RE: random sampling

>What kind of questions do you answer before they start?
I answer any and all questions from each group, rather than as a class.  I only had one group who asked meaningful questions like "what is red" and "is our suggested method a good method."  A number of groups didn't ask any questions.

>Can you elaborate a bit more on what you do after the students collect the data?
The students write up their findings and present them to the class using the Elmo to show their work.  I don't respond until after all groups have presented (maximum 5 minutes each, usually 2-3 minutes).  It is interesting because usually each group had at least one "good" component to their research and at least one "bad" one.  But between all the groups, we can piece together a good research study.

>I'm guessing you write the results on the board and discuss the sampling techniques but do you also lead them into what is good and not so good about each one?
I then discuss the bigger concepts such as defining the variables in a study and sampling methods.  There isn't much time left at the end of the class, but we address the basic methods of sampling and plusses/minuses with each method.  Their second day assignment is to read chapter one and address each bold print term both by the definition in the book and with reference to the library assignment.  Each group member is to individually develop a refined process for the library assignment.
At the beginning of the next class, I allow 5-10 minutes for each group to collaborate and decide on the process for the library assignment.  We then allow 2-3 minute presentations from each group.
It's in this second day that we nail down the sampling methods.  The first test has a number of questions from the first chapter, and few students missed any of those questions (which was a pleasant change from the norm).

>Do you give a grade for this?
No, I don't give a grade for this or any other classroom activity.  And, interestingly enough, no one asked about it.  The biggest concern they had was that I didn't know the "true" proportion of books in the library.  They wanted to know if they were right.  Which was a nice lead-in to inference from a sample to a population when the population parameter is not known.

>Also, do you have any handouts from Roxy's session you can share?
I am sure I have the handouts from her session.  I will look for them.

Tim
--

Tim Chappell
Mathematics Instructor
MCC-Penn Valley
3201 S.W. Trafficway
Kansas City, Missouri 64111
P 816-759-4215
F 816-759-4611
 


>>> "Moss, Cindy" <moss@...> 4/26/2008 2:08 AM >>>
Hi Tim,
 
This sounds like a great activity.  Can you elaborate a bit more on what you do after the students collect the data. I'm guessing you write the results on the board and discuss the sampling techniques but do you also lead them into what is good and not so good about each one? What kind of questions do you answer before they start? Do you give a grade for this?
 
Also, do you have any handouts from Roxy's session you can share?
 
Thanks,
Cindy

________________________________

From: Tim Chappell [mailto:Tim.Chappell@...]
Sent: Fri 4/25/2008 9:45 PM
To: mathedcc@...; Moss, Cindy
Subject: Re: random sampling



I picked up a great first day activity from Roxy Peck at the last AMATYC conference that might help.  I took roll at the beginning of the class the first day, and gave them this assignment.

Visit the college library and determine the percentage of red books in the library.  I will answer any and all questions you have before you go to the library.  Form groups of 3-4; make sure that at least one person in your group is familiar with the library.  You have 15 minutes in the library, and you may spend 15 minutes back in class to prepare your findings.  You need to develop your methodology before you leave so that you may quietly gather your data.

The activity was everything that Roxy said it would be and more.  I use Bluman's text for the course, and the first chapter is packed with fundamental topics.  This activity served as a springboard for many of those topics, including sampling methods.  Some students did cluster sampling, and at least they were wise not to pick the reference section.  Some students did systematic samplings by choosing every kth book on each shelf.  Some students did stratified samplings by splitting up and selecting books on each level of the library.  Some students asked library personnel for their opinion.  Some other insights were in defining what color is red, what percentage of the book had to be red, interrater reliability when students split up to cover more area quicker.  Students had not read the first chapter, but intuitively discovered many of those topics.  They seemed quite pleased when I used their group's findings to illustrate a topic.

I assign data projects throughout the course, and I have seen better studies with better analysis this semester than any other.  I attribute the improvement to this initial activity.

Tim
--

Tim Chappell
Mathematics Instructor
MCC-Penn Valley
3201 S.W. Trafficway
Kansas City, Missouri 64111
P 816-759-4215
F 816-759-4611



>>> "Moss, Cindy" <moss@...> 4/25/2008 3:12 PM >>>
I give a project to my statistics classes (online and face-to-face) where they have to challenge a claim by collecting data and analyzing it. The problem I am having is educating the students about random sampling. Almost all of them think that if they "say" they are collecting a random sample that it is random. I want them to come up with some sort of predetermined plan to collect data then realize that random sampling is almost impossible to achieve. What I used to do is give them an example of asking every 3rd (or so) person walking by them at a certain place on campus but then everyone used that exact same method and never even thought of the flaws in this system. Now I do not give them an example because I want them to think about it and not take randomness for granted.

Does anyone have any handouts or know of a text that discusses random sampling? I use Triola but there is only brief information on the definitions of various methods and flaws in some of them. It does not give suggestions on what randomness looks like.

Thanks,
Cindy Moss
Skyline College



****************************************************************************
* To post to the list: email mathedcc@... *
* To unsubscribe, email the message "unsubscribe mathedcc" to majordomo@... *
* Archives at http://mathforum.org/kb/forum.jspa?forumID=184 *
****************************************************************************