Running head: ONE PERCENT VS TWO PERCENT MILK

One Percent vs. Two Percent Milk

University of Maryland University College

Introduction There has always been an argument in my household regarding the varying degrees of milk and the differences in taste. According to John Bower (1995), “sensory evaluation is a basic tool in the practice of food science and is widely used in research and industry”. The data formulated from the sensory evaluation can come in many forms but the information is usually produced by verdicts of human subjects that form a taste panel. There is importance placed on the conclusions drawn from these sensory experiments (Bower, 1995).
Purpose of the Study An individual declares that she can discern the difference in one percent and two percent milk. We have designed an experiment by which this assertion can be evaluated. Two half gallon cartons of milk were purchased from the Giant Food store. According to the nutrition label of the Giant low fat brand or one percent the product has 110 calories and 2.5 grams of fat. The Giant two percent milk brand has a total of 122 calories and 4.8 grams of fat in each serving. Based on the nutritional information the milk selections are somewhat similar and therefore the null hypothesis of H0: u = the participant will be able to make a distinction between the two milk brands in 70% of the taste tests or u=.70. The alternative hypothesis, H1: u≠.70 in which the participant in the experiment is able to identify between the two milk samples.
Method
Participant The individual selected for this experiment was a 60-year old female who is a self-proclaimed expert in the different milk tastes. Over the past 60 years this individual’s breakfast routine has consisted of milk and cereal and has tried to make the switch from whole milk to two percent milk and then down to one percent milk due to the nutritional value but has been unsuccessful due to the claims regarding the distinction in tastes.
Design
The experiment was conducted over a holiday weekend with ten different trials. The cartons of milk were purchased the day before the experiment and placed in the refrigerator for the same amount of time.
Random Order Each trial was prepared in random order and in pairs. This allowed exclusive control over the distribution.
Temperature Control There was control over the temperature during this experiment. Both types of milk were purchased the same evening and also placed in the refrigerator the same evening. Even within the refrigerator the cartons of milk were placed side by side without anything immediately around to ensure that that milk was cooled evenly. Additionally, both cartons of milk were removed from the refrigerator at the same time to again control the temperature of the experiment.
Constants
In an effort to control this experiment both types of milk were purchased at the exact same store and they were from the exact same producer. It was also important to try to control the age of the product so both milk cartons had the exact same sell by date. All the Dixie cups also had the exact same quantity within each cup.
Data Collection
Materials
One half gallon of one percent Giant brand milk was purchased along with one half gallon of two percent Giant brand milk. The other materials included twenty one Dixie cups for the taste testing along with paper and pencil for recording the results.
Procedure
The day of the experiment was Saturday July 3, 2010 in the early morning prior to breakfast. The following are step by step instructions on how to conduct this experiment: 1. Start off by selecting a participant that claims to be a self-proclaimed expert in distinguishing between the different types of milk. 2. Purchase and/or gather all the following materials: one percent milk, 2 percent milk, Dixie cups, paper, pencil, and water. 3. Discuss the procedure with the participant so they have a clear understanding on the steps and expectations of their participation. 4. Send the participant to another room to ensure control over the experiment. 5. Line up the Dixie cups in pairs in two separate rows. The rows should equal to ten which is the number of trials that will be conducted. 6. Number each Dixie cup in a row either one or two. 7. Create an answer sheet that will show the type of milk that will be placed into each Dixie cup. 8. Take an additional Dixie cup and fill this with water and place on the table off to the side. 9. Next take both type of milk from the refrigerator and pour about halfway into each Dixie cup as denoted on the answer sheet. 10. Once this has been completed you can ask the participant to return to the room and take a seat at the table. 11. Have the participant clean their palate with a swish of water. This should also be done after each row of milk has been completed. 12. The participant can begin with the first row and the first Dixie cup tasting and swallowing the milk. After this is completed the participant is to select the type of milk, either one percent or two percent. The recorder of the experiment will keep track of the taster’s responses. 13. This routine is followed until all ten rows are completed.
Results
The participant was only able to distinguish between the two types of milk eight out of the ten trials. This resulted in an experiment success rate of 80%. The specific results of each trial are shown in Appendix A.
Data Analysis This experiment can also be referred to as a binomial experiment. Binominal experiments are ones that have a set number of trials and each trial only has two possible outcomes. The outcomes in this experiment would have resulted in either the selection of one percent milk or two percent milk. The probabilities within binomial experiments are also constant throughout with the expected success rate of 50% of the time.
Discussion
After reviewing all the results from this milk tasting experiment the participant was not able to always correctly identify the differences in the one percent vs. two percent milk. Prior to this experiment we estimated that the participant would only have success rate of seventy percent which was exceeded with the overall success rate of eighty percent.
Design Changes If this experiment needed to be conducted during another session the following are recommended for this experiment that could result in different outcomes. A. Changing the time of the experiment. This experiment was conducted during the early morning hours prior to breakfast. Since this is a sensory experiment there are factors that could skew the results. According to John Bower, appetite levels can in fact potentially change the outcomes. B. Changing the trials within this experiment. I would be interested to see the outcomes when the trials are truly independent and the participant does not have another selection to make a comparison.

References
Bower, J. (1995). Statistics for food science iii: sensory evaluation data. Part A-sensory data types and significance testing. Nutrition & Food Science, 6. p. 35-42. Retrieved July 11, 2010 from http://instructional1.calstatela.edu/dfrankl/CURR/ntrs455/PDF/Bower_Sensory_Eval_Da ta_III.pdf
Levine, D., Stephan, D., Krehbiel, T., Berenson, M. (2009). Statistics for Managers using Microsoft Excel, fifth edition. Upper addle River, NJ: Pearson Education Inc.

Appendix A

Experiment Results
|Trial/Row Number: |Dixie Cup 1 Identified As: |Dixie Cup 2 Identified As: |Was Experiment Correctly |
| | | |Identified? |
|1 |2% |1% |Yes |
|2 |2% |1% |Yes |
|3 |2% |1% |No |
|4 |2% |1% |Yes |
|5 |1% |2% |Yes |
|6 |2% |1% |No |
|7 |1% |2% |Yes |
|8 |1% |2% |Yes |
|9 |2% |1% |Yes |
|10 |1% |2% |Yes |

Appendix B

Data Analysis
|Probability: Number of ways an event can occur/total possible ways an |P(1% milk)=1/2=.5 |
|outcome can occur |P(2% milk) = ½=.5 |
|Probability of success on a single trial |.5 |
|Number of trials |10 |
|Number of successes (x) |8 |
|Binomial Probability (x=8) |0.044 |
|Cumulative Probability P(x8) |0.11 |
|Cumulative Probability P(x≥8) |0.05 |