Examples of Questions You Can Expect on Exam 2

Comment: These are illustrative examples of test problems. They are not meant to be a complete study guide. See StatPrimer, the HW assignments, and the "Can You Questions" for details regarding coverage.

Part A: Closed book

(1) Fill in the table cells below with the following choices: "correct retention," "correct rejection,""Type I Error," or Type II Error"

(2) Match each of the following terms with their definitions (below):

_____ statistical inference

_____ population

_____ power

_____ confidence

_____ beta

_____ alpha

_____ "null value"

_____ parameter

_____ sampling distribution of means

_____ confidence interval

_____ p value

(A) The "universe" of potential values to which we wish to infer

(B) The probability of a type I error

(C) The probability of a type II error

(D) The act of generalizing from a sample to a population with calculated degree of certainty

(E) A characteristic of the population.

(F) The hypothetical frequency distribution of all possible sample means based on repeated independent samples of size n from the same population.

(G) A range of values that has known certainty of capturing a parameter.

(H) The probability of not making a type I error

(I) The probability of not making a type II error

(J) The probability of obtaining a test statistic that is equal to or more extreme than the observed statistic, assuming the null hypothesis is true

(K) The value of the parameter as specified under a null hypothesis.

(3) What SPSS commands can used to compute summary statistics for continuous variables?

ANS: Several choices (e.g., Statistics | Summarize | Describe or Statistics | Summarize | Explore)

(4) What SPSS menu choice is used to create a new variable to capture the difference between two variables?

ANS: Transform | Compute

(5) The standard error a sample mean is a measure of its:

(A) spread

(B) location

(C) shape

(D) precision

Comment: Students often miss the above question. The correct answer is D; the standard error of the mean tells you how far the sample mean is likely to differ from the population. It is therefore a measure of the sample mean's precision (see StatPrimer p. 8.3, last paragraph, and p 8.4)

(6 How many degrees of freedom does a one-sample t test have when n = 20?

Ans: One-sample problems have n - 1 degrees of freedom.

(7) Which of the following factors determine power of paired t test?s

(A) sample size

(B) alpha level of the test

(C) standard deviation of the measurement

(D) "a" and "c"

(E) "a," "b," and "c"

Answer: E. Power is a function of A, B, and C, and is also a function of the expected difference in the population (assuming the alternative hypothesis is true).

(8 A 95% confidence interval for a mean has a 95% chance of capturing:

(A) "x bar"

(B) �

(C) s

(D) "sigma"

(E) "sigma squared"

Answer: B. Always remember that the confidence interval is trying to capture the parameter, not the statistic.

(9 A margin of error of �10 corresponds to a confidence interval length of:

(A) 5

(B) 10

(C) 20

(D) none of the above

Answer: C. The margin of error is half the confidence interval width.

(10) Name the two forms of statistic inference.

     (A) _______________________________________

     (B) ________________________________________

Part B: Open Procedure Notebook

These questions will be much like HW exercises! Here are some examples.

(1) A patient satisfaction survey finds a mean satisfaction score of 8.911 with a standard deviation of 4.667 (n = 25).

(A) Calculate a 95% confidence interval for � and interpret your results.(Show all work.)

(B) Calculate a 90% confidence interval for � and interpret your results. (Show all work.)

(2) The data set XXX (on the network in J:\Gerstman\hs167\) contains the variables X1, X2, etc.:

(A) Compute routine summary statistics for the variable X1

(B) Compute a 95% confidence interval for � of the variable X1.

(C) Test whether the population mean for the variable X1 is different from 100. List the null and alternative hypotheses. Let alpha = .01. Report the hypothesis testing statistics using suitable notation. Report your conclusion.

(D) Etc.

(3) An dietary intervention is done, with pre-intervention and post-intervention cholesterol levels as follows:

(A) Create an .SAV file with these data. Call this file "yourname.SAV" and turn it in on a floppy.

(B) Report summary statistics (n, "x bar", s) for the pre-intervention sample.

(C) Report summary statistics for the post-intervention sample.

(D) Report summary statistics for the change from pre-intervention to post-intervention.

(E) Test whether the change was significant. (List all four steps of the test.)

(F) Etc.