Instructions For this project, you will write a paper that involves designing, conducting, and analyzing results from your own experiment to test the sensory discrimination of one of your friends. Choose a friend who claims to be able to tell the difference between two very similar objects. You are free to choose the two objects, however here are some ideas to get you thinking along the right lines: Coke vs. Pepsi, Regular vs. Decaf coffee, Sweet 'n' Low vs. Equal sweetener, butter vs. margarine, or Ghiradelli chocolate vs. Hershey?s chocolate. Which you choose depends on whoever your ?expert? testing subject is. You will need to find an expert on your own, and will need to supply your own materials. The cost of your experiment will obviously depend primarily on the objects to be tested; testing to see if the subject can discern between bottled water and tap water will only cost a couple dollars, testing to see if the subject can discern between different brands of caviar will get pricey fast! Your paper should consist of a single document in Microsoft Word format. After you have completed your paper, you must submit it to Turnitin.com. Login and password details for Turnitin.com can be found in the classroom. 1. First, you should brainstorm to find a friend who is either a self-proclaimed expert or has some food/beverage related idiosyncracy(1). 2. Determine a protocol for administering the test a) What should be done about random variations in the items to be tested? For example, how do you prevent temperature change in the product, or to make sure each cup of coffee is equally sweet or has the same amount of cream? Before conducting your experiment, carefully lay out the procedure for administering the test, such as how the coffee will be made, how the wine will be decanted, and so forth. This is related to the important issue of ensuring that each trial is independent (i.e. the previous trials aren?t tainting your taster?s sense of taste). If you have ever been to a wine tasting, you?ll notice that you can tell the difference between the first few sips of wine, but once you get to the 11th and 12th glass, it?s all sort of a blur (and not due to inebriation!). Carefully consider how you will attempt to overcome this problem. b) How many trials should be used in the test (for this, see Levine Chapter 7 and 8.4, or read up on the Fisher exact test)? Should they be paired or independent(2)? In what order should the items be presented? Should the experiment be run in one sitting or spaced out over a number of days? Done properly, the experiment should be designed such that if the expert has no expert skills, the result will be wholly governed by chance. The number and ordering of the trials should allow an expert to prove his or her abilities while simultaneously preventing a fraud from succeeding. You will want to keep the binomial distribution in mind when making these decisions. c) What conclusion could be drawn from a perfect score or from a test with one or more errors? For the design you are considering, list all possible results of the experiment. For each possible result, decide in advance what conclusion you will make if it occurs. In determining this mapping of outcomes to conclusions, consider the probability that someone with no powers of discrimination could wind up with each outcome. You may want to make adjustments in your design to increase the sensitivity of your experiment. For example, if someone can?t distinguish decaf coffee from regular, then just by guessing, he/she should still be right half of the time and there will be a small chance, which you should calculate, of being right 100% of the time. On the other hand, if the taster possesses some, but not perfect, skill in differentiation, he or she will make some mistakes. 3. Write out an instruction sheet for your experimental process. Conduct a ?dress rehearsal? on somebody other than your subject to work out kinks in the process. After the practice run, determine whether or not you want to make changes in your instruction sheet to address any problems that arose. This practice run is an extremely important step; many people make big mistakes in their first attempt and historically, students who have taken this step seriously have scored much more highly on this project than those who do not. 4. You should now be ready to run your experiment. Record your results CAREFULLY, and note any unusual occurrences in the experimental process. It may be a good idea to keep track of the order in which the samples are served to your subject. 5. Numerically, summarize the results and analyze the data. Do they support or contradict the claim that the subject possesses no sensory discrimination? Use your list of all possible events and subsequent actions to come to a conclusion. Discuss the reasons behind the decision that you have made. For ideas on how to analyze the data, see Levine Ch. 9 or use the Fisher exact test. 6. What changes would you make to your experimental process if you had the opportunity to do it again? Why didn't you pick these issues up when you did your practice run? You are to work individually on this assignment. You can find definitions of academic dishonesty and plagiarism at http://www.umuc.edu/policy/academic/aa15025.shtml, along with the procedures I will follow should I suspect any transgressions of such nature. Suggested Report Format with Headings Purpose of the study The purpose of the study formally explains why the study was important to conduct and write the null and alternative hypotheses of the study (see Levine Ch. 9.1 and read the lecture material in the classroom from session 8). Method The Method section explains how the study was conducted to collect data. The section should include the following: 1. Participants of the study. Who are they? Where did you get the participants for the study? Which sampling design was used to select the participants? 2. Experimental design. Which experimental/sampling design was used in the study? 3. Data Collection. Explain the method used to collect data. 4. Data Analysis. For hypothesis testing, explain how you plan to calculate the data. 5. Step-by-step procedure. Explain the explicit steps from beginning to end for conducting the study. If another researcher wants to replicate the study, the directions must contain precise information for conducting the study. Data Analysis Show the results of your data analysis, including tables and charts as necessary. You should also discuss, in plain English, the results of the data calculation, both in terms of magnitude and significance. Discussion Discuss what you learned from the data collected. Did the results support or not support the null hypothesis? Appendix Place the charts, tables, and other relevant documents in an appendix at the end of your paper. Your paper should contain between 5 and 7 pages of text, plus any additional charts/tables/graphs found in the appendix. -------------------------------------------------------------------------------- (1) While something simple like butter vs. margarine or bottled water vs. tap water is a reasonable topic, students who have been more creative in their subject selection have found this paper to be rather enjoyable. Some examples of creative subjects: Testing to see is one call tell the difference between freshly grated parmesan cheese and the Kraft brand parmesan cheese (in the green tube) One individual's wife maintained that the white candies from a box of Good 'N' Plenty candy did not taste as good as the pink candies. The test was to see if his wife could tell the difference. One individual had a friend who ran a wine tasting social group. She tested whether or not her friend could discern between red and white wine by bouquet alone. Note that this does not say that one of the more "creative" ideas will necessarily make for a better experiment, paper, or grade. You are more likely, however, to enjoy the process of conducting the experiment and analyzing the data should you put serious thought into your test subject. (2) Paired vs. independent?In the context of a Coke vs. Pepsi expert, for example, in a paired test, for each trial you present one of each type at the same time and tell your subject to identify which one is Coke and which is Pepsi. If you were doing independent trials, in each trial they?d get exactly one cup, filled with either Coke or Pepsi, and be asked to identify which it is.
Paper#11663 | Written in 18-Jul-2015Price : $25