Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Research methods > Final exam > Flashcards

Final exam Flashcards

1
Q

what are the reasons to add a IV

A
  • identify boundary conditions
  • identify curvilinearity
  • test multiple treatments
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

identify boundary conditions

A
  • where does it stop mattering (stop having an effect)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

identify curvilinearity

A
  • does the effect have an optimal level (works till a certain point and then start going down)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

test multiple treatments

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

costs of adding levels of an IV

A
  • decreases power and increases the chance of type 2 errors
  • you need a larger sample size
  • need more resources
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

factorial designs

A

experiments with two or more independent variables

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

participant variables

A

a variable that is selected/ measures not manipulated
- usually a characteristic of the participant

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

reasons to add variables to an experiment

A
  • test boundary conditions
  • test a theory
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

factorial designs and boundary conditions

A

Does an IV affect different kinds of people, or people in different situations, in the same way?

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

boundary conditions

A

limits to an effect
- testing the generalizability of a casual variable
- testing moderators
- test of external validity

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

test a theory

A

does the study support your theory

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

costs to adding variables

A
  • decreases power and increases the chance of type 2 errors
  • need a large sample size
  • need more resources
  • complex
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

factorial designs

A
  • Independent group designs
  • Within-groups designs
  • Mixed designs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Between groups factorial design

A

No participant experiences more than one condition (they are in one of the four)

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

Within groups factorial designs

A

All participants experience all four conditions

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

Mixed groups factorial design

A
  • Some participants experience the 2 levels of the one variable and the other participants experience the two levels of the other variable
    - benefit less participants and time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

main effect

A

the overall effect of one independent variable on the dependent variable, averaging over the levels of the other independent variable

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

interaction

A

when the effect of one independent variable on the dependent variable depends on the level of another independent variable

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

how do you get the main effect

A
  • add the average score of the two levels of the IV and divide by 2
  • marginal mean
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

how do you get the interaction

A

Look at the difference between the levels of the independent variable for both the levels of the independent variables

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

crossover interaction

A

it depends

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

spreading interaction

A

especially

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

quasi-experiments

A

less experimental control than a true experiment
- some external factor

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

quasi-independent variable

A

Iv that researcher doesn’t have full experimental control over
- act of nature
- government policy
- participants themselves

25
quasi-experiment designs
- can't guarantee they are the same - Try to control all other variables Types - Posttest only - Pretest/ posttest
26
Non-equivalent groups posttest only design
- Can't be randomly assigned - Measure DV after
27
Non-equivalent groups pretest and posttest design
- can't be randomly assigned - Measuring DV before and after
28
quasi-experiment construct validity
Quasi independent variables tend to have good construct validity because they involve real-world manipulations/ experiences
29
quasi-experiment statistical validity
- Sample size and power - Effect size - Precision (CI)
30
quasi-experiment external validity
- Ecological validity- a study's similarity to real-world contexts - Tends to be high because quasi-experiments test real-world manipulations and experiences
31
quasi-experiment internal validity
- The degree to which a quasi-experiment supports a causal claim depends on the study design
32
trade-off for quasi-experiments
Quasi-experiments tend to give up some internal validity in exchange for some external validity
33
internal threats for quasi-experiments
- Regression threat - Attrition threat - Maturation threat - History threat - Testing threat - Instrumentation threat - Order effects - fatigue effects - carryover effects - practice effects - Selection effects - Demand characteristics - Researcher bias - Placebo threat
34
Solutions to internal validity threats for quasi-experiments
- Comparison groups - Counterbalancing - Double-blind/ masked study designs - Placebo control groups - posttest only - missing values analyses - variations of the measure - same coders and good codebook Another solution for selection effects - Waitlist control group
35
small N designs
studying only a few individuals - a lot of information from just a few cases
36
replication
- Will the same results be observed if the study is conducted again? - Crucial for statistical validity - A single study is only important if it can be replicated - Science progresses through replication of studies
37
direct replication
- repeat the original study as closely as possible
38
conceptual replication
- explore the same research question but use different procedures
39
replication-plus extension
replicate the original study and add variables to test additional question
40
replication crisis
- Accused Psychology of producing false-positives - Studies that have found effects but other researchers are unable to replicate it
41
Narrative synthesis
summarize the findings with words
42
Meta-analysis
a statistical analysis that creates a quantitative summary of a scientific literature - Examines strength of an effect across existing studies and replications
43
Systematic Reviews and Meta-analyses
- Systematically finds studies that fit eligibility criteria to answer a very specific research question - Conducted to reduce bias through explicit and systematic methods
44
why might a study not replicate?
1. problem with the replication attempt 2. problem with the original study 3. problems with the publication practices
45
underlying issue
- Researcher degrees of freedom - the flexibility researchers have in designing and conducting studies, and in analyzing data
46
Underreporting null findings and solution
- selectively reporting only variables, conditions or studies that showed statistically significant results (i.e. discarding data) - file drawer problem solution - make all study materials publicly available
47
harking and solution
- hypothesizing after results are known solution - make data publicly available
48
explanatory vs. confirmatory research
- Exploring lots of possibilities is easy - Confirming a specific hypothesis is hard - Explanatory research is not the same as harking because you have to be transparent (state that you don't have a hypothesis)
49
p-hacking and solution
fishing for an analysis that results in p < 0.5 - pre-registration
50
using small sample sizes and solutions
- more influenced by outliers - results are more imprecise and less replicable solutions - use larger sample size - conduct a power analysis to determine N
51
consequences of QRP
- Findings that do not replicate - Inflated evidence for hypothesis - smaller p-values and inflated effect sizes - Violate scientific norms - communality - disinterestedness
52
reasons QRP happen
- Not always intentional (biases creep in) - Looking for novel/ exciting findings - Can be more difficult to publish null findings
53
open science movement
The practice of sharing data and materials freely so others can verify results and collaborate and use data
54
principles of the open science movement
Transparency: make research visible Sharing: make research accessible and usable Inclusivity: involve and credit more contributors to research
55
Generalization mode
- Focus on external validity and ecological validity - generalizing from sample to population - generalizing from lab setting to real-world - Having a representative sample is crucial - Frequency claims- always in generalization mode - Association and causal claims- sometimes in generalization mode
56
Theory-testing mode
- Focus on testing theories and making association or causal claims - Prioritizes internal validity - External and ecological validity are not as important
57
Cultural Psychology
- Focuses on how cultural contexts shape the way a person thinks, feels and behaves - Work in generalization mode and challenge theory-testing mode
58
WEIRD
Western, educated, industrialized, rich and democratic

Research methods (2 decks)

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