research methods Flashcards
1
Q
x variable
A
independent
2
Q
y variable
A
dependent
3
Q
types of designs
A
between, within, mixed
4
Q
types of studies
A
experimental, non-experimental
5
Q
between-subject (+example)
A
1 person - 1 condition
6
Q
within-subject (+example)
A
1 person - multiple conditions
7
Q
mixed design 2 types
A
pretest-posttest
condition + improvement of dep. var.
8
Q
disadvantages of within-subj.
A
practice effect, fatigue, autonomic maturation, spontaneous recovery, carry-over effect (differential transfer)
9
Q
types of rendomisation of order and counterbalancing
A
intra-group and intra-subject
10
Q
types of intra-group counterbalancing:
A
complete and incomplete (Latin square)
11
Q
explain complete counterbalancing
A
all conditions in all possible orders (AB, BA)
12
Q
formula for Latin square
A
A,B,L,C,L-1,D,L-2,E,L-3,F,…
- if it´s an odd number of conditions - mirror it
13
Q
types of manipulation
A
direct (straightforward) and indirect (staged)
14
Q
five principles APA ethics code
A
- beneficence and nonmaleficence
- fidelity and responsibility
- integrity
- justice
- respect for people´s rights and dignity
15
Q
beneficence and nonmaleficence
A
risk-benefit, balance the benefits, avoid harm
16
Q
fidelity and responsibility
A
keep promises, build trust, if you say that you´ll do smth actually do it, be professional
17
Q
integrity
A
be honest, no fraud, stealing, misinterpretation, cheating
18
Q
justice
A
everyone benefits from the research, equal tretment, no discrimination, open access, peer review
19
Q
respect for people´s rights and dignity
A
privacy, confidentiality, consent, permission, self-determination
20
Q
deception
A
active misinterpretation of the nature of the study
21
Q
coercion
A
limits freedom to consent
22
Q
2 types of misrepresentation
A
fraud and plagiarism
23
Q
fraud
A
making up data
24
Q
plagiarism types
A
word-for-word, paraphrasing
25
one group posttest only design
pre-experiment: treatment gr. - ind. var. - dep. var
26
one group prettest-posttest design
pre-experiment: treat. gr. - pretest (dep.) - ind. - posttest (dep.)
27
nonequivalent control group design
- pre-experiment
| - 2 groups, only posttest
28
nonequivalent control group pretest-posttest des.
quasi experiment
treat. g. - pretest - ind. v - posttest
control g. - pretest - no treatment - posstest
(mixed design)
29
types of sampling
probability and non-probability
30
types of probability sampling
simple-random, stratified random, cluster
31
types of non-probability sampling
convinience, purposive, quota
32
sample-frame
list of all people from where we draw participants from our sample
33
strata
sub-populations
34
statistical validity
are the two means (A,B) different, can coincidence be rulled out
35
internal validity
depends if it´s a true, non or quasi exp.
36
external validity
can results be generalised; need random selection of participants and situations
37
simple-random sampling
- form a sampling frame
| - randomly select participants
38
stratified random sampling
- devide population from sample frame into strata
| - randomly select participants from each strata
39
cluster sampling
- identify clusters of individuals
| - randomly select a couple of clusters
40
convinience (haphazzard) sampling
- find participants where is convinient for you
41
purposive sampling
- representative participants who are typical of a certain population
42
quota sampling
- make stratas
| - decide afterwards who will be in each sub population
43
sampling bias
some groups are under/over represented
44
response bias
- low response rate
| - responders and nonresponders may differ
45
double-barreled question
2 questions in one
46
negative wording
are you NOT ...
47
loaded question
assumes your answer
| smth controversial
48
intra-group rendomisation
- for within-subject
- order of the conditons is chosen randomly for each participant
- each person gets each condition once
49
intra-subject rendomisation
- for within-subject
- each persin gets each condition multiple times
- condition is administered more than once
50
types of validity of a study
- statistical
- internal
- external
- construct
51
types of validity of measuring instruments
- content
- criterion
- construct
52
content validity
are the questions in the study a good reflection of all questions
53
criterion validity
can the test scores be used to predict something (a behavior)
54
construct validity of measurments
- does the test measure what we want it to measure and not something else
- prediction of relationship between variables
ex. does the intelligence test really test intelligence and not creativity or math solving
55
estimation of Cohen´s Kappa
- 100% agreement = 1
- 50% agreement = 0
- 75% agreement = 0.50
56
types of reliability
- test-retest (correlation)
- internal consistency (Cronbach´s Alpha)
- interrater (Cohen´s Kappa, Interclass corr.)
57
Test-retest
- give test multiple times to same individuals
| - correlation between scores
58
Internal consistency
- give a test once to a group
- multiple questions that measure the same concept
- Cronbach´s alpha
59
Interrater reliability
- extent to which raters agree with eachother
- Cohen´s Kappa
- interclass correlation (between 0 and 1)
60
meta-analysis
- combine findings from multiple studies to reach a conclusion
61
pros of naturalistic behavior
can observe unconscious behavior
62
cons of naturalistic behavior
- sometimes the behavior doesn´t occur often
- cannot know the motive behind the behavior
- difficult interpretation
- time consuming
- risk of reactivity
63
close ended questions pros
- limited number of answers
- easier to assign responses to values
- easier to interpret
64
question wording
- unfamiliar technical terms
- double barreled questions
- loaded question
- negative wording
65
oral interviewing pros
- suitable for open ended questions
| - clarification is possible
66
oral interviewing cons
- greater chance of non-responders
- sensitive to social desirability
- interviewer bias
- time consuming
- more costly
67
written interview pros
- less costly
- many people at the same time
- anonymous
- less sensitive to social desirability than oral interviews
68
peer review
- the study is reviewed by other scientists before being published
69
debriefing
- after the study
| - explanation of the purpose of the research
70
minimal risk
- no greater than risk in daily life
71
validity of a study
- the extent to which the conclusion is accurate
72
correlation coefficient
- between 0 and 1
| - numeral indicatior of the relatioship between variables
73
ceiling effect
- x has no effect on y bc the participants reached maximum performance level
- when the task is too easy
74
floor effect
- when the task is too hard
75
manipulation check
- for construst validity
| - attempt to directly measure whether x had the intendent effect
76
single-case experimental design (N=1)
- beh. is observed during baseline period
- treatment period is followed
- observe change in beh. from baseline to treatment
77
Reversal design
- ABA design
| - baseline - treatment - baseline
78
multiple baseline design across subjects
- beh. of several ppl is observed
| - manipulation is introduced at different times for each person
79
multiple baseline design across behavior
- one subject
| - several different beh. are observed over time
80
multiple baseline design across situation
- same beh. is observed in different environments
| - manipulation is introduced at different times
81
one group designs cons (pre-experiment)
- history
- maturation
- testing effect
- instrument decay
- regression towards the mean
82
history
any event that occurs between first and second measurment
83
testing effect
pretest changes behavior
84
regression towards the mean
- tendency for score to average out
| - high score becomes lower (vice versa)
85
interrupted time series design
tracking a long time before and after manipulation
86
cross-sectional method
- different ages/generations are studied
| - between-subj.
87
longitudinal method
- same group is observed as they age
| - within-subj.
88
quantitative research
- focus on behaviors that can be counted
| - statistical analysis of data
89
qualitative research
- focus on beh. from natural settings
- describe beh. in your own words
- based on interpretation
- state the levels
90
reliability
- are the results of my measurment repeatable
| - would we get the same results if we repeated the measurment
91
yea-saying (acquiescence response set)
agree with everything
92
nay-saying
disagree with everything
93
autonomy issues
when a person lacks the ability to make their own decisions and consent
94
matched pair designs
form pairs of participats who share the same variable (ex. age, height, personality trait)
95
Type of measures
self-report, behavioral, physiological
96
attrition/mortality
- when groups become non-equivalent due to high drop-out rates
97
Solomon 4 group design
1. experimental group - posttest
2. control group - posttest
3. prettest - experimental g. - posttest
4. prettest - control g. - posttest
98
demand characteristics
if the participants are aware of the hypothesis, they will do anything to confirm it
99
risk-benefit analysis
- stress
- physical harm
- loss of privacy and confidentiality
100
exempt review research
- research in which there is no risk
| - annonymous questionnaires, surveys, tests etc.
101
Basic VS Applied reasearch
basic - answers the questions abt nature of beh.
| applied - seeks to solve practical problems
102
Expedited review research
- minimal risk research
| - mostly biological/medical
103
Limited review research
- sensitive data is collected
| - participants need to be identified
104
Full review research
- greater than minimal risk
105
Developmental research designs
- Cross-sectional
- Longitudinal
- Sequential
106
Cohort effect
- disadvantage of cross-sectional design
| - a group of people born at the same time
107
Sequential method
- following groups of different ages for a long period of time
