A2 Research Methods

?
  • Created by: DonaJ2002
  • Created on: 14-01-20 10:34
Describe and explain case studies
Longitudinal(long period), detailed + depth, that involve gathering data the subject+their relatives. They analyse unusual individuals/events(rare disorder) or common cases.Interviews/obs/questionnaire=qualitative history/assess (e.g.IQ)=quantitative
1 of 43
Define and describe content analysis
Type of observational research where ppts. are indirectly studied via spoken interactions, written communication or media(books/TV)
2 of 43
Name and describe the first stage of content analysis
Coding large data into meaningful categories by counting number of times word(including synonyms)/phrase appears in text= quantitative data
3 of 43
TALK THEMANTIC ANALYSIS WITH MRS VERE!!!!!!!!!!!!!!!!!
.
4 of 43
Define reliability and give an example of how a data can be reliable
Consistency measure= if certain measurement is repeated and same result obtained
5 of 43
What are the 2 ways of assessing reliability?
1. Test-retest 2. Inter-rater observer
6 of 43
Explain what test-retest is and what indicates reliability in the test/task
Same person is tested twice by giving them the same test/task on multiple occasions- if the test/task is reliable, then results would be same/similar each time it's done
7 of 43
Explain inter-rater observer and state what indicates reliability
Multiple observers compare their obs. data by doing pilot study, to check observers are applying behavioural categories in the same way= observers should watch same event; independently record their data- if data similar=reliable
8 of 43
In test-retest and inter-rater observation what is done to the scores?
2 sets of scores are correlated
9 of 43
What should the correlation co-efficient be to indicate reliability?
Correlation coefficient should exceed +0.80= reliable
10 of 43
What are the 4 ways to improve reliability?
1. Rewrite/deselect some questionnaire items 2. Improved training for interviewer 3. Experiments use standardised procedures 4. Operationalise behavioural categories for observations
11 of 43
Explain why and how do you rewrite questionnaires that produce low test-retest reliability, to improve reliability?
Replace open questions that causes misunderstanding/misleading with closed questions ALSO, offer fixed choice alternatives that is less ambiguous
12 of 43
Explain what you should train interviewers to do, to improve reliability and state an alternative to training all interviewers
Training all interviewers to avoid misleading and ambiguous questions; if training is too expensive and not able then, use the same interviewer for all ppts.
13 of 43
Why are lab experiments reliable?
Strict control of variables, the procedure, conditions and standardised instructions
14 of 43
What should the observing categories be to improve reliability?
1. Non-overlapping 2. Measurable (e.g. pushing is less open to interpretation than aggression)
15 of 43
If observations overlap or are absent, explain what should be done and state the consequence of this
Different observers should use own judgement in deciding what to record and where; may end up with inconsistent results or if results are similar then it's reliable
16 of 43
Define validity
Has the researcher tested what they intended to do and is the result genuine and represent the real world
17 of 43
When can data be reliable but not valid- use IQ?
IQ test that produce same result every time when the same people are tested= reliable but it doesn't measure what it's supposed to measure= not valid
18 of 43
Explain what ecological validity is and how can you remember what this means?
Do findings generalise to other settings as well as the one in the study- including if the task is done in everyday life setting other than the lab? [REMEMBER: ECOLOGY relates to ENVIRONMENT]
19 of 43
Explain what temporal validity is and use Asch's study as an example. How can you remember temporal validity's meaning?
Do findings stay consistent over time? (e.g. Asch's study lacks temporal validity as it was conducted in conformist era in American history) [REMEMBER: Temporal and Time has a T!!!!!]
20 of 43
What are the 2 ways of assessing validity?
1. Face validity 2. Concurrent validity
21 of 43
Define face validity
Does the test measure what it appears to have measured? [REMEMBER: FACE= APPEARANCE and FACE VALIDITY= does it measure what it APPEARS to measure?]
22 of 43
How is face validity done?
Keep eye on measuring apparatus or ask expert to check it
23 of 43
Define concurrent validity and explain what indicates validity
Are the findings from your study similar to established studies and if they exceed +0.80 correlation coefficient= valid!!!!!!
24 of 43
List the ways you can improve validity
1.Control groups-experiments 2.Standardised procedures- experiments 3.Lie scale- questionnaire 4.Tell respondents all data is confidential 5.Defined, operationalised, non-ambiguous+ non-overlapping categories 6.Interpretive validity+triangulation
25 of 43
How does control group and standardised procedures improve validity in experiments?
1. Control groups= ensure effect on DV was due to IV. 2. Standardised procedures reduce impact of demand characteristics and investigator effects
26 of 43
How does lie scale and assuring respondents that all data is confidential, improve validity?
1. Lie scale reduce social desirability bias effects. 2. Reassuring respondents that data is confidential= reduce chances of them lying
27 of 43
How can interpretive validity and triangulation improve validity and for what type of research?
Interpretive validity via coherence(consistency) of reporting+inclusion(including) of direct quotes from ppts. Triangulation involves using number of different sources as evidence(interview data/diaries) to improve validity of qualitative research
28 of 43
Why are statistical tests used?
Statistical tests in psychology determine if there's a difference or association(correlation) in a research is statistically significant/ if it happened by chance
29 of 43
How do you choose a statistical test?
1.Does the study look for difference or association 2.Experimental design related(matched pairs/ repeated measures) or unrelated (independent groups) 3.Is data nominal/ordinal/interval 4.Imagine stats test table, use these info to decide which test
30 of 43
Draw the table of stats test and state what MNEMONIC to use to remember it!!!
.
31 of 43
Define nominal, ordinal and interval data
Nominal= Categories-each item appear in one category. Ordinal= data placed in ordered scale. Interval= based on numerical scales that include precisely defined and equal units (e.g. counting no.of obs. from tally chart/ time/temperature/length)
32 of 43
Explain why interval data is better than ordinal data
Ordinal lacks precision as it's based on opinion than unbiased measures. Interval data is better as more detail is preserved as scores not converted to ranks
33 of 43
Define null hypothesis
States there's no difference/correlation between the conditions
34 of 43
Explain when the null hypothesis is accepted
If stats test is not significant= accept null hypothesis and reject experimental hypothesis
35 of 43
State how to identify there's NO significance so that you can accept the null hypothesis, for each stats test
.
36 of 43
How can you remember which stats test shows that if the calculated value is equal to/more than shows significance=reject null hypothesis?
Rule of R= Stats tests with the letter ''r'' in them shows that calculated value has to be equal to/moRe than critical table value to be significant (=reject null hypothesis)
37 of 43
Define probability
Measure of chance a certain even will occur: 0= impossible and 1= certain
38 of 43
There are no statistical certainties in psychology so, what are used instead and define them
No statistical certainties in psychology so, we use significance level= point we accept/reject null hypothesis
39 of 43
What is the usual significance level we use in psychology
0.05 (5%)= probability of observed effect(result) occurred by chance is equal to/less than 5%
40 of 43
Explain type I error
Null hypothesis is rejected and experimental hypothesis is accepted= optimistic error/false positive as significant difference/ association is found when one doesn't exist
41 of 43
Explain type II error
Null hypothesis accepted but in reality the experimental hypothesis is true= pessimistic error/ false negative
42 of 43
Explain when both Type I and Type II errors are likely to occur?
Type I occur if significance level is too high (0.1= 10%) because significant values are not missed Type II occur if significance level is too low (0.01= 1%) because significant values are missed
43 of 43

Other cards in this set

Card 2

Front

Define and describe content analysis

Back

Type of observational research where ppts. are indirectly studied via spoken interactions, written communication or media(books/TV)

Card 3

Front

Name and describe the first stage of content analysis

Back

Preview of the front of card 3

Card 4

Front

TALK THEMANTIC ANALYSIS WITH MRS VERE!!!!!!!!!!!!!!!!!

Back

Preview of the front of card 4

Card 5

Front

Define reliability and give an example of how a data can be reliable

Back

Preview of the front of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »