IRAT Questions Exam 2 Flashcards
Which of the following represents 50% of the data around the mean when the data is normally distributed?
Interquartile Range
Which of the following is not part of descriptive statistics?
Making predictions based on data collected
What kind of data would the concentration values of glyphosate in soil be?
Ratio
Which of the following can be determined from an ROC curve?
Sensitivity and specificity of a diagnostic test
Incidience proportion is:
Number of new cases of a disease divided by total number in the population at risk
Which of the following is a nonparametric test?
Kruskal-Wallis
(hint: nonparametric tests are usually named after people)
Which of the following does not apply to normal distribution?
98% of normally distributed data is found within 2 standard deviations.
In hypothesis testing, we commit a type II error (beta) when we:
fail to reject the null hypothesis when it is false, possibly due to lack of power
Which type of distribution is Central Limit Theorem associated with?
Normal
Which of the following does not apply to correlation?
When data is of ranked type, the correlation coefficient is between 0 and 1.
In a simple (bivariate) linear regression, which of the following values for the R-squared corresponds to the situation where all of the data points in the scatterplot lie on the regression (or trend) line:
1
Which of the following is least likely to be considered a so-called “natural experiment” with confounding factors affecting the outcome of the study?
RCTs
What is the most commonly used statistical test in logistic regression to assess the statistical significance of the analysis?
Chi-square
In a regression analysis, the variable(s) that the researchers use to explain the final effect/outcome is known as:
Independent variable
Which of the following is false?
In logistic regression, coefficient estimates can be used to create hazard ratios.
The phrases below describe various concepts used in hypothesis testing, sample size calculation, and power analysis. Select the wrong match.
Probability of rejecting the null hypothesis when it is true (Power)
When computing sample size using power analysis, which of the following are conventional levels for alpha and power, respectively?
0.05 and 0.80
An investigational drug is being compared with a placebo for the treatment of hypertension. A research team is attempting to estimate the sample size necessary to detect a 12 mmHg difference in systolic blood pressure level with 90% power at the 5% significance level. The standard deviation, estimated from previous data, is assumed to be 12 mmHg. The team plans to conduct a two-sided hypothesis test (test for equality or test for difference). Prior to the conduct of the study, the research team decides that they want to be able to detect a 10 mmHg difference in systolic blood pressure instead of 12 mmHg. Holding everything else the same, what effect will this change have on the required sample size?
Increase required sample size
A researcher has collected data, performed the statistical test of interest, and now wants to see how powerful the test was (i.e., wants to answer the question of whether the test was powerful enough). Which of the following best describes this situation?
This is called retrospective power analysis and its use should be avoided.
Which of the following is the expression for power?
1-beta