Chapter 2: Foundations of research

Question 1

Variables vs constants

Answer 1

Varaiables: something that varies, meaning that it has at least two levels or values
If all participants are the same (fa. male), we keep ‘gender’ consistent
We use constants to check the effect on variables

Question 2

Independent vs dependent variables

Answer 2

Independent: the kind of variable researchers manipulate
Dependent: depends on the values of the independent variable and shows it’s effect

Question 3

Measured vs manipulated variables

Answer 3

Example: if we observe if people choose to have treatment for a disease or not (without us doing anything), we measure/observe the influence of nature, we manipulate it and see what changes

Question 4

Conceptual vs operational variables

Answer 4

Conceptual: abstract concepts
Operational: choices that you make as a researcher to make the abstract thing researchable or able to manipulate

Question 5

Studies make three types of claims

Answer 5

1. Frequency claims
2. Association claims
3. Causal claims

Question 6

Frequency claims

Answer 6

Say something about a frequency or a prevalence
Also associated with research called ‘descriptive research’
Each claim is about one variable that is measured (however, one study can make multiple frequency claims about multiple variables)

Question 7

Association claims

Answer 7

Talk about relationships between two or more variables
Also called ‘correlational research’ (doesn’t mean we always use correlations to explain associations)
Correlations can be used for causal claims, but not all correlations are causations!
Types of associations: positive vs negative, linear vs curvilinear, absent (null effect)

Question 8

Causal claims

Answer 8

Also talk about relationships between two variables, but here it’s a causality: one variable causes something in the other variable
Also called ‘experimental research’
We use stronger verbs for causations: causes, increases, decreases, leads to, changes

Question 9

The four big validities

Answer 9

1. Construct validity
2. External validity
3. Statistical validity
4. Internal validity
   → Depending on the type of claim being tested, some types are more/less important

Question 10

Construct validity

Answer 10

How well the variables in a study are measured or manipulated
The extent to which the operational variables in a study are a good approximation of the conceptual variables: do we have evidence that we are measuring what we want to measure?

Question 11

External validity

Answer 11

The extent to which the results of a study generalize to some larger population
We measure something in the lab and look at the possible generalization and to what extent we think we’ll find a similar situation in real life

Question 12

Statistical validity

Answer 12

How well the numbers support the claim: how strong the effect is and the precision of the estimate
Also takes into account whether the study has been replicated: if we state that there is a relationship between two variables, do we really see this relationship in the numbers and data

Question 13

Internal validity

Answer 13

In a relationship where one variable (A) and another (B), the extent to which A, rather than some other variable (C) is responsible for changes in B

Question 14

Four types of validity and frequency claims

Answer 14

Construct: how well has the researcher measured the variable in question?
Statistical: we get an estimate, but what is the confidence interval of the estimate? are there other estimates of the same percentages?
Internal: not relevant because frequency claims are not about causality
External: to what populations, settings and times can we generalize this estimate? how representative is the sample? was it a random sample?

Question 15

Four types of validity and association claims

Answer 15

Construct: how well has the researched measured each of the variables? reliable?
Statistical: what is the estimated effect size? how precise is the estimate? what do estimates from other studies say?
Internal: not relevant because association claims are not about validity (! no causal claims based on associations)
External: to what populations, settings and times can we generalize this estimate? how representative is the sample? how did we draw it from the population?

Question 16

Four types of validity and causal claims

Answer 16

Construct: how well has the researcher measured or manipulated the variables?
Statistical: what’s the estimated effect size? how precise is the estimate? what do estimates from other studies say?
Internal: was the study an experiment? does the study achieve temporal precedence? does the study control for alternative explanations by randomly assigning participants? does the study avoid internal validity threats?
External: to what populations, settings and times can we generalize this estimate? how representative is the sample? how representative are the manipulations and measures?

Question 17

Three criteria you need to meet to be able to say there is a causal relationship

Answer 17

1. Covariance: show that there is a relationship between the variables (before saying it’s a causal one)
2. Temporal precedence: cause needs to come first in time and the effect needs to come afterwards and so further in time
3. No alternative explanations: differences between groups are only due to differences in independent variable

Question 18

Spurious associations

Answer 18

Some variables may correlate, even though there is no causal relationship whatsoesver
We are often tempted to believe that correlations are a sign of causality

Question 19

Four causal models that can explain positive correlations (that are not causations)

Answer 19

1. Directionality problem: you don’t know whether variable A influences variable B, or the other way around
2. Feedback relations: the effect can go in both directions, variable A improves variable B and variable B improves variable A
3. Confouding: a third variable C that explains the relationship between variables A and B
4. Selection bias: there is a relationship because of variable C

Question 20

The Belmont report: three principles to use in experiments

Answer 20

1. Respect for persons: informed consent, protect vulnerable groups
2. Beneficence: look at the balance between risks and benefits for participants/society
3. Justice: fair balance between people participating in study and people benefitting from study

Question 21

APA guidelines: five general principles that apply to psychologists

Answer 21

1. Beneficence and nonmalifience
2. Fidelity and responsibility
3. Integrity
4. Justice
5. Respect for people’s rights and dignity

Question 22

Informed consent

Answer 22

Form/document describing procedures, risks and benefits of research
Also describes how data will be treated
Problem: more and more information is required to be in the form, so participants do not always read it

Question 23

Deception

Answer 23

Researchers sometimes withhold information from participants
Deception through omission: not disclosing all information
Deception through commission: lying
Sometimes necessary to avoid reactivity in participant
Participants may experience negative emotions when they learn about deception → may harm trust in scientific research
Debrief is necessary

Question 24

Debriefing

Answer 24

Required for studies that use deception
Often required for any study done in an academic context
Purpose: reestablish trust, give information about study, opportunity for participants to learn something from study

Question 25

Three types of research fraud

Answer 25

1. Plagiarism: research fraud because someone claims an idea to be theirs, but it's not 2. Fabrication: you fabricate data instead of collecting data through fa. an experiment 3. Falsification: you collect data, but you change a couple of things which results in different results and conclusions

Question 26

Prevalence of research fraud

Answer 26

1,97% admits to ever fabricating/falsifying data 14,12% states to know colleagues that ever fabricated data 33,7% admits to 'Questionable Research Practices' 72% states to know colleagues that use QRP's

Question 27

Ten Questionable Research Practices (QRP's)

Answer 27

1. Not reporting all dependent variables 2. Collecting extra data after first analysis 3. Not reporting all conditions 4. Stopping data collection when desired result is reached 5. Incorrectly rounding down p-values 6. Only reporting studies that 'worked' 7. Leaving out data until desired results are reached 8. Hypothesizing after results are known (HARKing) 9. Falsely claiming that certain variables did not influence results 10. Falsify data

Question 28

Reasons for fraud and solutions

Answer 28

Reasons: personality of researcher, rational choice by researcher, social context Solutions: regulations, norms, code of conduct, training, mentoring Main solution: open science

Question 29

Open science

Answer 29

Umbrella term used to refer to the concepts of openness, transparency, rigor, reproducibility, replicability and accumulation of knowledge → fundamental features of science Preregistration to avoid HARKing, sharing information, more attention to replication studies, open access to publications

Question 30

The three 'R'-s for animal research

Answer 30

1. Replacement: could we do this differently? without using animals? replace them with computer stimulations? 2. Reduction: can we reduce our sample of animals? 3. Refinement: can we refine procedures so they are less invasive?