The next day was test day. I had purchased these cookies for the experiment: Cookie type price/size cost/ounce overall class rating Pepperidge Farm $1.89/5 oz $.378/oz 2.79 Hy Vee Deli 2.49/9 oz .277 3.61 Archway 1.89/11 oz .172 2.29 Keeblers Deluxe 2.99/18 oz .166 3.29 Chips Ahoy 2.99/18 oz .166 2.75 Chip-a-riffic 2.49/18 oz .138 2.29 Shurfine 1.99/16 oz .124 2.71 Hy-Vee generic .99/12 oz .083 2.21 I purchased from two local markets. I used a deli cookie from Hy-Vee. Chip-a-riffic are packaged to look a lot like Chips Ahoy or the Keebler Deluxe. Shurfine is a store brand (Food 4 Less store) as is the Hy-Vee generic brand. The experiment was run this way. I randomly assigned each of the 8 cookies to a brown bag with letters A - H. Before class I put cookie pieces of about the same size (halves or thirds depending on the size of the cookie) in the bags in sufficient quantities. One student, who didn't want to eat cookies, assisted me in the classroom work. We passed out to each student a paper plate that the students divided into geometric sectors with letters A, ..., H. Then we passed bags up and down rows and everyone put a cookie shard of each letter on their plate in the proper sector. I imparted on them the importance of not colluding with their neighbors and essentially being quiet and taking it seriously and they did very well at this. I gave them each a glass of water. I gave each of the 8 sectors of the room (4 in each, there were exactly 32 students tasting) a different order to taste in -- just ABCDEFGH, cycled around 8 times -- to avoid any order effect. They were then allowed to go back and retaste any cookies as often as they liked. They were given the rating sheet below. We had decided as a class to rate on the different factors, and I chose a 1-5 scale (as in "Consumer Reports). I entered the data into Minitab. We spent class time analyzing results. Main findings: There is a modest correlation between quality and price for our sample: R = .49 for the ecological correlation of the 8 points above, R = .2 if you correlate at the level of individual subjects. We also found strong correlations between overall ratings and the other factors: for taste the correlation was .87, for texture .79, and for appearance .62. These are non-ecological correlations. We then spent some time on another day discussing what we'd learned and critiquing the design. Main conclusion: yes, there is a modest correlation. Design criticisms: Maybe don't need all factors since they correlate well with overall rating, giving "pieces" instead of whole cookies makes it hard to rate appearance, I should've stuck with hard cookies for the experiment (the deli cookies and the Archway are soft).
COOKIE DATA SHEET -- Your filling out of this form indicates consent to participate in the experiment. NAME: ________________________________________________ 1. I buy a package of commercial chocolate chip cookies on the average about: (a) once per week, (b) less than once a week but at least once per month (c) less than once per month CIRCLE ONE 2. My preferred brand of chocolate chip cookies is: (Put "no preference" in appropriate). 3. I am MALE FEMALE (cirle one) RATINGS (circle) according to: P = poor F = fair G = good VG = Very good E = excellent ID TASTE TEXTURE APPEARANCE OVERALL COMMENTS A P F G VG E P F G VG E P F G VG E P F G VG E B P F G VG E P F G VG E P F G VG E P F G VG E C P F G VG E P F G VG E P F G VG E P F G VG E D P F G VG E P F G VG E P F G VG E P F G VG E E P F G VG E P F G VG E P F G VG E P F G VG E F P F G VG E P F G VG E P F G VG E P F G VG E G P F G VG E P F G VG E P F G VG E P F G VG E H P F G VG E P F G VG E P F G VG E P F G VG E \PRE