You are on page 1of 14

Case Study - RES 342

Deb is the owner of Deb's Pizza Palace - an establishment that prides itself on preparing healthy but delicious pizzas sold by the slice. She is currently selling pizzas in a small shop on Monroe Ave. in Corvallis, but is considering an expansion of her current business to other campuses and needs help making this decision. She's collected some data and asks you to help her analyze and present it to the Small Business Administration loan officer who will either reject her proposal or provide funding for the proposed expansion. Each of the following 4 worksheets has a data set and one or more questions. Analyze the data, answer the questions, and report your conclusions and recommendations in a 1-2 page business memo written in Microsoft Word for Deb. Data calculations should not be placed in the meme - you should attach your Excel spreadsheets.

Portland 229.91 241.13 226.85 233.74 225.32 216.40 245.98 245.21 240.88

Pizza Sales Eugene Salem 300.56 267.24 244.40 265.10 272.84 283.77 291.20 255.92 300.20 259.59 284.00 246.13 323.24 307.94 307.40 293.56 292.28 275.81

Decision One: ANOVA Deb is contemplating three new locations for her pizza business to campuses in Portland, Eugene, and Salem A random survey of 27 piz shops in those locations is conducted with 9 stores surveyed in each location. (Deb hired college students to sit in the stores and eavesdrop the pizza ordering clerk). The number of pizzas sold per hour was talli and the students then computed the sales volume in actual dollars. Deb wants to know if there is a significant difference in location.

a) What are the null and alternate hypotheses? H0: all the means are equal Ha: at least one mean is different b) What are the degrees of freedom for the numerator and denominat 2 and 24 c) What is the critical value of F at the 5% level of significance? 3.40 d) What is the mean square for treatments? 7575 e) What is the computed value of F? 22.98 f) What is your decision? Reject H0, at least one mean is different Attach your answers to these questions to the business memo.

Anova: Single Factor SUMMARY Groups Count Portland 9 Eugene 9

If you find a significant difference, construct 95% confidence interval order to determine which location(s) might be different. Your recommendation to Deb should include the location(s) where she can ex the optimum sales. Sum Average Variance 2105.4 233.9333 102.5408 2616.12 290.68 504.4392 See below .

. Salem 9 2455.056 272.784 381.9985 Portland Eugene Salem 229.91 300.56 267.24 241.13 244.40 265.10 df MS F P-value F crit 226.85 272.84 283.77 2 7574.738 22.97746 2.66E-06 3.402826 233.74 291.20 255.92 24 329.6595 225.32 300.20 259.59 216.40 284.00 246.13 26 245.98 323.24 307.94 245.21 307.40 293.56 240.88 292.28 275.81 interval upper 241.72 307.94 287.81 interval lower 226.15 273.42 257.76 Portland is lower than the other two.

ANOVA Source of Variation SS Between Groups 15149.48 Within Groups 7911.828 Total 23061.3

for her pizza business - adjacent m A random survey of 27 pizza h 9 stores surveyed in each sit in the stores and eavesdrop on pizzas sold per hour was tallied s volume in actual dollars. Deb ference in location.

the numerator and denominator?

% level of significance?

o the business memo.

truct 95% confidence intervals in ht be different. Your e location(s) where she can expect

Group 1

Pizza Sales
Group 2 Hawaiian Slices sold without expensive ad campaign and promotion 19 24 23 20 19 19 19 17 16 15

Pepperoni Mushroom Olive Slices sold with expensive ad campaign and promotion 22 26 26 26 25 25 24 22 21 18

Deb is trying to decide whether to launch an ad campa would possibly boost sales. Her dilemma is whether th would pay for itself. She knows that sales of her Pepp Mushroom Olive and Hawaiian pizzas are remarkably c she decides to try a small version of the promotion on Pepperoni Mushroom Olive pizza and compare it to sale Hawaiian pizza.

You take a random sample of pizzas sold during a one h over the next week. Theresults are shown in the table of this box. Does the ad campaign make a difference?

Set up a model to test whether there are differences in mean satisfaction between groups. Use an alpha of .05 Step 1: H0: mean1 = mean2 Ha: mean1 not equal to mean2 Step 2: The test statistic is 3.587 Step 3: The critical value is 2.10 Step 4: The two tail p value is 0.002, which is low. We reject the null hypothesis. Step 5: Since we reject the null hypothesis, the means are not equal, so advertising does make a difference.

Be sure to fully explain your conclusion. in the business memo. You need to be correct, but it's equally important that your audience understand your explanation.

her to launch an ad campaign that Her dilemma is whether the campaign ows that sales of her Pepperoni n pizzas are remarkably consistent, so rsion of the promotion on the zza and compare it to sales of the

pizzas sold during a one hour period ults are shown in the table to the left mpaign make a difference?

oups. Use an alpha of .05 t-Test: Two-Sample Assuming Equal Variances

Group 1 Pepperoni Mushroom Olive Slices sold with expensive ad campaign and promotion Mean Variance Observations Pooled Variance Hypothesized Mean Difference df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail 23.5 7.166666667 10 7.522222222 0 18 3.587274406 0.001052987 1.734063592 0.002105973 2.100922037

Group 2 Hawaiian Slices sold without expensive ad campaign and promotion 19.1 7.877777778 10

a difference.

but it's equally important

Deb's final decision is which new flavor of pizza to introduce to her customers. She offers 4 new varieties during the next week at the Corvallis location and tallies the number of slices sold below.

Pizza
All Beef, All the Time Shrimp and Stuff Salmon, Marionberry & Hazlenut Soybean Surprise

Observed
37 22 36 5

Expected
25 25 25

(o-e)^2/e
5.76 0.36 4.84

16 25 chi square 26.96 What are the hypotheses for your research design? Test them using the five step model. On the basis of your results, which varieties (if any) should Deb roll out to all her new restaurants? Explain the rati

H0: the categories are uniformly distributed Ha: the categories are not uniformly distributed The chi square test statistic is 26.96 df = n-1 = 3 The critical value is: 7.815 The test stat is well above that, so we reject the null hypothesis. The categories are not uniformly distributed. The soybean surprise was a bust and is not good.

restaurants? Explain the rationale for your recommendation.

Cost/Slice($) to Product Pepperoni Mushroom Olive Garlic Chicken Hawaiian Vegetarian Vegen Cheese Stuffed Crust Thai Mexican Tomato Basil Pie Weight 31 29 31 21 26 25 47 29 26 22 Calories/slice 305 382 338 327 309 313 349 332 364 393 Fat grams/slice 9 16 14 9 10 11 13 10 17 19 make 1.51 1.53 1.51 1.2 1.23 1.28 1.23 0.9 0.85 0.8

Deb currently has a product line of 10 specialty pizzas. The data above shows overall population data for each line of pizza weight in ounces, cost of a slice to make in dollars, calories per slice, and fat grams per slice. Deb knows that many of her customers are weight conscious. She'd like to know if there a relationship between fat grams per slice and calories per slice. Is there? See the scatter diagram. Yes, there is a relationship. What is the coefficient of correlation? Is it significant at .01? How do you know? The correlation is 0.9185. H0: rho = 0 Ha: rho not equal 0 The test statistic is: r*sqrt((n-2)/(1-r^2)) 0.9185*sqrt((10-2)/(1-0.9185^2)) = 6.57 The t critical value (df = 8), two tails, is: t = +/- 2.306 The test statistic is well above that, so reject the null. The correlation is significant. How much of the variation in calories can be explained by the number of fat grams per slice? r^2 = 84.36% Build a predictive model for number of calories using fat grams. If a pizza has 15 grams of fat, how many calories should it have, using your model? The model is C = 7.815f + 241.17 Plugging in 15: 358.395 calories Can you think of any other correlations that might help Deb in her business planning? Are there any variables that seem to be missing?

correlati 0.918

slope 7.814 y inter 241.

variables that seem to be missing?

Avg. #Slices Sold per day 134 87 134 132 12 111 130 56 76 52 Profit/Slice 0.49 0.47 0.49 0.80 0.77 0.72 0.77 1.10 1.15 1.20

overall population alories per slice, and 20 15 Fat 10 5 0 0 100

Fat vs Cal

w if there a

200 Cal

300

400

correlation 0.918545584

ams per slice?

15 grams of fat, how

slope 7.814878893 y inter 241.1695502

nning? Are there any

500

You might also like