Biological Statistics
Homework 8
Due Thursday, November 9

You must type this and all other homework assignments. Do not e-mail the assignment to me; turn it in early (at 322 Wolf) for a foreseeable absence, or turn it in late after an unexpected absence from class.

1. Here are some data on the abundance of the tesselated darter, Etheostoma olmstedi, in streams in Maryland. The data are presented as number of fish per 75 meter section of stream. For each stream, data from two sections are given: one section adjacent to a pasture, and the other section not adjacent to a pasture. You want to know whether streams adjacent to pastures have the same density of tesselated darters as streams not adjacent to pastures. Analyze the data using: a regular t-test; a paired t-test; Wilcoxon's signed ranks test; and a sign test. Write a sentence or two interpreting the result of each test. Write a sentence or two explaining which you think the most appropriate test for these data would be, and why.

 

   stream            pasture   other
Andover Branch           0       31
Basin Run               52       21
Big Pipe Creek           2        0
Bynum Run               30       14
Carroll Branch          19       19
Delaware Run            10        0
East Br. Patapsco River  3        2
Gillis Falls             7        1
Gunpowder Falls          1        1
Israel Creek            22       30
Jones Falls             10        8
Little Deer Creek       21       60
Mason Branch            74      111
Middle Run              21        2
Northeast Creek         79       78
Patapsco River          54        4
Piney Run                4        2
Pomonkey Creek           2        0
Prettyboy Branch         2        0
Principio Creek         22       58
Red Lion Branch         55       23
Tuscarora Creek          1        0
Winters Rn              75      116

2. Write an exam question for which the correct answer is one of the tests that only involves attribute variables (tests of goodness-of-fit and independence). Make it in the same format as the questions on the exams: describe a biological experiment, mention the biological question to be answered, and ask which test is the best. Be sure to give the answer.

3. Write an exam question for which the correct answer is one of the tests involving a measurement variable (anovas, their non-parametric analogues, etc.).

4. I'm constantly trying to improve the web pages for this class. To help me, go to the web page listed below with a number that corresponds to your birth date and critique it. Your critique can consist of general comments typed in your homework, and specific comments that you write on a printout of the web page (spelling mistakes, unclear sentences, ugly format, etc.).

  1. Introduction
  2. Kinds of biological variables
  3. Probability
  4. Hypothesis testing
  5. Random sampling
  6. Exact binomial test
  7. Power analysis
  8. Chi-square test of goodness-of-fit
  9. Randomization test of goodness-of-fit
  10. Chi-square test of independence
  11. Fisher's exact test
  12. Small numbers in chi-square and G-tests
  13. Repeated G-tests of goodness-of-fit
  14. Central tendency
  15. Dispersion
  16. Standard error
  17. Confidence limits
  18. Student's t-test
  19. Introduction to anova
  20. Model I vs. Model II anova
  21. Testing homogeneity of means
  22. Planned comparisons among means
  23. Unplanned comparisons among means
  24. Estimating added variance components.
  25. Normality
  26. Homoscedasticity
  27. Data transformations
  28. Kruskal-Wallis test
  29. Nested anova
  30. Two-way anova
  31. Paired comparisons

Return to the Biological Statistics syllabus

Return to John McDonald's home page

This page was last revised November 2, 2007. Its URL is http://udel.edu/~mcdonald/stathw8.html