Quizizz 2 Attempt 1

Which of these best describes a classic p-value?

• Assuming the null hypothesis isn’t true, the probability of obtaining the same or more extreme result than one previously seen
• Assuming the null hypothesis is true, the probability of obtaining the same or less extreme result than the one observed
• Probability of the result being due to chance
• How effective an intervention is
• Assuming the null hypothesis is true, the probability of obtaining the same or more extreme result than the one observed

Respectively, Type 1 and Type 2 errors are…

• Rejection of a ‘true’ null & retaining a ‘false’ null
• Rejection of a ‘false’ null & retaining a ‘true’ null
• Retaining a ‘true’ null & rejection of a ‘false’ null
• Retaining a 'false' null & rejection of a 'true' null
• Rejection of a 'true' null & accepting a ‘false’ null

P-rep is…

• Analogous to a p-value
• A descriptive statistic
• Very useful
• A robust measure of replicability
• A measure of reproducibility

Alpha controls the rate of…

• Type 1 errors
• Type 2 errors
• Type 3 errors
• Reproducibility
• Confirmation Bias

Beta controls the rate of…

• Type 1 errors
• Type 2 errors
• Type 3 errors
• Reproducibility
• Confirmation Bias

Statistical power can be defined as…

• Assuming the alternative hypothesis is false, the probability that any test will reject the null hypothesis
• Assuming the alternative hypothesis is true, the probability that a specific statistical test will reject the null hypothesis
• Assuming the alternative hypothesis is true, the probability that any statistical test will reject the null hypothesis
• Assuming the null hypothesis is true, the probability of obtaining the same or more extreme result than the one observed
• How effective an intervention is

As a general rule, smaller the effect size of interest…

• Smaller the sample size needs to be for proper power
• It doesn’t matter what the sample size is for proper power
• Larger the number of variables needs to be used for proper power
• Larger the sample size needs to be for proper power
• All of the above

A direct implication of improper statistical power is…

• Increased T1E rate
• Decreased T1E rate
• Increased T2E rate
• Decreased T2E rate
• Decreased likelihood of a winner's curse

One indirect implication of improper statistical power is…

• Increased likelihood for T1E & winner's curse in significant findings
• Increased T1E rate & likelihood for winner's curse in nonsignificant findings
• Decreased likelihood for T1Es & winner's curse in significant findings
• Decreased T1E rate & likelihood for winner's curse in nonsignificant findings
• Decreased T1E rate and increased winner's curse rate

One indirect implication of improper statistical power is…

• Use of QRPs in a hunt for p<.05 findings
• Use of QRPs in a hunt for p>.05 findings
• Use of the hypothetico-deductive model
• Use of confirmatory research
• All of the above