Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Exp Psych > Exam 2 > Flashcards

Exam 2 Flashcards

(101 cards)

Start Studying
1
Q

experiment with one independent variable

A

IV= cause, DV= effect

ex. is there a connection btwn alcohol and aggression?

does using a cell phone behind the wheel impair driving?

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

interaction effect

A

whether the effect of the original IV depends on the level of another IV
- difference in differences

ex. hands free cell phone cause people to drive poorly
-interaction effect is whether effect of cell phone use depends on driver age

MOST IMPORTANT EFFECT

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

crossover interaction

A

when the impact of one IV reverses based on level of another IV- crosslike graph

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

spreading interaction

A

when effect of one IV on a DV is more pronounced at one level of another IV and weaker or non existent at another

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

factorial design

A

allow the testing of interactions because of the presence of two or more IVs
- can test whether IV affects diff kinds of people, or people in diff situations in the same way
- can show moderators
-can test generalizability and theories

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

main effects

A

effect of one IV on the DV ignoring all other IVs
-independent influence of factors
- differences

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

overall effects

A

main effects of all variables simultaneously
-= main effect?

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

marginal mean

A

arithmetic means for each level of an IV
- averaged over the levels of the other IV

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

independent-groups design

A

both IVs studied as independent groups

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

within groups design

A

both IVs manipulated within groups
- participants participate in all four combinations of design

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

mixed factorial design

A

one IV manipulated as independent groups and one IV manipulated as within groups

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

3 way design

A

a three way interaction is present whenever a two way interaction exists for one level of a 3rd IV but not for the other
-3 main effects
- 3 two way interactions
- 1 three way interaction

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

quasi experiment

A

when there is no control over IVs
- they have a DV and IV but lack full experimenter control over IV

can be within groups design thru nonequivalent control group or between groups design through interrupted time series deisgn

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

nonequivalent control group design post test only

A

participants in one group are exposed to a treatment and nonequivalent group is not exposed to treatment

a between subjects design w/ no random assignment, posttest only

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

interrupted time series design

A

tracking a long term period before and after point of intervention to assess effects

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

nonequivalent groups interrupted series design

A

taking set of measurements at intervals over a period of time both before and after an intervention of interest in two or more nonequivalent groups

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

quasi experiment: controlling for selection effects

A

wait-list design- participants plan to recieve treatment but are randomly assigned to do so at different times

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

quasi experiment: controlling for maturation

A

use of a comparison group

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

quasi experiment: controlling for history

A

not avoided but likelihood is assessed

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

quasi experiment: controlling for regression

A

random assignment eliminates this, but quasi does not use random assignment

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

quasi expeimnet: controlling for attrition

A

only including those who complete the study

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

quasi experiment: controlling for testing effect

A

use of comparison group and results of study

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

quasi experiment: controlling for instrumentation effect

A

use of comparison group and results of study

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

quasi experiment: controlling for observer bias

A

participants asked to self report

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
quasi experiment: controlling for demand characteristics
not avoided
26
quasi experiment: controlling for placebo effect
avoided if comparison group is placebo group
27
trade offs of quasi experiment design
might not ensure complete confidence internal validity of have strong ability to draw causal conclusions, ubt they provide opportunity to study in real world situations, take advantage of historical events, or conduct ethical investigation of intervention
28
interrogate quasi experiment by referencing construct, external and statistical validity
1. how well did the study manipulate/measure variables 2. can this study be applicable to another real life situation 3. how large were the group differences, and were the results significant
29
small n-designs
when researchers conduct experiments with few participants ex. case studies pros - usually have experimenter control cons - internal validity - external validity concerns
30
3 differences between small n and large n experiments
1. each participant is treated as a separate experiments 2. individuals data are presented 3. researchers decide if results are significant by repeating experiment on new participant
31
stable baseline design
small n design where researcher observes behavior for an extended baseline period before beginning a treatment or intervention - high internal validity, eliminates confounds
32
multiple baseline design
small n design where researcher staggers introduction of intervention across individuals, times, situations - behavior measured repeatedly during baseline and treatment period - rules out alternative explanations -two or more participants exposed to same treatment
33
reversal design
small n design where reserachers observe a problem behavior both with and without treatment but take treatment away for a period of time to see whether problem behavior returns- treatment is then added back to see if problem improves
34
evaluate four validities in small n designs
internal: can be very high if study is carefully designed external: can be problematic depending on goals of study- if goals are not to generalize to other populations its not an issue construct: can also be very high if definitions and observations are precise statistical: not always relevant or small n- usually use non parametric stats to evaluate
35
trade offs of small n design
limited generalizability but can be made to make causal claims
36
survey questions; open ended
allows participants to answer any way
37
forced choice question
only two options
38
rating scale question
based on what participants feel toward subject - rewuire selection of a numerical valure on a predetermined scale
39
population
entire set of people or things in which you are interested
40
sample
smaller set of people or things taken from a population
41
population of interest
when researchers refer to it, they are not talking about an entire population, but the population to which they want their research findings to generalize
42
biased sampling
1. researchers might study only those they can contact directly 2. researchers might only study those who volunteer
43
random sampling/probability sampling
every member of the population has the same chance of getting selected
44
simple random sampling
influences external validity randomizer.org
45
systematic random sampling
sampling every 2nd or 3rd person
46
cluster sampling
defining a population, cluster the population, randomly select a cluster and collect data from that cluster
47
stratified sampling
dividing population into homogenous sub population, randomly select individuals from each category, proportionate to their membership in the population
48
oversampling
intentionally overrepresenting one or more groups - variation of stratified
49
purposive sampling
purposefully sampling certain people ex. want people with brown hair, seek brunettes
50
convenience sampling
getting people who are close/convenient to participate
51
snowball sampling
one person tells their friends to do the study and so on
52
quota sampling
get a certain number of people to fulfill each group
53
representative samples for frequency claims
increases likelihood that the samples frquencies and proportions will accurately reflect those of the larger population
54
unrepresentative samples for frequency claims
55
external validity and sample size
sampling technique is more important than size, and external validity comes from the sampling method not the sampling size - the bigger sample is not always better
56
bivariate correlation
whether a linear relationship (positive, negative, zero) exists between two continuous variables - measures two variables in the same group of ppl
57
association claim and correlational study
correlational studies support association claims -correlational research involves measuring both variables
58
construct validity of association claims
-how well was each variable measured -does the measure have good reliability -is it measuring what its intended to measure -what is the evidence for face, concurrent disciminant, convergent validity
59
statistical validity of association claims
- strength and precision(estimates on small samples are less stable) of estimate? (effect size- large is usually best; small effect can compound over many observations making it important) - has it been replicated? - are there outliers? (most problematic when there are extreme scores on both variables) - restriction of range? (when there is not a full range of scores on one variable) - is a zero association curvilinear?
60
curvilinear association
correlation coefficient is zero or close to zero and the relationship between two variables isnt a straight line
61
moderator
when the relationship between two variables changes depending on the level of another variable feature of experimental situation that might change the effect size ex. while social media use can predict levels of loneliness, the relationship mey be stronger for adolescents than for older adults
62
association vs causal claim
causation requires an association between variables, but association does not mean causation - causal claims (internal validity, temporal precedence, covariance)
63
cross sectional correlation
test to see if two variables measured at the same time are correlated - different participant same variable - does not establish temporal precedence
64
autocorrelation
test to see if corelation of each variable with itself across time - participant, same variable, same time
65
cross lag correlation
tests to see if earlier measure of one variable is associated with later measure of the other variable - different participants, same variable, different time - establishes temporal precedence
66
multiple regression analysis
statistical technique that rules out third variables -allow researchers to control for another variable
67
criterion variables
the variable being predicted - dependent variable
68
mediator
a variable that helps explain the relationship between two other variables - a study does not need to be correlational to have a mediator - tested via multiple regression
69
pattern and parsimony
researchers can investigate causality by using a variety of correlational studies that point in a causal direction parsimony- degree that a theory provides the simplest explanation of a phenomenon
70
mediation hypotheses
why are these two variables linked - why or in what way are two variables linked
71
moderator hypotheses
are these two variables linked in the same way for everyone, or in every situation - for whom is the association strongest
72
direct replication study
researchers repeat an original study as closely as they can to see whether the effect is the same in the newly collected data
73
replication
part of interrogating statistical validity - gives study credibility
74
conceptual replication studies
researchers explore the same research question but use different procedures - the conceptual variable are the same, but the operationalization may be different
75
replication plus extension study
researchers replicate their original experiment and add variables to test additional questions
76
one study, many labs
multiple lab groups work together to replicate a study
77
many labs, many studies
many lab groups try and replicate many different studies
78
meta analysis
when researchers collect all the studies on a topic and consider them together - uses quantitative techniques to create a mathematical summary of the literature
79
file drawer problem
issue with meta analysis where it might overestimate the true effect size because null effects havent been included in the analysis
80
questionable research practices
-underreporting null findings - harking - p-hacking- removing different outliers
81
transparent research practices
open data and open materials- sharing data and materials so others can use them - preregistration- publish hypotheses and methods before collecting data
82
external validity; other participants and other settings
the more settings and more participants, the greater the generalizability meaning greater external validity other settings - ecological validity: an aspect of external validity in which the focus is on whether a lab study generalizes to real world setting
83
generalization mode
when researchers want to generalize the findings from the sample in the study to a larger population of interest- primary concern is external validity - important for probability samples - applied research - frequency claims - associatio and causal claims can also be generalization mode
84
theory-testing mode
looking as association or causal claims to investigate whether there is support for a particular theory ex. contact comfort theory
85
generalization mode and the real world
external validity is important in generalization mode so having a representative sample is important - ecological validity is also important when considering generalizing to nonlab settings
86
theory testing mode and the real world
external validity and real world applicability are not as important
87
observer bias
when the observer sees what they expect to see
88
observer effects
when participants confirm observers expectations
89
preventing observer bias and effects
researchers train observers - codebooks for ratings -use multiple observers for interrater reliability reactivity: - blend in - wait it out - measure the behavior's results
90
masked research designs
observers can give unintentional cues that influence behavior - blind - observers are unaware of the purpose of the study and the conditions the participants are assigned
91
semantic differential
similar to likery but is not about agreement - use an adjective to reflect attitude ex. easy/difficult, fast/slow
92
ideal question in a survey
1. measures underlying concept it is intended to tap (construct validity) 2. it doesnt measure other concepts 3. it means the same thing to all respondents
93
question wording matters
-appropriate level of difficult -ask same question from different perspectives - offer clear alternatives - are not leading - are short and simple - dont show bias
94
double barrelled question
asking two questions at once
95
negative wording
ex. perople who do not drive with a suspended license should never be punished confusing
96
survey question order
earlier items can influence the way respondents interpret later items - ask sensitive questions near the end - randomize order of similar questions - make different version of the survey
97
response set
participants adapt a consistent way of answering all questions
98
yea saying
agreeing with all questions solution: include reverse worded items
99
nay saying
disagreeing with all questions solution: include reverse worded items
100
fence sitting
choosing all neutral solution: remove neutral choice
101
trying to look good
participants give answers that make them look better than they really are, decreasing construct validity solution: ensure participants know survey is anonymous, include items that identify socially desirable responders with target items

Exp Psych (8 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions