12. A bakery is testing to see if the height of a onelayer cake (in inches) is affected by the baking
temperature and the rate at which the batter is stirred. They decide to investigate the effects of
combinations of two temperatures (300 degrees F and 350 degrees F) and three stirring rates (60 rpm,
90 rpm, and 120 rpm) on the height of the cake. They bake 5 cakes at each combination of baking
temperature and stirring rate.
a. Identify the subjects, factor(s), levels, treatments, and the response (outcome) variable
Subjects: 30 cakes
Factors: temperature and stirring rate
Levels: temperature has 2 levels (300, 350); stirring rate has 3 levels (60, 90, 120)
Treatments: There are 2x3=6 treatments:
b. Evaluate this experiment in terms of the 3 criteria we used in lecture.
i. Make comparisons: Set up to try to make good comparisons cleanly between the
temperatures and stirring rates.
ii. Avoid bias: Can’t tell if they controlled for biases; some things to watch for would be
baking time, type of cake (same for all treatments); ingredients for the cakes (all exactly
4
60 rpm
90 rpm
120 rpm
300 degrees
Trt 1
Trt 2
Trt 3
350 degrees
Trt 4
Trt 5
Trt 6
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Documentthe same), the person baking the cakes, where they bake the cakes, location of the cake in
the oven, how they measure height of the cake, etc.
iii. Have enough data. As far as having enough data, I think no, since there are only 5
cakes for each combination. Would like to see more (say around 20 at least)
c. How can you make this experiment better? Justify your answer.
Lots of answers here; biggest thing is to raise the sample size; have 1020 cakes per
treatment (this may be hard to do from a practical standpoint), and talk about how
you’ll handle the issues in part b above.
13. A researcher wants to study the effects of computer activities vs. pencil and paper activities on
learning math for preschool children. Thirty children will participate in her study; 15 children using
computer activities and 15 children using pencil and paper activities. She wants to compare the
performance of the two groups. To form the two groups she will choose between two procedures.
Procedure 1
: She brings the children into a room with 30 desks; 15 desks have computers and
15 desks have a paper and pencil. She tells the children to go to whichever table they want.
Procedure 2:
She flips a coin for each child. Heads = computer group and Tails = paper and
pencil group. Each child goes to their assigned group; when one group is filled the rest go to the
other group.
Which procedure is the best to use and why? (If they are equally effective explain why.) Use the 3
criteria we learned in lecture to answer this question.
Procedure 2 is much better, because it uses random assignment of children to treatments.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '11
 Johnson
 Statistics, researcher

Click to edit the document details