Critically Thinking About Psychological Research

Question 1

What three consequences of misconceptions of NHST often occur ?

Answer 1

• Overestimate the importance of effects
• False positive
• False negative

Question 2

What is meant by p-hacking?

Answer 2

researcher degrees of freedoms that lead to the selective reporting of significant p-values

Question 3

What is meant by HARKing?

Answer 3

Hypothesising after the results are known

Question 4

Name some ways to p-hack

Answer 4

trying multiple analyses, deciding when to stop data collection outside of the predetermined time and including or not including data based on the effect it has on the p-value

Question 5

What is meant by a p-curve?

Answer 5

The number of p-values you would expect to get from each value of p

Question 6

How are p-curves based on data computed?

Answer 6

By a statistical summary of p-values which are derived from studies

Question 7

How can you compute p-curves based on statistical theory?

Answer 7

The curve is affected by the effect size and the sample size but has a characteristic shape. When there is no effect size the curve is flat - all p-values are equally likely. For effect sizes greater than 1, the curve has an exponential shape because smaller ps occur more often than larger ps