MacLin's Advanced Statistics Blog

This is a site for our advanced (graduate) statistics class...

Wednesday, August 30, 2006

SPSS Student Version

Hi Dr. MacLin, I am interested in purchasing the SPSS statistical software package. ...How much did you say that the student price was for the package, and where can I go to get it? Thanks!

Here is the URL to a campus site with will give you a little more information about where to buy the "Grad Pack" for SPSS - it is about $200 and well worth it.http://www.csbs.uni.edu/tech/spss.htmlHope that helps,--Dr. M

Homework 2.25 to 2.59

Exercises 2.25 to 2.59 are due on Thursday - you can post a comment if you have any questions. If you cannot get in sign a password - just check the anonymous button to make a comment.

Monday, August 28, 2006

Sabin 227 Free Hours

Question: How do I find out when the Sabin 227 room is free so my graduate students can use it if they need to?

Response: Here are the classes in Sabin 227 this semester:
http://www.csbs.uni.edu/tech/labs/schedule/request/classes.txt
Monday - Thursday we have open hours 9am to 10pm, other than the scheduled
classes and other one-time reservations:
http://www.csbs.uni.edu/tech/labs/schedule/reserved.asp
On Friday it is 9am to 5pm. Sabin 223 has no classes, so is only subject to
reservations.

Class Mailing List

We are using a mailing list that I received from the registrar that I use to send important emails to the class. If you are not receiving emails, please send me your email address so I can add it. Also, if you have an email address that you check more often than your UNI account, please let me know so I can use that instead.
Take care,

--Dr. M

Sunday, August 27, 2006

Adding to the blog

Advanced Statistics

I wanted to add to this blog so I clicked on the "Blog this!" button on the top of the page. I made sure to make a meaningful title, and then I added this text.

Chapter 2 Homework

Hi Class,

I met with some students yesterday about the exercises due for Tuesday - they we doing a great job except for a couple of the questions where Howell was asking them to make graphs that SPSS does not necessarily do (ex overlay of histograms). Remember Howell is not a book on SPSS so he is going to ask you to think that relate to what he discussed in the chapter...
One thing you could do is to generate the two histograms in SPSS and then draw the overlay by hand using the SPSS output - or for that matter take one of the graphs from SPSS and then hand draw the other one over it (sometimes you have to get creative - and there is always the do-it-by-hand option).
One of the other tings that came up was to make a back to back stem-and-leaf display. The book tells you how to do this, but anyone in their right might mind would rather have SPSS do this. So how do you get SPSS to do the labor of constructing the display (Hint: Analyze > Descriptives > Explore) but you will find that Explore won't do back to back. Again with the histograms, one could have SPSS generate the displays and then make the back to back by hand using the SPSS output. Or you could take one of the displays and hand draw the second display onto the output. Get the idea.

In the real world you might have a professor, a boss, a reviewer, etc. tell you to make a back-to-back stem and leaf display - you can't just say SPSS won't do it - You could 1) talk them out of it by recommending a different way to display the data, 2) you could go back to the books and figure out how to do one entirely by hand, 3) you could tell the editor that the reviewer is being unrealistic and risk loosing the publication, 4) put something together giving your best shot and tell them it is the best you can do, how do they recommend proceeding, 5) phone a friend and ask how they did it, or 6) ...?

Depending on the circumstances and the consequences of not producing the desired display - there are a number of outcomes, but the bottom line is to give it your best shot and then move on to the next thing to get done.

As we move on in the class and you advance your knowledge of statistics and as you gain more experience with your computerized tools (SPSS, Excel, etc.) you will get better on sorting these sorts of things out.

I hope this helps,

--Dr. M