Statistics 200 Honors
Elementary Statistics
Fall 1999

Assignments


General Information - Assignments - Syllabus - Overheads - Computer Labs

Policy on homework assignments: Assignments are given in most classes, usually to be turned in at the next class. Because of the short amount of time allotted for you to finish each homework assignment (one day in some cases), the assignments will be short. I expect that your solutions will be well-written. A good rule of thumb is that a person with no knowledge of the assignment should be able to infer what each question was and understand its solution after reading what you turn in. Grades will reflect how complete your work is, not how correct it is. Late homework is not accepted.

Date AssignedDate DueAssignmentSolutions
Tues., Aug. 24Aug. 25 Read: pp. 14-16 on histograms, pp. 267-277 (sect. 3.4)
Do: Nothing today. Buy the Moore and McCabe book if you haven't!
Solution
Set 1
(pdf or
postscript)
Wed., Aug. 25Aug. 27 Read: pp. 70-79 on normal distributions, etc.
Do: p. 87, exercise 1.76; p. 89, exercise 1.88
Fri., Aug. 27Aug. 31 Read: pp. 376-383 (don't worry much about the technical explanation on p. 380)
Do: p. 391, exercise 5.4; p. 395, exercise 5.18
Tues., Aug. 31Sept. 1 Read: pp. 397-405. Pay particular attention to the Central Limit Theorem.
Do: p. 408, exercise 5.26
Wed., Sept. 1Sept. 3 Read: pp. 434-445, with particular focus on 444-445.
Do: p. 448, exercise 6.6; p. 449, exercise 6.10
Fri., Sept. 3Sept. 7 Read: pp. 453-468 and learn the definitions of all terms in bold.
Do: p. 468, exercise 6.26; p. 470, exercise 6.32
Solution
Set 2
(pdf or
postscript)
Tues., Sept. 7Sept. 8 Read: pp. 624-632
Do: p. 646, exercise 9.10
Wed., Sept. 8Sept. 10 Read: pp. 640-641 (reread pp. 624-632 if you need to!)
Do: From the survey data set we looked at in computer lab last Thursday, choose two categorical variables measured on the sample and perform a chi-square test of the independence of those variables. Begin by summarizing the data in a two-way table, then write a clear report following the steps you did in p. 646, exercise 9.10. I encourage you to use Minitab; feel free to include output from Minitab, but NEVER include output from a statistical package without elaborating on it.
Fri., Sept. 10Sept. 14Read: pp. 51-53. Learn all about sample standard deviation, s.
Do: p. 61, exercises 1.54 and 1.55
Solution
Set 3
(pdf or
postscript)
Tues., Sept. 14Sept. 15Read: pp. 586-595. The standard error and z-statistic formulas should look familiar.
Do: Buy the smallest bag of M&Ms you can find and bring it to class on Wednesday. Make sure to buy regular, non-peanut M&Ms in the dark brown bag.
Wed., Sept. 15Sept. 17Read: pp. 601-606. Do not be intimidated by the ugly-looking formulas for standard error!
Do: From the survey data set we looked at in computer lab on Sept. 2, choose a yes-no question of interest to you. Assume that there were two separate subpopulations sampled, males and females. Test to see whether the proportion of yes in the male subpopulation is the same as the proportion of yes in the female subpopulation against the alternative that they're not the same using the test for 2 proportions. Now set up a cross-tab table and perform a chi-square test for the same question. Are the p-values you get for these tests the same? Comment on the results of the tests in plain English. Feel free to include output from Minitab, but NEVER include output from a statistical package without elaborating on it.
Fri., Sept. 17Sept. 21Read: pp. 504-512. Know what "standard error" means.
Do: p. 524, exercise 7.4
Solution
Set 4
(pdf or
postscript)
Tues., Sept. 21Sept. 22Read: pp. 513-517. Review pp. 504-512 if you have to.
Do: p. 526, exercise 7.16; p. 527, exercise 7.20
Wed., Sept. 22Sept. 24Read: pp. 537-546. Look for similarities with the previous 2 nights' reading on t statistics and t tests.
Do: p. 556, exercises 7.48, 7.49, and 7.50. Please note that 7.50 involves the results of the previous two problems. Please read exercise 7.50 and think about it before you begin the other two exercises.
Fri., Sept. 24Sept. 28Read: pp. 475-480
Do: Choose a quantitative variable from the survey data set we've been looking at in computer lab. Also, identify two subgroups in the sample, such as males and females (you may use some variable other than gender to split the sample if you wish). Your goal is to perform a 2-sample t-test to compare the means of these two groups. Use Minitab as needed.
  1. Produce a stem and leaf plot for each group to see whether there are any serious problems with lack of normality or outliers.
  2. Perform a 2-sample t-test, stating the relevant hypotheses.
  3. Explain your findings in (1) and (2). In particular, comment on whether the assumptions made in performing the t-test seem to hold and on what, if anything, you discovered.
Solution
Set 5
(pdf or
postscript)
Tues., Sept. 28Sept. 29Read: pp. 79-83.
Do: p. 91, exercise 1.98
Wed., Sept. 29---Review the syllabus and overheads for tomorrow's test.
Tues., Oct. 5Oct. 6Read: pp. 312-321
Do: p. 322, exercise 4.38; p. 324, exercise 4.44.
Solution
Set 6
(pdf or
postscript)
Wed., Oct. 6Oct. 7Read: pp. 326-332. Learn what the law of large numbers is.
Do: p. 340, exercise 4.50; p. 342, exercise 4.58.
Thurs., Oct. 7Oct. 14Read: pp. 346-356
Do: p. 360, exercise 4.78; p. 361, exercise 4.82.
Solution
Set 7
(pdf or
postscript)
Thurs., Oct. 14Oct. 15Read: pp. 126-131.
Do: p. 132, exercise 2.22; p. 135, exercise 2.32.
Fri., Oct. 15Oct. 19Read: pp. 135-141.
Do: p. 148, exercise 2.36; p. 150, exercise 2.40.
Solution
Set 8
(pdf or
postscript)
Tues., Oct. 19Oct. 20Read: pp. 154-163
Do: None
Wed., Oct. 20Oct. 22Read: pp. 199-201
Do: p. 173, exercise 2.58 (write this up as if it were a short report)
p. 207, exercise 2.96
Fri., Oct. 22Oct. 26Read: pp. 662-671
Do: p. 697, exercise 10.6
Solution
Set 9
(pdf or
postscript)
Tues., Oct. 26Oct. 27Read: pp. 671-678
Do: p. 697, exercise 10.8
Wed., Oct. 27Oct. 29Read: pp. 681-691 (Do not spend time learning the complicated formulas)
Do: p. 703, exercise 10.24. In part (d), your residual plot should suggest to you that the linear model is missing some structure. Comment on how you can see this from the residual plot.
Fri., Oct. 29Nov. 2Read: pp. 712-718
Do: None; there will be a Reading Assessment Test (RAT) over the reading on Tuesday.
Solution
Set 10
(pdf or
postscript)
Tues., Nov. 2Nov. 3Read: pp. 719-730
Do: Make sure you have a topic for your final project.
Wed., Nov. 3Nov. 5Read: pp. 744-top of 747
Do: p. 734, exercise 11.4 (Note: the data set for this problem is available online.)
Fri., Nov. 5Nov. 9Read: pp. 747-753
Do: The direction of an association between two variables can be reversed if we account for a third, lurking variable. What do we call this phenomenon when it involves categorical variables?

Do the following in Minitab:

  • Generate 3 columns C1, C2, C3 of 100 rows each of standard normal random numbers.
  • Call the C1 column y.
  • Let column C4 equal 3*C1 + 4*C2. Call this new column x1.
  • Let column C5 equal 8*C1 + 4*C2 + C3. Call this new column x2.
Now answer the questions below to turn in:
  1. Plot x1 vs. y in a scatterplot. What is the direction of the association?
  2. Perform a multiple regression of y on x1 and x2. What is the direction of the association of x1 and y after the effect of x2 is accounted for?
  3. Explain how the results of 1 and 2 reveal a reversal of association as referred to above.
  4. For a given city, let x1=number of lifeguards employed in that city last year and y=number of drowning deaths last year. It is reasonable to assume that these variables are positively correlated. Explain why this is true, then suggest a third variable, x2, with the property that for fixed values of x2, x1 and y will be negatively correlated. Explain how this example illustrates the same phenomenon as in the Minitab example above.
Solution
Set 11
(pdf or
postscript)
Tues., Nov. 9Nov. 10Read: pp. 753-759
Do: None
Wed., Nov. 10Nov. 12Read: pp. 753-762 (in case you didn't already read 753-759 from yesterday)
Do: p. 782, exercise 12.10. For part (c), I'd like you to create the ANOVA table by hand instead of using a statistics program. I've done most of the tedious calculations; you just have to show how to use the following information to create the ANOVA table:
Nematodes010005000 10,000
Sample Mean10.6510.425 5.65.45
Sample Variance12.65/36.6275/3 4.64/39.41/3
Treating the 16 observations as a single sample, we get an overall mean of 8.03125 and a sample variance of 133.9744/15.

When you complete the table, you should check that you obtain an F statistic of 12.07974. Use Table E to tell what this reveals about the p-value.

Fri., Nov. 12Nov. 16Read: None
Do: p. 788, exercise 12.24
Solution
Set 12
(pdf or
postscript)
Tues., Nov. 16Nov. 19Read: None
Do: p. 365, exercise 4.98; p. 735, exercise 11.8
Wed., Nov. 17Nov. 18Read: None
Do: Study for test tomorrow; decide between paper and presentation for final project by Friday, Nov. 19.
Fri., Nov. 19Nov. 23Read: 769-772
Do: p. 780-781, exercises 12.5, 12.6 (Note: The numbers in this data set are slightly different from the ones that I used on your exam. Use the Tukey multiple comparisons procedure available in Minitab with a family error rate of 5%.)
Solution
Set 13
(pdf or
postscript)
Tues., Nov. 23Nov. 30Read: 762-769
Do: p. 783, exercise 12.12 (Remember, there's a summary of the relelvant data set on this web page next to the assignment for November 10.)
Solution
Set 14
(pdf or
postscript)
Tues., Nov. 30Dec. 1Read: 483-487
Do: p. 493, exercise 6.64
That's all the assignments for the semester.


Last modified: December 3, 1999
dhunter@stat.psu.edu