This lab assignment involves using inferential statistics in Excel to analyze yield strength data from two steel alloys, ALLOY 1 and ALLOY 2. Students will obtain confidence intervals and conduct hypothesis tests to compare the mean strength of each alloy to specifications, compare the mean strengths of the two alloys, and determine if a high-pressure/temperature treatment increased the strength of ALLOY 2 rods. The data is provided in an Excel file and students must state hypotheses, calculate and report test statistics and p-values, and discuss assumptions of the analyses.
This lab assignment involves using inferential statistics in Excel to analyze yield strength data from two steel alloys, ALLOY 1 and ALLOY 2. Students will obtain confidence intervals and conduct hypothesis tests to compare the mean strength of each alloy to specifications, compare the mean strengths of the two alloys, and determine if a high-pressure/temperature treatment increased the strength of ALLOY 2 rods. The data is provided in an Excel file and students must state hypotheses, calculate and report test statistics and p-values, and discuss assumptions of the analyses.
This lab assignment involves using inferential statistics in Excel to analyze yield strength data from two steel alloys, ALLOY 1 and ALLOY 2. Students will obtain confidence intervals and conduct hypothesis tests to compare the mean strength of each alloy to specifications, compare the mean strengths of the two alloys, and determine if a high-pressure/temperature treatment increased the strength of ALLOY 2 rods. The data is provided in an Excel file and students must state hypotheses, calculate and report test statistics and p-values, and discuss assumptions of the analyses.
This lab assignment involves using inferential statistics in Excel to analyze yield strength data from two steel alloys, ALLOY 1 and ALLOY 2. Students will obtain confidence intervals and conduct hypothesis tests to compare the mean strength of each alloy to specifications, compare the mean strengths of the two alloys, and determine if a high-pressure/temperature treatment increased the strength of ALLOY 2 rods. The data is provided in an Excel file and students must state hypotheses, calculate and report test statistics and p-values, and discuss assumptions of the analyses.
In this lab assignment you will use inferential tools available in Excel to compare the yield strength of two steel alloys. You will estimate the strength of each alloy with a confidence interval and apply a test of significance to see whether the observed mean yield strength exceeds the target value. Finally, you will compare the mean yield strengths of the two steel alloys. One-sample and two-sample inferences will be applied to the data. You will also discuss the assumptions necessary to make the inferences valid. It is recommended to get familiar with the Lab 4 Instructions before you start working on the assignment.
Comparison of Yield Strength of Two Alloys
A manufacturer of metal rods used in construction must make reasonably certain that the rods satisfy some strength specifications. If the strength requirement is not met, the production process is discontinued while adjustments are made. One recognized way of monitoring production is to take a random sample of several rods from each hour's production and to determine their average yield strength. When the production process is operating properly, the mean yield strength depends on various factors that may change from hour to hour. Random samples of size 30 will be taken every hour to determine any change in the value of the mean.
The manufacturer has just developed two new alloys, ALLOY 1 and ALLOY 2 and would like to see which of them has better strength qualities. In order to compare the yield strength of the two alloys, 30 rods made of each alloy were randomly selected and subjected to the strength test. The yield strength of each rod in the two groups was recorded.. After the strength test, the 30 rods made of ALLOY 2 were subjected to a combination of high pressure and temperature and their strength values were determined again.
The data are available in the Excel file lab1.xls located on the STAT 235 Laboratories web site at http://www.stat.ualberta.ca/statslabs/index.htm (click Stat 235 link and Data for Lab 4). The data are not to be printed in your submission.
The following is a description of the variables in the data file:
Name of Variable Description of Variable
ALLOY 1 The yield strength of rods made of ALLOY 1, ALLOY 2 The yield strength of rods made of ALLOY 2, ALLOY 2 +TREATMENT The yield strength of ALLOY 2 rods subjected to high pressure & temperature.
1. First you will obtain the summary statistics for each sample and you will obtain a confidence interval for the mean strength of each alloy.
(a) Use the Descriptive Statistics tool to obtain the summary statistics and a 95% confidence interval for the mean strength of each alloy. Paste the summary statistics into your report and report the 95% confidence intervals for each alloy. Use the summaries to compare the two alloys. Which of the two alloys has better strength qualities? Explain briefly.
(b) According to the specifications, the mean strength of each alloy is required to exceed 64 ksi. Is there any indication that the mean strength of either alloy is below the required threshold value of 64 ksi? Refer to the 95% confidence interval for each alloy to answer the question. Explain briefly.
1 2. Do the data provide evidence that the mean strength of each alloy exceeds the threshold value of 64 ksi? Now you will answer the question by carrying out the appropriate statistical tests.
(a) Carry out an appropriate test to check the above claimusing the data for each alloy. In particular, state the null and alternative hypotheses in terms of the population parameters, obtain the value of the test statistic, specify the distribution of the test statistic under the null hypothesis, and obtain the p-value of the test. What do you conclude? Notice that as there is no appropriate feature in Data Analysis to carry out the test directly, you will have to calculate the value of the test statistic and the corresponding p-value by entering appropriate formulas into Excel worksheet. Lab 4 Instructions may be useful in this part.
(b) What are the assumptions about the distribution of strength required to make the tests in part (a) valid? Would it be safe to use the inferential procedures even if some of the assumptions were violated? Explain briefly. It is not required to verify the assumptions with Excel.
3. In this part you will compare the mean strength of ALLOY 1 and ALLOY 2 rods. Do the data provide any evidence of a difference in the mean strengths of ALLOY 1 and ALLOY 2 rods?
(a) Answer the above question by carrying out the appropriate test in Data Analysis menu. Before you choose an appropriate test, you might refer to the output in Question 1 to decide whether the test with equal variances would be appropriate. State the hypotheses tested and the distribution of the test statistic under the null hypothesis. Report the value of the test statistic provided by Excel, the P-value, and your conclusion.
(b) What are the assumptions about the distribution of strength required to make the tests in part (a) valid? Would it be safe to use the inferential procedures even if some of the assumptions were violated? Explain briefly. It is not required to verify the assumptions with Excel.
4. The thirty ALLOY 2 rods were subjected to a combination of high pressure and temperature. In this question you will estimate the effect of the treatment on the mean strength of the rods.
(a) Do the data provide evidence that the treatment increased the mean strength of the ALLOY 2 rods? Answer the question by carrying out an appropriate test in Excel. In particular, define the null and alternative hypotheses; report the value of the test statistic and the p-value of the test provided by the Excel output.
(b) Use the Descriptive Statistics feature in Data Analysis menu to obtain a 95% confidence interval for the mean change in strength of ALLOY 2 rods after the treatment. You may create first a new variable EFFECT defined as the difference in strengths ALLOY 2 +TREATMENT rods and ALLOY 2 rods. Is the interval consistent with the outcome of the test in part (a)? Explain briefly.
(c) What are the assumptions necessary to make the test in part (a) and confidence interval in part (b) valid? Explain briefly.
(d) Is the effect of the treatment independent of the initial strength of the rods? In order to answer the question, obtain the plot of the variable EFFECT versus ALLOY 2 measurements. What do you conclude?
2 MARKING SCHEMA TO THE LAB 4 ASSIGNMENT
Proper Header and Appearance: 10 points
Question 1
(a) Summary statistics: 3 points 95% confidence interval for each alloy: 3 points each (6 points total) Comparison and indicating the better alloy: 3 points
(b) 95% confidence intervals and the threshold value of 64 ksi: 2 points each alloy (4 points total)
Question 2
(a) Test for ALLOY 1: 6 points In particular: hypotheses: 1, distribution: 1, test statistic value: 2, p-value: 1, conclusion: 1) Test for ALLOY 2: 6 points
(b) Assumptions: 2 points Validity of the inferences: 2 points Large sample size and Central Limit Theorem effect: 2 points
Question 3
(a) Test to compare ALLOY 1 and ALLOY 2 strengths: 6 points In particular: hypotheses: 2, distribution: 1, test statistic value: 1, p-value: 1, conclusion: 1.
(b) Assumptions: 2 points Validity of the inferences: 2 points Large sample size and Central Limit Theorem effect: 2 points
Question 4
(a) Test for the treatment effect: 6 points In particular: hypotheses: 2, distribution: 1, test statistic value: 1, p-value: 1, conclusion: 1.
(b) 95% confidence interval for the treatment effect: 3 points Consistency of the interval with the outcome of the test in part (a): 2 points
(c) Assumptions: 2 points
(d) Plot of effect vs. ALLOY 2 strengths: 6 points Comments: 2 points
Handbook of Marketing Scales - Multi-Item Measures For Marketing and Consumer Behavior Research-SAGE Publications, Inc-William O. Bearden, Richard G. Netemeyer