align=middle width=80%

 Student's version HTML Format Word Format Computer Supplements Excel Format Text Format
>

Which paper towel is more absorbent?

John E. Boyer*, Deborah J. Rumsey**, Christopher R. Bilder***, and Christopher J. Malone*

* Department of Statistics
Kansas State University
Manhattan, KS 66506

** Department of Mathematics
Ohio State University
Columbus, OH 43210

***Department of Statistics
Oklahoma State University
Stillwater, OK 74078

Statistics Teaching and Resource Library, February 7, 2001

© 2001 by John E. Boyer, Deborah J. Rumsey, Christopher R. Bilder, and Christopher J. Malone, all rights reserved.  This text may be freely shared among individuals, but it may not be republished in any medium without express written consent from the authors and advance notification of the editor.

This group activity focuses on conducting an experiment to determine which of two brands of paper towels are more absorbent by measuring the amount of water absorbed. A two-sample t-test can be used to analyze the data, or simple graphics and descriptive statistics can be used as an exploratory analysis. Students are asked to think about design issues, and to write a short report stating their results and conclusions, along with an evaluation of the experimental design.

Key words: Two-sample t-test

MATERIALS

Each group (2-4 students) needs the following: glass, ruler, stop watch, tongs, water supply, several sheets of each of two brands of paper towels

TIME

One 48-minute class period

OBJECTIVE

Conduct an experiment to compare the absorbency properties of two brands of paper towels by measuring the amount of water each can absorb.

DIRECTIONS

Appoint, with the agreement of the team members, one person who will always fold and dip the paper towel in the glass, and a second person who will read all the water measurements. One person in the group (it may be one of those listed above) will need to record the data for the group. Any other group members assist in the experiment and take notes.

Start by deciding how many measurements you will make on each brand. (A minimum of 4 is required.) After deciding, establish a randomization scheme which provides the order in which observations will be taken.

 When you are ready to start have the person assigned to the measuring duties put an amount of water into the glass, measuring it carefully with the ruler. (About 10cm in a common 12oz. glass works pretty well.)
 Have the person appointed to dip the paper towel describe the method by which they will insert the towel into the glass (are they going to fold it carefully first, or will they wad it up, or will they just kind of cram it in?).
 Also have that person describe when they will remove the towel from the water (how long will they leave it in, will they toss it out as wet as possible, will they let it drain a little, will they let it drain until it stops dripping?)

Now proceed with the first observation. As soon as the towel has been discarded, have the measuring person measure the amount of water left, and record the difference between this value and the initial amount of water as the amount absorbed by the towel.

Repeat this process, duplicating the conditions at each trial as nearly as possible. Record the data in two columns, one for each brand. Beside each observation record the observation number (1 for first, 2 for second, etc.). You may look back at these numbers later to see if there was any trend across time in the way you took measurements.

Discuss with the group appropriate techniques for looking for differences in the absorbencies. Which graphical displays will help you explore the data? Which hypothesis test can be conducted? Does there appear to be a difference between the two brands of paper towel?  Conduct the hypothesis test, draw a graphical display of your data, and draw conclusions.

Discuss with your group any problems that came about in conducting this experiment. If you had the opportunity to redesign this experiment, what changes would you have made, and why? Items to think about: who carried out the experiment (same person each time?); how the experiment was conducted (same way each time?); how the measurements were made (was this accurate and consistent?); was randomization used?; are the paper towels you used in the experiment a representative sample of the whole population of paper towels from each of these two brands? What problems arose that you can prevent next time?

ASSESSMENT

Each person in the group turn in a short typed report (2-3 pages) outlining the design of your experiment, the data you collected, the analyses, graphical display of the data, conclusions, and a brief assessment of how the experiment was designed and conducted.

Provide the appropriate computations and graphics, and write a couple of sentences in the way of conclusions. (Remember, your ultimate goal here is to decide whether or not there are absorbency differences. Your conclusions should reflect that.) Briefly discuss any changes you would make in your design, if you had the chance to do the experiment again. Although the members of your group will all have the same general results, please write up this report in your own words.

• 25% appropriate design; answering issues 1-8 above
• 25% conducting the experiment; recording results
• 25% analysis of results; providing statistics
• 20% clear explanations, appropriate conclusions
• 5% professional look (typed, professional-looking data display, spell-grammar checked)
• TEACHER NOTES

Devise a method to organize groups quickly, and have materials ready. Organize groups of appropriate size so everyone has a job to do. As the activity takes place, be prepared for questions regarding what to do if a paper towel tears, etc. Encourage them to solve their own problems. Emphasize that problems in data collection come up often, and to come up with solutions that are reasonable in terms of the design issues. Be sure to identify which brand was which after the activity is over- they will ask!

Editor's note: Before 11-6-01, the "student's version" of an activity was called the "prototype".