Two one-sample t-tests of the same hypotheses each give . Study 1 reports and Study 2 reports . Quantity A: Study 1's sample size. Quantity B: Study 2's sample size.
Exam 2 Mock: Part I
A z-test of observes . Compute the estimated effect size . Give the exact decimal.
A study reports . On the course scale, the estimated effect is:
A test of whether a new checkout layout changes a grocery chain's mean checkout time gives p = 0.03 and . Which conclusion is best?
A one-sample t-test with 36 observations gives . Which is the most plausible estimated effect size?
A website redesign raises the mean number of pages viewed per visit, with and from millions of visits. Which statement fits the course's view?
To estimate power, a researcher simulates 1,000 studies under an assumed true mean. Results: 180 gave very strong evidence, 230 strong, 150 some, and 440 not enough. What is the estimated power (strong or very strong evidence)?
A team plans a study with 80% power for an expected true effect. It then learns the true effect is probably only half as large. To keep 80% power, the sample size must:
Holding everything else fixed, select every change that increases the power of a test when the null is wrong.
A random sample gives a 95% confidence interval for of (0.41, 0.53). A second random sample, four times as large, gives the same . What is the margin of error of its 95% interval? Give the value to 2 decimal places.
A pollster quadruples the sample size and gets the same . The margin of error of a 95% interval:
A 95% interval for the mean number of texts a city's teens send per day is (61, 73). What is the value 67, and what parameter does the interval estimate?
The multiplier for a 90% interval for a population mean, based on 20 observations, comes from:
A 95% interval for a proportion has standard error 0.02. Its margin of error is closest to:
Which research question is best answered with a hypothesis test rather than a confidence interval?
From a random sample of a state's adults: "We are 95% confident that the true mean weekly hours of exercise for the state's adults is between 3.1 and 4.0 hours." This statement is:
"If we repeated this survey many times, about 95% of the intervals would contain the sample mean from our original survey." This statement is:
Three simulations each draw 100 intervals of the same sample size for one parameter. Plot P has 9 intervals that miss, Plot Q has 19, Plot R has 2. Rank the plots from lowest to highest confidence level.
A 95% interval for is (−3.1, 0.4). Which is the most plausible two-sided p-value for testing with the same data?
A two-sample test of gives a positive t and a two-sided p-value of 0.006. Which is the most plausible 95% interval for ?
Using the same data, an analyst changes the confidence level from 90% to 99%. The interval becomes:
Researchers randomly assign students to study with pop or with classical music, then record each student's quiz score. Which describes MUSIC GENRE?
A Welch test of mean commute minutes (group 1 = suburban, group 2 = urban workers) gives . What does the test statistic mean?
For the commute study, the two-sided p-value is 0.04. Which is a correct definition?
A 95% interval for is (1.5, 4.0). What is the 95% interval for ?
Dog owners in a city take more daily steps than non-owners. Why could the age of the person be a confounder?
Overall, Bakery X has a higher mean customer rating than Bakery Y. Split by weekday versus weekend visits, Bakery Y has the higher mean in both groups. What best explains this?
To show that smoking status is associated with employment sector (office versus outdoor work), select every comparison that is correctly built.
A random sample of a city's adults is randomly assigned to receive a text reminder or no reminder before a library book is due. Late returns differ strongly between groups. What can be concluded?
An observational study of one million randomly sampled adults finds that coffee drinkers live longer on average (p < 0.0001). What is the main reason this does not establish that coffee extends life?
30 original concepts-and-skills questions across Lectures 08–14 in the real Part I mix: single answers, select-all, and Quantity A versus Quantity B comparisons. 80 minutes, help card only, no calculator. Not an official exam. Original questions written in the style of this course. Not official exam content.