Thinking Critically about Psychological Theories

Question 1

What assumptions are made when commenting on causality from correlational studies?

Answer 1

Strong assumptions about all the potential variable that are involved in affecting the variable you are studying, about the direction of the relationship and the role these variables play ( mediator etc)

Question 2

When trying to find the pure effect of one variable (sex) on another (aggression) which is mediated by another variable (testosterone) which causes another variable (baldness), which of these variable should we condition for?

Answer 2

Not testosterone because it would nullify the effect and not baldness because it would have a similar effect since it is correlated too highly. So none.

Question 3

What is meant by berkson’s paradox?

Answer 3

When two diseases have a positive correlation between them but are not actually related

Question 4

What is a possible effect of conditioning on a variable that is both a mediator and a collider when there is a third predictor variable?

Answer 4

Can create a spurious correlation between two other predictor variables

Question 5

Name a way of making your assumptions explicit

Answer 5

Through a theory

Question 6

What can often implicitly affect how we theorise ?

Answer 6

Our statistical techniques

Question 7

What is meant by the hot hand fallacy?

Answer 7

Representativeness causes people to expect good throws from people who have already throwing a good series

Question 8

What is the opposite of hot hand fallacy?

Answer 8

Gambler’s fallacy

Question 9

What fallacies are posed by drawing conclusions from null hypothesis testing? (3)

Answer 9

false dilemma, straw target and common practice

Question 10

What is the difference between verbal and formal models?

Answer 10

Verbal models: explanations of psychological phenomena with words.
Formal models: explanations of phenomena in things like mathematical equations or computer code.

Question 11

Name four advantages of formal models

Answer 11

Designing strong tests of theorys, sharpening research questions, going beyond linear theories and addressing real world problems

Question 12

How does formal models help design strong tests of theories?

Answer 12

Models enable researchers to make competing quantitative predictions, which in turn lead to strong comparative tests of theories. Instead of comparing a hypothesis against chance, competing theories can be tested against each other. It is possible to quantify how much a model’s predictions deviate from empirical observations.

Question 13

What is the problem with research questions from verbal theories?

Answer 13

Verbal theories, if left underspecified, can be used post hoc to ‘explain’ almost any observed empirical pattern

Question 14

How does formal models help sharpen research questions?

Answer 14

Formal quantitative predictions are often not easy to grasp by intuitive reasoning: computer simulations are often needed. Often it is only when starting with modelling that one learns what a theory really predicts.

Question 15

How does formal models help with addressing real world problems?

Answer 15

Some designs (such as factorials) can lead to tests of theories under conditions that have little to do with the real-world, which results in a lack of external validity. Modelling allows researchers to deal with natural confounds without destroying them: they can be built into the models.

Question 16

What is meant by trading on equivocation?

Answer 16

deliberately exploiting the ambiguity and vagueness of a sentence.

Question 17

What is meant by Bononi paradox?

Answer 17

increased complexity of a model can cause us to be able to better predict the outcome value, without a better understanding of what we are modelling. The model becomes too complex.

Question 18

What is meant by overfitting?

Answer 18

when a certain model fits the current data so well, that it is likely that it won’t fit any new data.

Question 19

What is the ‘cost’ of overfitting?

Answer 19

Overfitting is at the expense of the generalisability of a model.

Question 20

Which models are often more robust to overfitting?

Answer 20

Simple/linear models are usually more robust.

Question 21

What is meant by abductive reasoning?

Answer 21

Inference to the best explanation

Question 22

What is meant by descriptive adequacy?

Answer 22

the extent to which a theory is in line with available empirical data.

Question 23

Why should a theory be precise as possible?

Answer 23

so that other theorists can interpret it easily and unambiguously

Question 24

What does this precise explaining of a theory involve?

Answer 24

This involves defining constructs and relationships between them well.

Question 25

What is meant by saying that a theory should be coherent and consistent?

Answer 25

a theory should be free of logical flaws and all its components should form a coherent whole. One must also look for circularity.

Question 26

How can one pass the Prediction and falsifiability criteria for theory evaluation?

Answer 26

It should be possible to reasonably reject a theory through critical tests

Question 27

How can one pass the Postdiction and explanation criteria for theory evaluation?

Answer 27

a theory should provide a genuine explanation of existing results.

Question 28

How can one pass the parsimony (Occam’s Razor) criteria for theory evaluation?

Answer 28

a theory should be as simple as possible / not unnecessarily difficult. Only the causes or constructs that are needed to explain a phenomenon should be posited.

Question 29

What is meant by the breadth of a theory?

Answer 29

the extent to which a theory is applicable to a broad range of phenomena or is restricted to a limited domain. Theories should attempt to be as broad as possible, while maintaining the other criteria of a good theory.

Question 30

What is meant by the originality of a theory

Answer 30

The extent to which a theory is new or essentially a restatement of an existing theory. Two theories based on opposite assumptions can make exactly the same predictions for an experiment.

The degree to which it generates new testable predictions

Question 31

What is meant by the usability of a theory?

Answer 31

Usability = the extent to which a theory is useful in addressing societal problems.

Question 32

What is meant by the rationality of a theory?

Answer 32

a theory should make claims about the architecture of the mind that seem reasonable in light of evolutionary theory.

Question 33

What is meant by a potential falsifier?

Answer 33

any potential observation statement that would contradict a theory.

Question 34

When is a theory more falsifiable than another?

Answer 34

A theory is more falsifiable than another when the class of potential falsifiers is larger.

Question 35

Why are theories that are more falsifiable preferred?

Answer 35

Theories that are more falsifiable are preferred because it takes less data to falsify such a theory. Simple theories (such as straight lines instead of curves) are easier to test.

Question 36

How can a theory be more falsifiable?

Answer 36

A theory can gain in falsifiability not only by being precise, but also by being broad in the range of situations to which the theory applies. The greater the universality of a theory the more falsifiable it is, even when the predictions are not very precise.

Question 37

What is meant by the degree of corroboration?

Answer 37

A theory B can be called more strongly corroborated when it is more falsifiable than a theory A, and it passed the same test as A.

Question 38

What is meant by Ad hoc?

Answer 38

a revision or addition to a theory that decreases its falsifiability. According to Popper, one should only revise or introduce theories that increase the degree of falsifiability of the system.

Question 39

Name three ways in which statistical bias enters?

Answer 39

(1) things that bias the parameter estimates (including effect sizes); (2) things that bias standard errors and confidence intervals, and; (3) things that bias test statistics and p-values.

Question 40

What are the three forms of circular reasoning?

Answer 40

Repeating the premise in the conclusion
Premise presupposes the truth of the conclusion
Premise is logically or semantically equal to the conclusion

Question 41

What is meant by versimilitude?

Answer 41

A theory that is more falsifiable (excludes more and has been tested in several different situations) has more verisimilitude

Question 42

Which of the criteria are more important than the which criteria?

Answer 42

Dexcriptive adequacey, precision, coherence & consistency, falsifiability, Post-diction (explaining a theory after data collected) > parsimony, originality