Stat 1000: Tips for Assignment 3

Published: Thu, 02/23/12

 
Please note that my second midterm exam prep seminar for Stat 1000 will be on Saturday, Mar. 10, in room 100 St. Paul's College, from 9 am to 9 pm .  I am now ready to take registrations.  Please click this link for more information about the seminar and to sign up if you are interested:
Grant's Stat 1000 Exam Prep Seminars 

Join Grant's Tutoring on Facebook or follow Grant on Twitter.

Simply go to www.grantstutoring.com and click the Facebook and/or Twitter icons.
 
If you ever want to look back over a previous tip I have sent, do note that all my tips can be found in my archive.  Click this link to go straight to my archive:
 
Grant's Updates Archive
 
Did you miss my Tips on How to Do Well in this Course? Click here
 
Did you miss my Tips for Assignment 2? Click here
 
If you are taking the course by Distance/Online (Sections D01, D02, etc.), click here for my tips for your Assignment 3.
 
If you are taking the course by classroom lecture (Sections A01, A02, etc.), click here for my tips for your Assignment 3.
 
Tips for Assignment 3 (Sections A01, A02, etc.)
 
You will need to study Lesson 4: Density Curves and the Normal Distribution in my book to prepare for this assignment (that would be Lesson 2 if you have an older edition of my study book).
 
To learn Lesson 4 and do this assignment, you need Table A from the textbook.  That can be downloaded from the Resources section of Stats Portal.  Here is a direct link to download this table:
Table A
 
Questions 1 and 2 are very similar to the concepts I teach in questions 1 and 2 in Lesson 4.
 
Question 3: I teach statistics vs parameters right at the start of my Lesson 4.
 
Question 4: Study Lesson 4, questions 5 and 6.
 
Question 5: Although they don't specifically say so, use the 68-95-99.7 Rule.  See my questions 3 and 4.

 
The rest of the assignment is a good run-through of X-Bell curve concepts taught in the latter half of Lesson 4.
 
Question 5:
To determine if data is normally distributed, you should construct a Normal Quantile Plot.  If the data on a normal quantile plot looks like linear, then the distribution is normal.  If the data looks nonlinear, the distribution is not normal.  
 
To make a Normal Quantile Plot in JMP:
Open a "New Data Table" and enter the given data in Column 1.  Double-click column 1 and name it something like "Weight" and make sure the Data Type is Numeric and the Modeling Type is Continuous and click OK.  Now select "Analyze, Distribution" and select "Weight" and click the "Y, Columns" button and click OK.  You should now see a histogram and stuff.  Click the red triangle next to "Weight" and select "Normal Quantile Plot" to get the graph you want.  You can remove any of the other outputs if you wish by clicking the red triangle and deselecting the other outputs (such as Histogram Options, etc.) and/or clicking the blue triangles.
 
Tips for Assignment 3 (Distance/Online Sections D01, D02, etc.)
 
Study Lesson 4 in my study book (Density Curves and the Normal Distribution; in older editions of my book this is Lesson 2) to learn the concepts involved in Assignment 3.
 
To learn this lesson and do this assignment, you need Table A from the textbook.  That can be downloaded from the Resources section of Stats Portal.  Here is a direct link to download this table:
Table A
 
Questions 1 and 2 are just like my questions 5 and 6 in Lesson 4.  Note, a "cumulative proportion" is a left area on the bell curve.  Put another way, cumulative proportion is P(Z < z).
 
Question 3 is a good run through of X-bell curve problems that I teach in the latter half of Lesson 4.
 
For the JMP part of the assignment, here are some tips:
Open a "New Data Table" in JMP.  To copy and paste the data into JMP, in the toolbar at top select "Edit" then "Paste with Column Names".  Double-click the "gpa" column heading and make sure the Data Type is Numeric and the Modeling Type is Continuous, using the drop-down menus to fix that if necessary.  Double-click the "sex" column heading and make sure the Data Type is Character and the Modeling Type is Nominal, using the drop-down menus to fix that if necessary.  Click OK.
 
To get side-by-side boxplots: In the toolbar at the top, select Analyze then select Fit Y By X.  In the pop-up menu, highlight the gpa column and click the "Y, Response" button.  Highlight the sex column and click the "X, Factor" button.  Click OK.  You will then see a graph with a vertical array of dots for the males (1) and the females (2).  Click the red triangle next to "Oneway Analysis ..." and select "Display Options".  You will then be able to select "Box Plots" in the Display Options sub-menu.  They want you to remove the Grand Mean line, so click the red triangle again and select Display Options and deselect Grand Mean.
 
To get normal quantile plots: They want you to make a normal quantile plot for the males and a separate plot for the females.  In the toolbar at the top, select "Analyze" , then "Distribution".  In the pop-up menu, highlight "gpa" and click "Y, Column".  Highlight "sex" and click "By".  Click OK.  You now see separate histograms for the males and for the females.  Click the red triangle next to "gpa" for both the males and for the females, and select "Normal Quantile Plot" to get a separate normal quantile plot for the males and females.
 
A normal quantile plot checks to see if a sample's distribution appears to be normal.  If the data follows a normal distribution, the normal quantile plot will look like a rising diagonal line.  If the plot looks curved rather than linear, there is evidence the data is not normal.  However, none of that seems to matter in your question, all they ask is which students are clear outliers, which, personally, I think is much easier to see from your side-by-side boxplots than from your normal quantile plots.