominated by one or two people

Question 30

Group interviews - Strengths

Answer 30

• Rearcher can control pace and scope of discussion
• Plan a schedule that allow a focus
• Rearcher can change the focus of the discussion at any point.
• Create a situation that reflects how people naturally share ideas

Question 31

Group interviews - limitations

Answer 31

• Rearcher has to be in high control to maintain focus which requires skill
• May be problems with representativeness
• Group interviews are at risk of ‘groupthink’ - people feel pressure so say what they think the rest of the group/researcher want to hear.
• Risk of reflecting a ‘group concensus’ rather than revealing what individuals really believe.

Question 32

Observation

Answer 32

• Observational methods are seen to be more valid because data is gathered by seeing how people behave rather than on trust that people do what they say they do.

Question 33

Overt Participant observation

Answer 33

• Researcher takes part in the behaviour being studied.
• People know they are being studied.
• Weber called this ‘Verstehen’ - to understand by experiencing.
• Mead described it as the researchers ability to ‘take part of the other’ and see things from their viewpoint (empathy).
• Researcher joins the group openly
• Usually conducts the research with the permission/cooperation of the group.

Question 34

Overt participant observation - strengths

Answer 34

• Researcher can ask questions, take notes and observe openly.
• If group involves hierarchy they can access all levels.
  eg. Vankatesh’s study of a Black American gang was only possible because the gang leader ‘sponsored’ him.
• Sponsorship makes it easier to separate roles of participant and observer - it reduces the chances of the researchers becoming so involved in a group.
• Going ‘native’ is when observers become too involved and end up becoming participants.
• Researchers build a highly detailed picture of the lives they are describing - increases validity.
• Observer’s role is clearly defined so less risk of involvement in criminal behaviour
• Researcher can pull back from behaviour without losing trust in those being studied.

Question 35

Overt participant observation - limitations

Answer 35

• Group can refuse to be observed
• Requires lots of time, effort, money
  eg. Venkatesh spent 8 years on his study
• Hawthorne effect can occur
  eg. Venkatesh witnessed a fake punishment as the higher level gang members put on a show for his benefit.
• Researcher involvement may be superficial to allow true understanding
• not participating in all behaviour may affect the extent that the researcher is truly experiencing how people normally behave.
• Risk the researcher will become to involved.
• Studies using this method are impossible to replicate.
• Difficult for researchers to accurately record behaviour
• What is recorded is selective and interpreted.

Question 36

Covert participant observation

Answer 36

• Researcher secretly joins the group
• Subjects are unaware they are being studied.
• Main aim is to experience behaviour in it’s natural setting - to watch how people normally behave.
• Researcher must balance the roles of researcher and participant without revealing their true role.

Question 37

Covert participant observation - strengths

Answer 37

• May be the only way to study groups that wouldn’t norally allow them to be studied. eg. Criminal/deviant groups, closed groups, religious groups, defensive groups
• Avoids observer effect
• Researcher gains understanding of the meanings and motives within a group and can make sense of behaviour.

Question 38

Covert participant observation - limitations

Answer 38

• Goffman’s study identified 3 major problems:
  1. Getting in - some groups may be invitation only, have entry requirements, characteristics must match those observed.
  2. Staying in - researcher may not have access to all areas. Researcher has to learn the culture and dynamic of the group quickly, be able to mix easily with strangers, create a convincing back story.
  3. Getting out - may be difficult to stop participating. People have gained trust.
• Research cannot be replicated = not reliable
• Recording data frequently is difficult

Question 39

Non-participant observation

Answer 39

• Observing from a distance
• Subjects don’t know they are being observed
• Researcher usually does not become personally involved
• Used to produce quantitative data eg. number of times people are observed carrying out a certain act.

Question 40

Non-participant observation - Strengths

Answer 40

• Easy access
• Allows research on people who may not want to be studied.
• Researcher can be objective
• Study people in their natural habitat
• Researcher gets to see everyday behaviour

Question 41

Non-participant observation - limitations

Answer 41

• Not easily replicated - affecting reliability
• Observing from a distance may produce data that fails to capture the depth of their behaviour - may lack validity
• Raises ethical issues - observing without permission (no consent)