Research Methods ✅ Flashcards

(121 cards)

1
Q

Experimental hypothesis

A

A statement of what is expected
Predicts difference or correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Null hypothesis

A

A statement of no difference
Any difference or correlation is due to chance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Directional hypothesis

A

Direction is predicted

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Non directional hypothesis

A

No direction predicted

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Extraneous variables

A

Could affect outcome if not controlled eg weather

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Cofounding variables

A

Extraneous variables if not controlled

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Operational

A

Clear
Specific
Measurable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Participant variables

A

Mood
Age
Gender

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Situational variables

A

Weather
Noise
Light

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Labatory experiment

A

Takes place in highly controlled environments
Doesn’t have to be a lab
Researcher controls IV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Field experiment

A

Takes place in a natural everyday setting
Researcher manipulates IV and records effect on DV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Natural experiments

A

Researcher takes advantages of pre existing IV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Quasi experiment

A

IV based on existing difference between people (age, gender)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Strengths and limitations of lab experiments

A

Strengths:
High internal validity, reliable

Limitations: lack generalizability, low external validity, demand characteristics, lack mundane realism

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Strengths and limitations of field experiments

A

Strengths:
High mundane realism, high external validity

Limitations:
Invasion of privacy, ethical issues

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Strengths and limitations of natural experiments

A

Strength
High external validity, no ethical issue

Limitations
Pps must be randomly allocated, researcher won’t know effect (DV)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Strengths and limitations of quasi experiments

A

Strengths
High internal validity
Reliable

Limitations
Can’t randomly allocate
Confounding variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Investigator effect

A

Any effect of investigator’s behavior has on research outcome

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Randomization

A

Use of chance in order to control effects of bias when designing materials and deciding order of conditions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Standardization

A

Using same procedures and instructions for all pps

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Experimental design

A

How we divide our participants

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Independent measure designs

A

Participants only take part in 1/2 conditions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Repeated measure designs

A

Pps take part in both conditions of the experiment together

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Matched pairs design

A

Pps are matched in each condition for characteristics that may have an effect on their performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Strengths of limitations of independent measure design
Strength No order effects, pps are less likely to guess aim Limitations Individual differences
26
Strengths of limitations of repeated measure design
Strength Pp variables are controlled, less pps needed Limitation More order effects, may improve on 2nd task
27
What are order effects
Boredom Fatigue Practice
28
What are demand characteristics
Guessing the aim
29
Strengths of limitations of matched pairs design
Strength No order effects or demand characteristics Limitations Time consuming, expensive, never matched exactly, pre test required
30
What is counterbalancing
An attempt to control the effects of order in a repeated measure designs, half participants experience conditions in one order, vice versa
31
Sampling
How to select pps for a study
32
Target population
A group of people drawn from population to represent who you would want to do the study on
33
Opportunity sampling
Involves asking a group of people who fit the criteria of your study if they are available
34
Random sampling
Everybody in target population has an equal chance of being chosen
35
Stratified sampling
Classifying population into categories then sample is proportional within population
36
Self selected sampling
Participants volunteer when u ask or in response to an advert
37
Systematic sampling
A sampling frame is used, a list of people in the target population that has been organized and then picked by a sampling system
38
Strengths and limitations of random sampling
Strength Free from researcher bias Limitations Difficult and time consuming, target population is hard to obtain May be unrepresentative Some may refuse
39
Strengths and limitations of systematic sampling
Strength Avoids researcher bias Fairly representative Limitations Time consuming
40
Strengths and limitations of stratified sampling
Strength No researcher bias Representative Generalise findings Limitations Doesn’t reflect ALL people
41
Strengths and limitations of opportunity sampling
Strength Convinient Limitations Unrepresentative Researcher bias
42
Strengths and limitations of volunteer sampling
Strength Minimal input from researcher Limitations Volunteer bias
43
Informed consent
Participant must be aware and happy to take part in
44
deception
No infomed consent, lying, not telling pps abt expected results and how they’ll be used
45
Debriefing
Should explain the study, expected results, pps results, ask if they want to withdraw results or feel distress
46
Competence of researcher
Understand implications Knowing ethical guidelines Getting advice when not confident Qualified Safe How and where to store data
47
Right to withdraw
Pps should be asked beginning, end and during if they want to withdraw
48
Confidentiality and anonymity
Data shouldn’t be discussed No names in published data Confidentiality- can be traced back to names if required as researcher knows Anonymity- can’t be traced back, no names given
49
Protection of participants
Pps should not suffer physical or psychological harm Right to withdraw No greater risk from harm than before Shouldn’t be made to feel embarrassed or inadequate
50
Alternate ways of consent to deal with deception and informed consent
Presumptive Prior general Retrospective
51
Presumptive consent
Ask a similar group if study is okay and if they say yes, presume it’s okay
52
Prior general consent
They give their consent to take part in several studies, including the one you want to conduct
53
Retrospective consent
Asked for consent after the study
54
Controls for individual differences
Sample- large and random Design- use repeated measures or matched pairs
55
Controls of situational variables
Standardise - keep everything the same for each pp Counterbalance
56
Controls for demand characteristics
Deception- lie about aim, distractor questions Single blind- only experimenter knows which condition they are in
57
Experimenter variables
Tone of voice Body language Facial expressions Gestures
58
Controls of experimenter variables
Double blind- third party involved as neither experimenter or pp knows what condition they are in Inter- rater reliability- independent raters rate the same behaviors as researcher does to check reliability
59
What is the simplest data form
Nominal- separated into categories (Tall and short)
60
Middle level of data
Ordinal (ordered subjectively) Tallest to shortest
61
Highest level of data
Interval Publicly recognised numerical scale Calculates mean and standard deviation (Actual measurements of height)
62
Covert observation
Behaviour is watched and recorded without knowledge or consent
63
Overt observations
Behaviour is watched and recorded with knowledge
64
Naturalistic observations
Take place in a natural situations
65
Controlled observations
Can take place where some variables are controlled and manipulated by experimenter
66
Structured observations
Determine the behaviours to be observed Determine the sampling to be used
67
Time sampling
Observations may be made at regular time intervals and coded
68
Event sampling
Keep a tally chart of each time a behaviour occurs
69
Unstructured observations
Observer records everything that happens
70
What are the 4 formats of questions for a questionnaire
Likert type scale Rating scale Fixed choice option Open question
71
How would you go about designing a survey
- questions must be worded so all participants can understand them, only valid if everyone gets it -pilot studies should be carried out to decide what needs changing -consider the sample
72
Social desirability bias
Unlikely to give true response to personal questions, but prefer to give a more socially acceptable answer
73
Aquiescence bias
Likely to agree with something regardless of their true feelings
74
Coefficient of correlation
Mathematical value instead of “it looks positive”
75
the coefficient of correlation numbers
0.00 No relationship +1.00 positive -1.00 negative
76
Advantages of correlations
-may indicate connection where experimental evidence is hard to obtain -good start point for future research -no manipulation of variables -high ecological validity
77
Disadvantages of correlation
Do not prove casual relationship Could be 3rd variable
78
Single blind review
Where reviewers know the identity of the author but authors do not know the identity of reviewers
79
Double blind review
The identity of both the author and reviewer is kept hidden
80
Open review
Author and reviewer know each others identity
81
Advantages of single blind reviews
Allows impartial reviewing, scrutiny and decisions Prevents any influences from the author towards the reviewing process
82
Disadvantages of single blind review
May cause unnecessary critical and harsh judgement on the authors work Concern of the author over the reviewer publishing the work earlier by delaying process
83
Double blind review advantages
Prevents reviewer bias Value for the content and work of author rather than reputation
84
Disadvantages of double blind review
Chances of identifying author through report
85
Advantages of open review
Encourages more open and honest reviewing Prevents plagiarism Prevents harsh and malicious comments
86
Disadvantages of open review
May discourage honest reviewing due to fear of arguments that can affect reviewers reputation
87
What is the 3 D decision
How to decide a statistical test
88
What is the acronym for the 3 d decision
Carrots Should Come Mashed With Swede Under Roast Potatoes
89
What does the 3 d decision acrynonm stand for
Chi squared Sign test Chi squared Man Whitney Wilcoxon Spearman’s rho Unrelated t test Related t test Pearsons r
90
How to do the sign test
N= total number of people with in/decrease S= number of most +/- Step 1: Decide if it’s one or two tailed test Find 0.05 column under that tail test Step2: Trace your finger down the column for 0.05 until you come to whatever you N is Step3: you will then find critical value of x tailed test where n=x at 0.05 level of significance is x Step 4: Compare calculated value with critical value. For the difference to be significant the calculated value must be equal to or less than the critical value
91
Type 1 error
Significance level set too high Accept alternative hypothesis where it should have been recjected
92
Type 2 error
Significance level set too low Reject alternative hypothesis where it should have been accepted
93
Reliability
The extent to which a test produces consistent findings every time it is done
94
how do we assess reliability with the test-retest method
Test-retest method- do the same test with the same pps again at a different time. Sufficient time between tests so pps don’t remember answers
95
How do we assess reliability with Inter rater reliability
When more than one person is assessing behaviours and believe there is consistency on what was seen. May use categories
96
Internal reliability
The extent of which a measure is consistent within itself: different parts of the same test should give consistent results Eg. Attitude scales or personality tests
97
How can you assess internal reliability using split half method
Split the test in 2, and give it to the same pp, the 2 halves should have similar results
98
How can we improve reliability in questionnaires
Test retest then compare two sets of data Correlation needs to be +0.8 ——- questions may needs to be rewritten to reduce ambiguousness of interpretation
99
How to improve reliability in interviews
Use same interviewer each time Avoid leading or ambiguous questions Structured interviewed
100
How to improve reliability in experiments
Precise replication
101
How to improve reliability in observations
Operationalising behavioural categories Categories should not overlap eg hugging and cuddling
102
Validity
True to life
103
Face validity
Measure is scrutinised to determine weather it seems to measure what it is supposed to Eg. Does an IQ test look as though it is measuring IQ
104
Internal validity
Results must be due to manipulation of IV and not another factor Demand characteristics can threaten internal validity
105
Construct validity (internal)
When the measure successfully measures the concept it is supposed to measure
106
Concurrent validity (internal)
When the results of the study show a very similar result to what was studied in another recognised test
107
External validity
Related to factors outside the study such as generalising to other settings, other populations and other eras
108
Ecological validity (external)
If can be generalised to real life situations and settings
109
Temporal validity (external)
Concerns whether findings from research stand the test of time- are they still relevant 20 years later?
110
How can validity be improved in experiments
Control group to see changes are due to IV manipulation Ensure procedures are standardised to minimise pp reactivity and investigator effects Use double/single blind procedures- reduces demand characteristics, bias and investigator effects Use double
111
How can validity be improved in questionairres
Incorporation of a lie scale- ensures reliability of answers and control for effects of social desirability bias Ensure pps are advised their data will remain anonymous
112
How can validity be improved in observations
Covert observations Behavioural categories are not too ambiguous, broad or overlapping
113
Paradigm
Shared assumptions and agreed methods within a scientific discipline Eg evolution
114
Paradigm shift
A change in the dominant theory within a scientific discipline
115
Objectivity
All sources of personal bias are minimised so as not to distort the research process
116
Empirical
Gathering of evidence through direct observation and experience
117
Replicability
Repeating scientific procedures to test validity of findings
118
Falsifiability
We can only demonstrate the truth of a scientific principle by demonstrating is it untrue
119
Meta analysis
Research abt research
120
Strength and limitation of primary data
Strength- authentic and direct, specific targets Limitation- time consuming in planing preparing and carrying out research
121
Strength and limitation of secondary data
Strength- inexpensive, minimal effort Limitation- outdated, varies in quality and accuracy