P-Values
Confidence Intervals
Type 1 Error
Type II Error
Power
100
A p-value is the probability of finding the observed results if this is true.
What is the null hypothesis
100
A 95% ___________ means that we would expect 95% of the intervals from similar samples we take to have the population mean within the interval.
What is a confidence level
100
A Type 1 Error is when you reject the null hypothesis but the null hypothesis is actually true. This is also known as ____ _____.
What is a false positive
100

Type II Error?

What is fail to reject a false null hypothesis

100
Power is the probability of correctly interpreting the null hypothesis how?
What is rejecting the null when it is in fact false
200
A p-value is calculated from the results of a study in comparison to this term that is used to describe the pre-test probability.
What is alpha
200
T/F: A 95% confidence interval indicates that there is a 95% chance that the population mean is within the interval.
What is false
200
What is the probability of committing a Type I Error?
What is alpha
200
How does power correlate with Type II error?
What is 1-beta = power
200
How does power change if you decrease alpha?
What is decreases
300
If a p-value is less than the chosen significance level, then how would you interpret the null hypothesis?
What is reject the null
300
A confidence interval is a measure of getting the same result over and over again, which means it is measuring this statistical term.
What is precision
300
In the demonstrated picture, which label choice points to the alpha?
What is Choice B
300
In the demonstrated picture, what would represent the power of test?
What is 1-A? (A is beta)
300
Researchers input the ideal power, alpha, # of tails, effect size, and SD into statistical software to determine what variable?
What is sample size
400
Besides statistical significance, what are the two other measures of significance with which we are concerned?
What are clinical significance (cost, likelihood that patients will use it, side effects) and practical significance (effect size)
400
If the confidence interval contains the null hypothesis, then how do you interpret the study results?
What is fail to reject the null
400
All else equal, what does bringing the arbitrary cutoff between null and alternative hypothesis closer to the alternative hypothesis mean do to alpha?
What is decreases alpha
400
How does increasing effect size (detection size) affect the chance of making a Type II error assuming alpha is held constant?
What is decreases?
400
How does increasing the standard deviation of a sample affect the power?
What is increases
500
In the paper: treatment failure occurred in 18.6% of the children who received amoxicillin-clavulanate, compared to 44.9% who received the placebo. The p-value for these results must have been less than what conventional significance level?
What is 0.05
500
In the demonstrated absolute percentage point difference graph of confidence intervals, which letter choice indicates the least significant confidence interval?
What is Choice C
500
Freestyle question! Please explain the following text in the context of the otitis media/antibiotics study. Taken from the statistical analysis section: We estimated that with 260 patients, the study would have 90% power to detect an absolute reduction of 15 percentage points in the rate of treatment failure in the amoxicillin-clavulanate group as compared with the placebo group, assuming a 25% rate of treatment failure in the placebo group, with a type I error of .05
With a total control + experimental N of 260, the experimenters would be able to correctly reject the null (saying that there is a significant, over 15% difference in rate of treatment failure in favor of patients given the antibiotics when the difference actually exists) 90% of the time, assuming that there is a 25% rate of failure with the placebo group, with a 5% chance of type I error in rejecting the null (saying that there is a significant difference in outcome between treatment and placebo groups) when there is none in actuality.
500
How does increasing the sample size or improving precision of your assessment tool affect the chances of making a Type II Error?
What is decreases?
500
If the probability of failing to reject a false null hypothesis is 10%, then what is the power?
What is 90%
M
e
n
u