Chapter 3

Question 1

What are the 3 types of claims?

Answer 1

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

Question 2

What types of variables are there?

Answer 2

1. Manipulated
2. Measured
3. Unethical - can not be measured or manipulated because of ethics

Question 3

What makes a concept hard to operationalise?

Answer 3

If it is an abstract construct that can not be directly measured (seen, touched or felt). Most researchers develop a creative or elegant operational definitions to capture the variable.

Question 4

What are some ways claims can be made?

Answer 4

They can be made based on personal experience/observation, rhetoric, textual etc.

Question 5

Frequency claims

Answer 5

Measured by proportion, percentage or count. They focus and only have 1 variable - all being measured not manipulated. Does not give any direct info about relationships

Question 6

Association Claims

Answer 6

Involves 2 (at least) related variables. They are often measured, not manipulated. They can often be used to make predictions if there is a relationship. A stronger relationship also means a stronger prediction. Words include: link, associate, correlate, predict, tie to and be at risk for

Question 7

Associations and slopes

Answer 7

Positive associations have a positive slope, negative is negative and zero means, well, zero (horizontal/vertical best fit line)

Question 8

Causal Claims

Answer 8

It has at least 2 variables. If there is 0 associations with 2 variables, there is no causal claim to it no matter what. Words include: cause, enhance, affect, decrease, and change. To determine causality, 2 variables must be correlated, the causal variable should come first and the outcome variable later and that there are no other explanations for the relationship

Question 9

What are the 4 big validities?

Answer 9

1. Construct Validity
2. External Validity
3. Statistical Validity
4. Internal Validity

Question 10

What makes a claim valid?

Answer 10

If it is reasonable, accurate and justifiable

Question 11

The construct validity of frequency claims

Answer 11

Researchers must establish that each variable has been measured reliably and that different levels of a variable accurately correspond to true differences.

Question 12

The external validity of frequency claims

Answer 12

how well does the sample chosen represent the whole population?

Question 13

The statistical validity of frequency claims

Answer 13

the value given from a single study is not an objective truth but an estimate - but hopefully still a precise one. Precision is often captured by a confidence interval/margin of error of the estimate. The point estimate in frequency claims are usually percentages. With more estimates, statistical validity improves.

Question 14

The construct validity of association claims

Answer 14

You need to assess the construct validity of each of the 2 variables and determined how well each was measured.

Question 15

The external validity of association claims

Answer 15

Can the results be generalised to other populations, contexts, times or place?

Question 16

The statistical validity of association claims

Answer 16

How strong is the estimated association and how precise is that estimate especially with other estimates of the same association. Studies with smaller samples often have less precise intervals that reflect uncertainty.

Question 17

What are the three criteria for causation?

Answer 17

1. Covariance
2. Temporal precedence
3. Internal validity

Question 18

Experiments and causal claims

Answer 18

Usually, experiments are used to support causal claims by controlling other possible variables. A variable is often manipulated (the one thought to be the cause) and the other is measured (the consequence). Manipulating means assigning participants to be at one level or the other.

Question 19

How can the method of an experiment establish temporal precedence and internal validity?

Answer 19

By manipulating the independent variable, it ensures that the causal variable comes first. They also have the potential to control for alternative explanation and variables.

Question 20

What is often the most important validity for causal claims?

Answer 20

Internal validity

Question 21

Which of the four validities is the most important?

Answer 21

Depending on the goals, some validities are proven higher than others. Especially with the inverse relationship of internal and external validity. It also depends on time and cost