Chapter 10: Research Design Explained
- You are now at the Chapter 10 section of the book's student Web site. Here you can
- Look over the concept map
of the key terms and look at a glossary of the key terms.
- Test yourself on the chapter objectives
- Test yourself on the key terms.
- Take the Practice Quiz. (a short, general quiz over the entire chapter).
- Take 15-item quizzes over the first part of the chapter.
- Take 11-item quizzes over the second part of the chapter.
- Take 4-item quizzes over the third part of the chapter.
- Take 9-item quizzes over the fourth part of the chapter.
- Take 10-item quizzes over the fifth part of the chapter.
- Do the interactive end-of-chapter exercises.
- Review by looking at the key issues and reading the chapter summary.
You may also look at these very short reviews the
basics of significance testing and the
the t test.
- Do a t test using a statistical calculator or get a better understanding of the t test by going through this short, colorful tutorial (Under the graph, you will see a "Population 1" heading and a "Population 2" heading. We would have labeled these columns "Sample 1" and "Sample 2". Use the arrow buttons to change the mean, variance, and "population" [sample] size. If you are having trouble thinking of values, click on the tutorial's "Parameters" tab at the left of the screen. After you are done playing with values, click on the tutorial's "Simulation" tab to see if your conclusions about the impact of difference between means, variance, and sample size are correct.)
Note that online calculators make doing some advanced statistical analyses
simple. For example,
- if you wanted to calculate effect size from your t test , you could use this effect
- if you had unequal variances and wanted to do Welch's t test,
you could use this online calculator.
- if you wanted to do a power analysis to determine what your chances of
obtaining a significant effect were given a certain sample size, you could
use this online
calculator. (In the left column, find the "Power" folder, and click on
"Mean Difference." Then, click on the "Enter New Data" button. Estimate your
two means, put in your sample size, and estimate your standard deviation
[1/6 of the range is a good guess, so if you had a 7-point scale, 1 would be
a good guess because your range could be 6 [7-1] and 6/6 = 1).
You may want to explore the chi-square test, a statistical test that is especially useful when participants responses are
type of response (e.g., helped or did not help; chose either red, green, or blue) with the following tutorials.
Tutorial 1 (more general)
Tutorial 2 (more detailed)
- If you are confused about the central limit theorem (p. 408), this 10-minute
video from the Khan academy
may help you.
- Find out how to conduct a field experiment by reading "Web Appendix: Field experiments."
- Get a better idea of the steps involved in conducting a study by reading "Web Appendix: Conducting a Study."
- If you want to write up your method section, you may wish to consult the
tips on writing a method section in Chapter 15 as well as the Method section
part of the checklist in Appendix A.
- If you want to write up the results of a simple experiment, you may wish to look at these
download this tutorial.
- Get a better understanding of random assignment by going through this
short (it shouldn't take you more than 10 minutes) but useful
- Get some appreciation of how random error could make two groups differ--and
how large sample sizes tend to reduce the degree to which random error will
cause big differences between groups--by playing with this
- See how easy it is to create a working model of a simple experiment by
watching this short,
If you were really going to do an experiment using this technique, it
would be better to create separate web pages for each condition,
randomly assign participants to condition, and then e-mail participants
assigned to one group the link to one web page and the participants assigned to
the other group the link to the other web page. By giving different participants
different links, you could use Survey
Monkey to do experiments. Another way to use Survey Monkey to do
experiments is, as Blaine Peden explains
here, is to take advantage of its branching feature. Reminder: Please do
not do an experiment--or any research--without your professor's permission.
hints--and some answers--to the end-of-chapter exercises.
Back to Research Design Explained Home Page