Research methods Flashcards
(106 cards)
1
Q
Outline what a self-report is.
A
methods of gathering data where participants provide information about themselves without interference from the experimenter
2
Q
Two types of self-reports.
A
interviews and questionnaires
3
Q
Types of questions used in self-reports
A
open-ended - They do not have fixed responses, and so they allow the participant to answer however he/she wishes.
closed questions- restrict the participant to a predetermined set of responses and generate quantitative data.
4
Q
Structured vs unstructured interview.
A
Structured interview- questions are pre-determined. Interviewer cannot go off topic and ask to for the ppt to elaborate.
Unstructured interview- interviewer has some idea of the questions they will ask but can make up questions in the interview to ask participant.
5
Q
Outline what an interview is
A
A method of data collection whereby an interviewer asks questions either face to face, video call or on the phone. Verbally asking and answering questions.
6
Q
Strengths of using questionnaires.
A
- once designed, easy to administer & analyse
- large amounts of info can be collected within a short period
- ppts feel comfortable answering- confidential
7
Q
Limitations of using questionnaires.
A
- problems with wording- ppts may interpret questions in different ways.
Researcher can also infer answers incorrectly to fit their investigation (researcher bias) - can only be given to those who can read or write
- can be lead down a rabbit hole by vast answers (open ended questions)
8
Q
Strengths of using interviews.
A
- flexibility; complex issues can be explored which may not be possible via other RM’s
- tackling sensitive topics; can investigate topics that are private or personal
9
Q
Limitations of using interviews.
A
- interpreting data; qualitative data can be difficult to analyse and may be bias from interviewer. (researcher bias)
- interviewers presence may effect response from ppt- demand characteristics.
10
Q
Validity issues with self-reports.
A
- social desirability bias; ppts alter the their response to present themselves in the best way possible
- internal validity; leading or ambiguous questions can affect ppts responses
- interviewer bias; how the researchers behaviour could influence ppts responses
- content validity; do the questions actually measure what is intended to be measured
11
Q
Assessing and improving validity in self reports.
A
- using face validity; does the questionnaire look like it’s measuring the right thing.
- concurrent validity; comparing new questionnaire results with results from previous established tool on the same topic. Ppts complete both at the same time and results are compared.
12
Q
Reliability issues with self-reports.
A
- External reliability; consistency of results obtained, ppts should be able to take questionnaire again & get similar results.
- Internal reliability; extent to which questions are consistent with each other.
- Test-retest method; asking ppts to retake the test at a later date and comparing results.
13
Q
What is correlation?
A
relationship between two or more variables and how one affects the other.
14
Q
Types of correlation.
A
- Positive correlation; As one variable increases the other increases.
- Negative correlation; As one variable increases the other decreases.
15
Q
Explain how to measure the correlation.
A
Psychologists use a stats test called correlation coefficient to measure the strength. Can range between -1.0 & +1.0.
The number represents the strength of the relationship between the variables, nearer the number is to +1.0 or -1.0 the stronger the relationship.
16
Q
What is validity?
A
measure of the truth; does the test measure what it says it’s measuring. If you can generalise the results then it is valid.
17
Q
What is internal validity?
A
how much the findings of the DV actually has to do with the manipulation of IV and not other factors (like EVS)
18
Q
What is external validity?
A
weather the results can be generalised beyond the study (ecological validity and population validity)
19
Q
What is population validity?
A
can the results be generalised to the population
20
Q
What is ecological validity?
A
can the results be generalised to real life and real world. to be high then both the setting and the task which the study takes place in must have mundane realise. (similar to ‘everyday life’)
21
Q
What is face validity?
A
an independent expert looks at the measure being used and asses weather the study is measuring what it intends to (just by looking at it) (judging a book by its cover)
22
Q
What is concurrent validity?
A
comparing the new procedure with a similar procedure that has been done before, where validity has already been established. If the scores correlate as a strong positive correlation (+0.80) then the test is deemed to b e valid.
23
Q
What is temporal validity?
A
weather results from a study or a theory remain the true & valid over a period of time. (weather the results can still be generalised to a modern day society)
24
Q
Improving validity- experiments.
A
- to ensure that IV has an affect on DV and not EV’s, a control group is used.
- control of variables is needed through standardisation; in order to reduce investigator effects and demand characteristics. Single-blind and double-blind procedures can be used.
25
Improving validity- questionnaires.
- a lie scale within the results is used within the questionnaires and psychological tests. These are additional questions which assess how truthful someone is in their answers, helps control social desirability bias.
- anonymity can encourage truthfulness, increases validity.
26
Improving validity- observations.
- behaviour categories which have broad, ambiguous or overlap may reduce validity, so they must be clear and operationalised.
27
Improving validity- qualitative methods.
- interpretative validity needs to be demonstrated. the researcher has to show that their interpretation matches the ppts reality, using direct quotes on their report.
- gathering evidence from from multiple sources (family/friends, diary entries and observations) can strengthen case studies & interviews because interpretation can be validated, this process is called triangulation.
28
What is reliability?
making sure the results and the tests are consistent (if the experiment was to be repeated multiple times, would the results be the same each time)
29
Assessing reliability- Test re-test method
- re-administering the same test to the same people in the same conditions but on different occasions. (results will be the same if test is reliable)
- occasion that the tests are administered must not be too far away from one another so that answers completely change but not to close together that ptps remember the answers.
30
Assessing reliability- Inter-Observer reliability
- relevant to observational methods
-different researches must be able to interpret data in the same and come to the same conclusion in order for it to be reliable
- using behavioural categories
31
Improving reliability- Questionnaires
- if a low correlation is produced, some questions may need to be removed or rewritten
32
Improving reliability- Interviews
- unstructured interviews have low reliability
- use same interviewer where possible
33
Improving reliability- Experiments
- instructions need to be standardised
- Ev's controlled as much as possible
34
Improving reliability- Observations
- Behavioural categories need to be operationalised, must be measurable and self-evident
- using more than one observer (interobserver reliability)
35
What is a lab experiment?
- controlled conditions
- manipulates IV
- measures DV
36
What is a field experiment?
- natural conditions (ppt's in natural environment)
- manipulates the IV
- measures the DV
37
What is a natural experiment?
- natural conditions (naturally occurring, something is already happening)
- IV naturally occurring
- measures the DV
38
What is a quasi experiment?
- controlled OR natural conditions
- IV is the difference between people (individual differences)
- measures the DV
39
What is an independent variable?
- manipulated by the researcher to test effects on another variable
40
What is a dependent variable?
- measured by researchers to see how to see how it's been affected DV
41
What are extraneous variables?
- extra variables which are controlled to prevent effect on DV
42
What are cofounding variables?
- type of EV which acts as an IV; systematically affects all ppt's in the same way, altering the DV.
43
What is a pilot study?
- a study before the real one done on a smaller scale to see if the real one will run smoothly, it checks all variables and allows changes to be made.
44
Evaluate lab experiments
strengths;
-high control of EV's
-high internal validity
-high reliability
limitations;
-low external validity (ecological)
low mundane realism (artificial setting and task)
-more likely to have demand characteristics (ptps changing behaviour as they are aware there being tested)
45
Evaluate field experiments
strengths;
-researcher still has some control- manipulates the IV
-high eco validity
-less demand characteristics (only if ptp is unaware)
limitations;
-could be lots of EV's
-more likely to have ethical issues (lack of consent)
46
Evaluate natural experiments
strengths;
-high eco validity
-provides unique insight
limitations;
-lack of control over EV's
-low internal validity & reliability
- limited generalisability- may not be relevant to rest of population
47
Evaluate quasi experiments
strengths;
- compares different groups of people to test for differences
-can be carried out in controlled or natural setting
limitations;
-lack of manipulation of UV in natural setting
- lacks eco validity if in lab
- EV's as IV is ptp characteristics
- limited generalisability
- may have bias
48
What is operationalising variables?
defining specifically how they are used in a study
49
What is an independent measures design?
- choosing different people to go in each condition.
- should be done by random allocation
50
Describe how you could carry out random allocation for group designs?
- give numbers to all ptps
- put numbers in hat/random no. generator
- select numbers and allocate to conditions in random manners. (first 5 to condit A, next 5 to condit B)
51
What is an repeated measures design?
- same ptps take part in both condit's
- counter balancing used to avoid order effects.
52
What is counterbalancing in repeated measures?
- both conditions completed but in diff order
- eg. 12 ptps do condit A & then condit B
- other 12 do condit B & the condit A
- this cancels order effects out but does not rid of demand characteristics.
53
What is a matched pairs design?
- pairs of ptps matched in terms of variables/ characteristics (age, gender...)
- one of the pair is placed in one condit the other pair in the other condit.
54
Evaluate Independent measures design.
strengths;
- avoids order effects
- avoids/ reduces demand characteristics (guessing aim of study and changing behaviour)
limitations;
- requires large sample
- cannot control IV's which become EV's that affect DV. (invalidating results)
55
Evaluate Repeated measures design.
strengths;
- controls IV's
- fewer ptps required
limitations;
- strong chance of order effects (counterbalancing reduces this)
- demand characteristics
56
Evaluate matched pairs design.
strengths;
- effects of IV differences can be reduced
- avoids order effects and demand characteristics
limitations;
- exact matches are difficult/impossible to find
- requires large number of ptps.
57
What is a sampling frame?
list of potential ptps from target population, from this list a sample of ptps can be drawn
58
What is a target population?
population of people the research is aimed at. You select sample from this using a sampling technique
59
What is a population?
total group from whom your going to take your sample from
60
What is opportunity sampling?
- sample from which you have the opportunity of studying... most available (friends/family)
- NOT RANDOM
61
Evaluation of opportunity sampling.
strengths;
- easiest
- no prev prep needed (practical advantage)
limitations;
- inevitably biased because sample from small population.
- can be unrepresentative (lacking validity)
62
What is volunteer sampling?
- ptps respond to advertisement to take part in experiment (newspaper, social media etc)
- sample is whoever volunteers up to the number of ptps required.
63
Evaluation of volunteer sampling.
strengths;
- less researcher bias as ptps pick themselves
- usually quite representative
- ptps willing to take part
limitations;
- chance of volunteer bias (type of person who your going to attract, more likely to be extroverted, don't mind volunteering, an lack representativeness)
64
What is random sampling?
- gather a sampling frame, randomly select ptps and assign them to conditions (all ptps have equal opportunity of being selected)
65
Evaluate random sampling.
strengths;
- all ptps have equal chance of being chosen (unbiased), selects ptps in an objective way
limitations;
- could be unrepresentative (despite being randomly selected, it's not guaranteed)
- takes a long time (not practical)
66
What is systematic sampling?
-ptps selected on a pre-determined system
- you have a sampling frame, you then select every nth person from target population, this forms your sample.
67
Evaluation of systematic sampling.
strengths;
- unbiased as ptps selected using an objective system
limitations;
- could be unrepresentative despite random selection
- not truly random unless you select a random person to begin with.
68
What is stratified sampling?
- layers or sub group
- have your sampling frame, identify the sub groups of your population
- population on a smaller scale in proportion to actual population.
- from these populations, select these ptps randomly.
69
Evaluation of stratified sampling.
strengths;
- completely representative of target pop due to proportioning and random selection
limitations;
- very time consuming and hard to do (not practical)
- lots of info required (info about ptps to categorise them)
70
What are ethical issues in psychology?
Conflict between what researcher needs to do to conduct useful research and the ptps rights. Researcher then has to decide what the right or wrong thing to do is.
71
What is informed consent?
ptps pov- have a right to know what they are consenting to (includes what they are going to do and what will come from it)
researchers pov- giving this info can cause demand characteristics.
72
What is deception?
(lying about aim or not giving full detail)
ptps pov- prevents ptps from giving valid consent therefore violates rights and is unethical.
researchers pov- researchers may need to do this to avoid demand characteristics, may lead to mistrust of psychologists/ people of the profession.
73
What is right to withdraw?
ptps pov- can withdraw even after originally consenting, shouldn't feel pressured.
researchers pov- responsible for reminding ptps of their right to withdraw, should tell them at the start, during the study, after the study. If ptp withdraws after study their data needs to be removed.
74
What is risk of harm (physical & psychological)?
ptps pov- could be harmed physically or mentally (humiliated/distressed). Should not be exposed to harm that they wouldn't experience in their day to day life (BPS guidlines)
researchers pov- researchers responsibility to minimise harm. Not always possible to avoid harm, but in planning it must be evident that most harm minimised.
75
What is confidentiality?
ptps pov- confidentiality is a legal right, personal data, only be recorded if it is kept confidential.
researchers pov- should do all they can maintain the confidential info of their ptps, not revealing personal details that can identify ptp when published. Researchers anonymise their results, or use fake names/ initials.
76
What is privacy?
(your right to decide how much/ what info you want to share publicly)
ptps pov- what they choose to do with their personal info
researchers pov- can be difficult to maintain, when conducting social/ field experiments. BPS states that 'public behaviour can be observed without ptps knowledge.
77
What is a debrief?
- researcher should always debrief ptps after a study, especially if used deception
- debrief provides; full info about the aim
check on well-being
- give ptps opportunity to ask questions
- gives another to withdraw if they choose to.
78
What is presumptive consent?
- way to deal with deception
- this means asking another similar group if they would consent to taking part if they would consent to taking part if they knew the aim. the researcher the can assume that their own ptps would also consent.
79
What are ethical guidelines?
- researchers must follow ethical guidelines outlined in BPS code of conduct
- need to conduct a cost-benefit analysis to judge the costs of the ptp's of doing the research and the benefits of research findings for society.
80
What is overt?
ptps know they are being observed (obvious)
81
What is covert?
ptps do not know they are being observed (covered)
82
What is a participant observation?
observer is part of observed group (observing from within)
83
What is a non-participant observationm?
observer is away from group (outside observation)
84
What is a structured observation (controlled)?
- using a system to observe specific behaviour like behavioural categories or sampling procedures
85
What is an unstructured (naturalistic)?
- observe everything you see
- you are likely to collect more qualitative data.
86
What is behavioural catergories?
- observer ticks the relevant category when one of the behaviours are displayed, like a tally chart.
All behaviour categories need to be;
- operationalised; breaking the behaviour being studied into set of components.
-objective; no inferences can be made about behaviour (no question of what category behaviour belongs to)
- mutually exclusive; no overlapping, not having to mark two categories at once.
87
What is event sampling?
- observer has behavioural categories- these are the events they are sampling
- during total observation time, they record in a table each time behaviour occurs.
- data collected likely to be a tally.
88
What is time sampling?
- during total observation time, observer records what behaviours occur at pre- determined time intervals.
- do not record any behaviours that occur outside of these time intervals
89
Evaluate observations.
strengths;
- high external validity
- practical method (eg. can study children which isn't ethical too do in an experiment)
- fewer demand characteristics (covert)
- replicable (structured)
limitations;
- cause & effect; little control over EV's, cannot conclude what was manipulated in order to effect DV.
- observer bias; can be reduced by inter-observer reliability.
- replication; lack of control over variables, never can be exactly repeated
- ethics (covert & non ptp); invasion of privacy and informed consent when ptps don't know.
- observer bias; observing behaviour & interpreting it in a way to fit study.
90
What is a case study?
- in depth study of one individual or a group/ event. Uses a range of methods to collect data (interviews, experiments etc). Usually done over a long period of time (longitudinal).
91
Name examples of case studies used in psychology.
- HM; memory- types of LTM
- KF; memory- evidence for WMM
- little hanz; approaches- freud psychodynamic theory
- 44 Juvenile thieves; attachment- Bowlby
92
Evaluate case studies.
strengths;
- collects rich, in-depth data
- looks at real behaviour rather than in a controlled setting
- looks at unique cases, not unethical as not an experiment. EG. split brain patients (biopsych)
limitations;
- risk of ethical issues (lack of confidentiality, risk of harm, lack of consent)
- hard to generalise as it is usually unique. Idiographic method.
93
What is content analysis?
- converts qualitative into quantitative data
- researcher creates behavioural categories (coding) and analyses data, creates tally at the end of behaviour displayed.
94
What is thematic analysis?
- keeps data in qualitative form
- data reviewed repeatedly and identifies emerging trends (themes that keep coming up)
- researcher places behaviours into broader themes (not pre-determined behavioural categories)
95
Evaluate content analysis.
strengths;
- easily replicated using same coding system
- statistical analysis can be conducted because data is now quantitative
limitations;
- cause & effect cannot be determined (as only data)
- researcher bias may be present
96
Evaluate thematic analysis.
strengths;
- provides more in depth analysis than content analysis
- allows for categories to emerge from data rather than being pre-determined
limitations;
- can be subjective
- being so flexible in analysis may make it difficult to decide what aspect of data to focus on
97
What are levels of measurements (stats)?
- scale used to describe features/properties of data from research
98
What is nominal data?
- categories with no ordering or direction
- answer is usually yes or no
- can only get frequency (tally)
- cannot find mean, median or mode
- eg. marital status
99
What is ordinal data?
- ordered categories (rankings/order/scaling)
- eg. student grades, tallest to shortest
100
What is interval/ratio data?
- differences between measurements, no true zero
- standardized intervals (temperature, exam score)
101
What is the purpose of stats test?
- you can calculate the probability that your research is significant, therefore weather your results are due to chance
- p value= 0.05 (5% chance or less you are wrong)
- allows you to reject the null hypothesis or accept it.
102
What is the set phrase for accepting/rejecting the null?
' as the calculated value (put value here) [is/is not] significant at p ≤ 0.05, we [can reject/ must accept] the null hypothesis and conclude that there [is/is not] a difference in ............................ (whatever the variables being tested were).'
103
What is a Type I error?
false positive- you think your results are significant but they are not.
(used the wrong critical value eg. 0.1, too broad)
104
What is a Type II error?
false negative- you think your results aren't significant but they are
(too harsh of a critical value eg. 0.01)
105
What is an alternative hypothesis?
- a statement that predicts what will happen
106
What is a null hypothesis?
- prediction that states that the alternative hypothesis is wrong- statement of no effect.
