You are on page 1of 13

Managerial Statistics WEB CASE

Click to edit Master subtitle style

4/28/12

Chapter 7
1). Data Collection procedures followed CCACC to form its conclusions are flawed, because theyve followed Convenience Sampling They should have followed Simple Random Sampling to make their conclusions more accurate. Also sample size used should be more than 5 units. 2) (a) Oxford Os mean weight: (360.4 + 361.8 + 362.3 + 364.2 +371.4 )/5 =(1820.1)/5 = 364.02
4/28/12 Oxford Alpine frosted flakes mean weight:

Parameters

Oxford Os 368 364.02 15 -0.59 -1.64

Alpine 368 368.04 15 0.006 -

(b)

Z Z(Critical)

4/28/12

3). It will be incorrect to conclude about the filling process of the two cereals, as the conclusion is drawn using biased samples of 5 units each.

4). Yes, the request is reasonable because allegations backed with ample evidence.

is not

5). Yes, such techniques can be used if we take an unbiased sample by using simple random sampling. In such scenario, if there is any cheating case it can be proved by CCACC.

4/28/12

Oxford Cereal chapter 9


The Null &Alternate hypothesis are as follows H0 : < 368 H1 : > 368 As depicted the p value 0.07<0.05 hence we cannot reject the null hypothesis The independent testing is thus valid and the results prove that oxford cereals are not cheating its customers
4/28/12

Oxford Cereal chapter 9


The confidence level for the mean weight of the box contents is 367.16 < < 373.71 This implies that the contents of the box could weigh anything up to 373.71 grams,hence Oxford cereals claim that many boxes contain more than 368 grams is not surprising and is true. Possibility that the public experiment and the CCACC claim can both be right exists 4/28/12 because the sample size taken in both cases could have been

Chapter 10- OXFORD CEREALS


Q. On a recent purchase of a new sample of cereal boxes--ten produced at Plant #1 and ten produced at Plant #2. It was discovered that the average weight of Plant #1 boxes was 372.44 grams, but that the average weight of Plant #2 was 365.64 grams, less than the 368 grams claimed. Even after the recent public experiment about cereal box weights, the CCACC remains convinced that Oxford Cereals has mislead public. 1. 2. Do the CCACCs results prove that there is a statistical difference in the mean weights of cereal boxes produced at Plant# 1 and 2? Perform the appropriate analysis to test the CCACCs hypothesis. What conclusions can you reach based on the data?
4/28/12

Solution
Samples are randomly and independently drawn. Populations are normally distributed Population variances are unknown If population variances are equal, the pooled-variance t test is appropriate Here, use a separate-variance t test, which includes the two separate sample variances in the computation of the test statistic We have 2 sets of data where the variances are unknown, so we have 2 options here. We can either apply POOLED-VARIANCE t-test or SEPARATE VARIANCE t-test. So we will apply F-test here to decide for the2same.

S1 F= 2 S2

4/28/12

DATA
Plant 1 375.28 368.19 356.98 366.1 379.84 365.51 376.39 398.66 353.09 384.37 Plant 2 364.51 367.95 368.62 370.25 345.45 359.36 379.64 378.71 358.14 4/28/12 363.74

F-test
F-Test Two-Sample for Variances Variable 1 372.441 180.8843 10 9 1.787974 0.199863 3.178893 Variable 2 365.637 101.1672 10 9 Mean Variance Observations df F P(F<=f) one-tail F Critical one-tail

From the F-test we can see that variances are different, so we will apply separate variance t-test. 4/28/12

SEPARATE VARIANCE t-test


t-Test: Two-Sample Assuming Unequal Variances Variable 1 Variable 2 Mean 372.441 365.637 Variance 180.884277 101.167201 Observations 10 10 Hypothesized Mean Difference 0 df 17 t Stat 1.2811504 P(T<=t) one-tail 0.10867112 t Critical one-tail 1.73960673 P(T<=t) two-tail 0.21734224 4/28/12 t Critical two-tail 2.10981558

Interpretations
Two-tail test: Hypothesis: H0: 1 = 2 H1: 1 2 i.e., H0: 1 2 = 0 H1: 1 2 0 From the results, we can see that the t-value is less than 4/28/12 the t-critical, so we can accept our hypothesis.

THANK YOU.

4/28/12

You might also like