QCM : Understanding Statistical Errors and Reproducibility — 11 questions

Question 1

1. In frequentist hypothesis testing, what does the null hypothesis stance represent?

A decision to keep all findings regardless of the test outcome

A default claim that there is no effect to compare against

A claim that the effect is already proven to exist

A rule that a result is significant whenever p is below 0.05

Explication

The null hypothesis stance is the default no-effect claim used as the starting point for the test. Rejecting it means moving away from the no-effect stance, not proving the effect in advance.

Answer

A default claim that there is no effect to compare against

Question 2

2. What does a null hypothesis stance represent in statistical hypothesis testing?

A claim that there is an effect to be found

A default claim that there is no effect to compare against

A hypothesis that is always accepted regardless of data

A statement that effects are always present in data

Explication

The null hypothesis stance is a default claim that there is no effect, serving as a baseline against which evidence for an effect is tested. It is not an effect claim itself, nor is it always accepted or assumed effects are present.

Answer

A default claim that there is no effect to compare against

Question 3

3. Which outcome is a true negative in the study outcome framework?

The study finds an effect even though no effect truly exists

The study finds an effect and an effect truly exists

The study finds no effect and no effect truly exists

The study finds no effect even though an effect truly exists

Explication

A true negative occurs when the study does not find an effect and the true state is also no effect. The other options describe a true positive, false positive, or false negative.

Answer

The study finds no effect and no effect truly exists

Question 4

4. What does rejecting the null hypothesis in a statistical test imply about the presence of an effect?

It suggests there is evidence for an effect.

It indicates the effect is definitely real.

It confirms the null hypothesis is true.

It proves the null hypothesis is false.

Explication

Rejecting the null hypothesis suggests there is evidence for an effect, but it does not confirm the null is false or prove the effect is real; it simply indicates the data are unlikely under the null.

Answer

It suggests there is evidence for an effect.

Question 5

5. What is a Type I error?

Concluding that no effect exists when the true state is an effect

Finding a result that exactly matches the true population effect

Repeating a study and obtaining the same result again

Concluding that an effect exists when the true state is no effect

Explication

A Type I error is a false alarm: the test says there is an effect when in reality there is none. The second option describes a Type II error instead.

Answer

Concluding that an effect exists when the true state is no effect

Question 6

6. What is the primary purpose of calculating a P value in hypothesis testing?

To assess the likelihood of obtaining the observed data if the null hypothesis is true.

To measure the effect size in the population.

To estimate the probability that the alternative hypothesis is correct.

To determine the probability that the null hypothesis is true.

Explication

The P value indicates the probability of observing results at least as extreme as those obtained, assuming the null hypothesis is true. It does not directly measure the probability that the null hypothesis itself is true or false.

Answer

To assess the likelihood of obtaining the observed data if the null hypothesis is true.

Question 7

7. What happens when the null hypothesis is false but the test fails to reject it?

A true negative occurs

A Type II error occurs

A Type I error occurs

A true positive occurs

Explication

Failing to reject a false null means the study misses a real effect, which is a Type II error. A Type I error would require rejecting a true null instead.

Answer

A Type II error occurs

Question 8

8. When was the high failure rate in reproducibility of psychological studies first reported by the Open Science Collaboration?

2012

2014

2011

2015

Explication

The Open Science Collaboration first reported an approximate 60% failure rate in reproducibility in psychology studies in 2015, highlighting the reproducibility crisis.

Answer

Publication bias involves selective reporting of significant results, while lack of transparency refers to incomplete disclosure of methods and analyses.

Answer

John Ioannidis

Answer

It increases the likelihood of false-positive findings due to flexible analytic choices.

Question 9

9. How does publication bias differ from lack of transparency in scientific research?

Publication bias involves selective reporting of significant results, while lack of transparency refers to incomplete disclosure of methods and analyses.

Publication bias occurs only in clinical trials, whereas lack of transparency is exclusive to psychological studies.

Publication bias is a form of human error, while lack of transparency is related to researcher degrees of freedom.

Publication bias is about the unintentional errors in data collection, whereas lack of transparency concerns deliberate withholding of data.

Explication

Publication bias involves selectively publishing studies with significant results, whereas lack of transparency pertains to incomplete disclosure of research methods and analyses, which can both distort the scientific record.

Question 10

10. Who is credited with proposing the concept of publication bias and its impact on scientific research?

John Tukey

John Ioannidis

Robert Rosenthal

Karl Popper

Explication

John Ioannidis is well-known for his work on publication bias and its effects on the reliability of scientific literature, highlighting how selective reporting can distort evidence.

Question 11

11. What is a primary consequence of researcher degrees of freedom in scientific studies?

It reduces the need for transparency in reporting methods.

It guarantees the reproducibility of results across different studies.

It eliminates the risk of publication bias by standardizing procedures.

It increases the likelihood of false-positive findings due to flexible analytic choices.

Explication

Researcher degrees of freedom allow for multiple justifiable analysis choices, which can lead to selective reporting and an increased risk of false-positive results, thereby compromising the integrity of scientific findings.

QCM : Understanding Statistical Errors and Reproducibility — 11 questions

Questions et réponses du QCM

1. In frequentist hypothesis testing, what does the null hypothesis stance represent?

2. What does a null hypothesis stance represent in statistical hypothesis testing?

3. Which outcome is a true negative in the study outcome framework?

4. What does rejecting the null hypothesis in a statistical test imply about the presence of an effect?

5. What is a Type I error?

6. What is the primary purpose of calculating a P value in hypothesis testing?

7. What happens when the null hypothesis is false but the test fails to reject it?

8. When was the high failure rate in reproducibility of psychological studies first reported by the Open Science Collaboration?

9. How does publication bias differ from lack of transparency in scientific research?

10. Who is credited with proposing the concept of publication bias and its impact on scientific research?

11. What is a primary consequence of researcher degrees of freedom in scientific studies?

Révisez avec les flashcards

Approfondir avec la fiche

Cours similaires

Physiologie du tissu nerveux

Organisation du système nerveux

Catégories des médicaments vétérinaires

Plans et régions du corps humain

Respiratory Physiology and Diagnostics

Clinique de la périnatalité

Crée tes propres QCM