Session 6:
Homework

Problem H1

Conduct your own comparative study. It can be an observational study or an experimental study. The questions below should serve as guidelines as you proceed.

 a. What question would you like to answer with your study? b. What population are you seeking to answer this question for? For example, the population might be teachers at your school, or boxes of Brand A raisins.

 c. Come up with a design for your study, one that seeks to remove sources of potential bias. (See Design 5 of Problem C4 for a possible design strategy.) Remember that to answer your question, you will need two sets of data to compare. Make sure that you collect enough data to analyze; you should use at least 12 subjects (16, if possible). d. If you've chosen an experimental study, what treatment will you be investigating? Remember that the goal of an experimental study is to judge the effectiveness of a particular treatment.

 e. Determine Five-Number Summaries and box plots for each set of data you collect, and compare the box plots side by side.

 f. Does your data answer the original question? Do your results suggest any new questions? If the data you collected did not answer the question, how might you modify the study to answer the question more thoroughly?

Suggested Readings:

Noether, Gottfried (1994). Mental Random Numbers: Perceived and Real Randomness. In Teaching Statistics at Its Best (pp. 40-41). University of Sheffield, England: Teaching Statistics Trust.
This article first appeared in Teaching Statistics <http://science.ntu.ac.uk/rsscse/ts/> and is used with permission.

 Download PDF File: Mental Random Numbers: Perceivedand Real Randomness

 Session 6: Index | Notes | Solutions | Video

© Annenberg Foundation 2015. All rights reserved. Legal Policy