You will analyze data that shows some details of a sample of resale price of 600 GM
Question:
You will analyze data that shows some details of a sample of resale price of 600 GM CARS. Your data set was randomly resampled from a big data set that consist of more records. So, please be aware that each group will have its own data set. You will only focus on the relevant variables and answer each question.
The data includes:
NUM- The unique index number of each auto
Mileage – usage of the car (mileage in miles)
Price Price of each car sold
Make – The make of the car – Buick, Cadillac, Chevrolet, Pontiac, SAAB, Saturn
Model – The model of the car - Century, Lacrosse, Lesabre, etc
Trim - different versions of the same model with different features: Sedan 4D, CX Sedan 4D, Custom Sedan 4D, Limited Sedan 4D, Sedan 4D, DHS Sedan 4D, DTS Sedan 4D, etc.
Type – different car types: Sedan, Coupe, Convertible, Hatchback, Wagon
Doors – number of doors, 2, 4.
Engine Engines with 4 cylinders, 6 cylinders or 8 cylinders power.
Liter – Car engine size measures in liters. Engine size is the volume of fuel.
Sound – Car audio speaker where 1- upgraded speaker, 0-standard speaker
Cruise – Cruise control is a system with an automatic control of the car speed where 1 – there is cruise control and 0 – no cruise control
Leather – Car seat cover where 1 – there are leather seats and 0 – no leather seats
How the Project is graded
Your submission will be graded based upon the following factors: substance, presentation, accuracy, grammar and clarity. A demonstration of effort is the driving force of this assignment. Assignments will be compared to discern levels of effort and excellence.
As a minimum, your report must include the following:
- Title page: [1] title [2] submission date [3] group number and the file name of the data set used [4] names of each group member plus their student number, [5] course code (i.e.: QMS202) [6] Submitted to “Instructor’s name”
- Your project must be submitted online via D2L under Group Discussion.
- The answer to each question will begin on a new page. State the question (cut and paste).
- Cut and paste all relevant SPSS outputs in the write-up section at the bottom of your answer to each question. Do not send the reader to appendices to find them.
- A complete write up of your chosen hypothesis test must include your assumptions, analysis of results and your conclusions. You must use both approaches (critical value and p-value approaches) to make your statistical decisions.
- Not using the exact dataset assigned to your group will result in getting a zero mark for the project. If you use data from another group, both your group and the other group will receive a zero mark. The data for each group is for their group’s use only.
Group Size
This project must be done in groups with 2 to 5 members only. This means that the project report must be a result of team effort. It is your responsibility to find your group members online via D2L under Communication → Groups. Your instructor (or D2L) has already assigned you a group number.
THERE ARE 4 QUESTIONS in this project. The following table shows the naming convention for the data set for each group. Please state the dataset name on your project.
DATA ASSIGNMENT | |
Group # | Dataset |
1 | QMS202Group_1 |
2 | QMS202Group_2 |
3 | QMS202Group_3 |
4 | QMS202Group_4 |
5 | QMS202Group_5 |
6 | QMS202Group_6 |
7 | QMS202Group_7 |
8 | QMS202Group_8 |
9 | QMS202Group_9 |
10 | QMS202Group_10 |
11 | QMS202Group_11 |
12 | QMS202Group_12 |
13 | QMS202Group_13 |
IMPORTANT:
Each GROUP HAS ITS OWN unique Data Set. You have been assigned to a group with a specific group number by your instructor (or D2L). Your group is assigned to a data set consists of 600 records of the resale cars. Individual projects (teams of 1) are NOT permitted. Contact my graduate assistant at qms102@ryerson.ca if the remainder of your team decides not to submit a project.
Question 1(10 marks)
i) Use the variable "Price" from your data to construct the confidence intervals for the estimate of population mean “price” of SEDAN car, at both 90% and 95% levels. Interpret your confidence intervals.
ii) Did you make any assumptions when constructing your confidence intervals? If yes, which assumptions; if not, why?
Question 2 (10 marks)
Consider the claim that the average resale price of cars at the time the data was collected was equal to $21,000. Use the variable “Price” to test this claim. (Use the 10% level of significance).
Question 3(10 marks)
[i] Based on your data, is the mileage of used cars with 4 cylinders significantly MORE THAN the mileage of used cars with 6 cylinders? Test at the 3% level of significance. Note that SPSS only performs a 2-sided test.
[ii] Provide possible reasons why you should expect to find a significant statistical difference between the prices of these 2 groups.
Question 4 (20 marks) Note the mark difference.
[i] Based on your data, is there a significant difference in prices among the different Type? State all your hypotheses and conclusions clearly in the standard format.
[ii] What are your results and conclusions from the Levene test?
[iii] Conduct the Tukey test (at the 5% level of significance) and describe the conclusions you derive from this test.
[iv] If one assumed that the cost of resale cars was independent of Type of the car, then what choice should one have made a decade ago to maximize the resale value of these cars, based on this data? What pair of Types had the most significant difference?
[v] This is old data and current trends may differ. Research current Type preferences for GM cars and give your team’s advice and reasons as to what Type of GM car one should buy and NOT buy if the only objective is to maximize the resale value.
- SPSS PROJECT HINTS: avoid these pitfalls
- The most common and biggest error is to assign one question to each person and put all parts together. The outcome is almost always of very poor quality and receives a very low grade. Our exams & tests also assume that each of you is expert in all facets of this project. You must check each other’s work- and fully understand it. It is a TEAM effort.
- The 2nd most common error is to postpone the assignment so late that you do not have time to complete it. That is a sure way to do badly in this course.
- The 3rd most common mistake is to fail to monitor your team members. You must learn to manage teams and make sure that you have all of the data and reports at the same time.
Here are the other common mistakes:
- You misread the question
- You used the wrong test (e.g., Using a Z test instead of a t test)
- Your test was in the wrong direction (or H1 has > or < instead of ≠)
- The null hypothesis or the alternative hypothesis (or both) was wrong
- You came to a wrong conclusion
- You used the wrong data (or Incorrect inputs)
- Hypothesis missing μ or π or has the wrong one
- Ha contains one of {= ,≤ or ≥} OR Ho contains one of {>,<,or ≠}
- You used Sample data in your hypotheses
- Failed to check the requirements to use a test
- Misread p-values or comparison of p to α is wrong
- Reaching a conclusion, i.e., rejecting H0, when p > α
- There is no technical conclusion (or a wrong one)
- There is no managerial conclusion (or a bad one)
- This test is a one-sided test (not 2-sided)
- You must take ½ of the Sig value from SPSS for a 1-sided test
- You failed to state the problem and/or define the variables.
- A printout of your DATA IS MISSING! It had to be included!
- Missing LEVENE TEST of homogeneity
- Forget to discuss or check for normality
Applied Regression Analysis and Other Multivariable Methods
ISBN: 978-1285051086
5th edition
Authors: David G. Kleinbaum, Lawrence L. Kupper, Azhar Nizam, Eli S. Rosenberg