Deck 7: Sample Size, Effect Size, and Power

Question 1

Researchers study the relationship between interpersonal violence and health in college age women. The selected an alpha of 0.05. The researchers examined levels of social support for abused versus non abused women. A p value of 0.19 is reported. Based on this information, you know:

A) There is a statistically significant difference in social support between the two groups.
B) There is no significant difference in social support between the two groups.
C) This is a clinically significant result.
D) In this study women who were abused were statistically less likely to report social support.

Answer 1

B

Question 2

Researchers study the relationship between taking Vitamin C and life expectancy A p value of 0.43 is reported. If the researchers were incorrect about this conclusion it would be an example what?

A) A type I error.
B) A sampling error
C) A type II error
D) An overenrolled study

Answer 2

C

Question 3

Researchers studied symptom distress and palliative care designation among a sample of 900 hospitalized patients. They reported that women were significantly more likely to experience distress from fatigue when compared to men. Which statement is correct?

A) The p value associated with this test was > alpha.
B) If this result were incorrect it would be a type I error.
C) If this result were incorrect it would be a type II error.
D) This result has to be correct because it is statistically significant.

Answer 3

B

Question 4

Researchers studied symptom distress and palliative care designation among a sample of 900 hospitalized patients. They reported that women had higher average distress scores from anxiety (2.44) then men (2.20) but that this association was not significant. Which statement is correct?

A) The p value associated with this test was < alpha.
B) If this result were incorrect it would be a type I error.
C) If this result were incorrect it would be a type II error.
D) This result may not be correct because women are more distressed by anxiety.

Question 5

A study reports that the abuse of bath salts (synthetic cathinones) is associated with developing candida. If this result is an error you know this means:

A) The sampling methodology was biased.
B) A type II error was made
C) This is a clinically significant result.
D) A type I error was made.

Question 6

A small study reports that there is no association between taking metronidazole and alcohol sensitivity. A larger study then correctly identifies the association. You now know that in the first study:

A) The alpha level should have been increased.
B) A type II error was made
C) A clinically significant result was found.
D) A type I error was made.

Question 7

A study reports an association between feeding livestock antibiotics and drug resistance in humans. You know this means:

A) The study had adequate power to find the association.
B) The study was underpowered.
C) The researchers did not include enough subjects in their study.
D) The alpha was only 0.10 which is why they reported statistical significance.

Question 8

You are reading a pilot study that reports taking glargine insulin prevents glycemic spikes. The p value is 0.08. You know this means:

A) The selected alpha was 0.05.
B) The selected alpha was greater than 0.08.
C) The selected alpha was less than 0.08.
D) A type II error was made.

Question 9

After completing the power analysis the researchers determine they need a sample size of 400 to have adequate power in their study. After enrolling subjects they have a lower response rate than anticipated and they only enroll 320. Inadequate enrollment may increase the risk of:

A) Systemic bias
B) A type I error
C) A type II error
D) Statistical Significance

Question 10

A study reports that synthetic marijuana use is associated with altered mental status and tachycardia. You know this means:

A) There was an inadequate sample size.
B) There was adequate power to detect a difference.
C) A type I error occurred.
D) A type II error occurred.

Question 11

If you are very concerned about making a type I error what would be the best alpha to select?

A) 0.10
B) 0.05
C) 0.01
D) One that is greater than your p value.

Question 12

A small study reports energy drinks are associated with hyperadrenergic syndrome with an effect size of 0.7. You are funded to complete a larger study on the issue. You know:

A) This is a small effect size so you will only need a small sample to detect it.
B) This is a moderate effect size.
C) This is a strong effect size so you will need a large sample to detect it.
D) This is a strong effect size so you will only need a small sample to detect it.

Question 13

A small pilot study reports consumption of energy drinks are not associated with adverse outcomes in children. However, a larger follow up study correctly reports that energy drinks are associated with death in children under age 5. You know this means:

A) The large study did not have enough power to detect the effect size.
B) The small study was a type I error.
C) The small study may have found the same results if it had a smaller sample.
D) The small study was a type II error.

Question 14

If a study reports INCORRECTLY that an association exists this can only be what type of error?

A) Type I
B) Type II
C) Type III
D) Unable to determine

Question 15

If a study reports INCORRECTLY reports that there is no association or difference this can only be what type of error?

A) Type I
B) Type II
C) Type III
D) Unable to determine

Question 16

If a study has an alpha of 0.05 and a power of 0.80 you know the chance of making a type II error is:

A) 80%
B) 20%
C) 5%
D) Unable to determine

Question 17

A study reports that taking Glucophage is associating with lower postprandial hyperglycemia (p<0.01). You know this means:

A) A type I error was made
B) A type II error was made.
C) A statistically significant association was found.
D) A clinically significant association was found.

Question 18

Your patient completes a DXA scan and the report indicates her left hip bone density has a Z score of -1.4 with a p value of 0.001. You know this means:

A) Your patient's left hip bone density is significantly below average.
B) This is a clinically significant difference.
C) Your patient's left hip bone density is average.
D) Your patient's left hip bone density is significantly above average.

Question 19

If a study has an inadequate sample size you increase the risk of:

A) a type I error
B) a type II error
C) a small effect size
D) a large effect size

Question 20

A pilot study examining weight bearing exercise and bone density detects an effect size of 0.8. You are planning the next study and know this means:

A) you will find a clinically significant effect
B) if you enroll a large sample you may be spending more than you need to
C) if you have a small sample you will overestimate the effect
D) if you have a small sample you will not have enough power to detect this effect size