Jan 082013
 

One thing that I have struggled with is to make review meaningful for AP Stats. I absolutely hate the “here is the review, the test will be similar” wink wink, where all you do is change some numbers and voila, there is the actual exam.

That doesn’t seem honest to me, when I am supposed to be preparing them to think critically and learn material. It doesn’t seem honest for regular classes and it is doubly dishonest in AP classes. This year I have really been thinking and working on this issue more in my classes. To that end, this is my review project this year. 1 day in, and I think it is mostly successful.

I started off with a data set. I took it from my proficiency materials, stripped out anything that could be used to identify a particular learner, and then deleted 20 digits and put in blanks. The blanks were filled in using digits from the learners own student id number. Now, the learners can discuss HOW to do the problems, but they cannot get answers from anyone else. I can easily check their answers using JMP or another statistics program if I question their work. The nice thing, is almost everyone’s graphs will look similar, but not identical. Similar enough I can help w/o needing to do the entire process because I know what it is supposed to look like.

image

Now comes the questions. heh heh heh. 10, simple, innocuous questions.

For instance, here are the first 3 questions:

1. Describe the 3 variables completely, including types of variable, appropriate graphs of the values and complete description of the distributions, including all appropriate statistics. J K L

2. A passing score on the math test is a 242, while for the reading test a 300 is required. Is passing the proficiency exam in either math or reading independent of gender? Construct appropriate graphs to help explain why or why not. J K L

Create a bar chart of reading and math scores broken down by gender. Explain what the graph says about the pass rates of males and females.

The “JKL” at the end is the Wingding’s font for “smiley face” “straight face” and “frowny face”. They can rate themselves on each question.

The files are embedded here in both docx and pdf format if you would like copies.

How did it go, you ask?

I am giving them 3 class periods to work on it. I am glad, because after day 1, they accomplished most of question number 1!

I was stunned by them asking “How do I make a histogram?”. Like we didn’t do it 100 times at the beginning of the year.

Ooops, that is the problem. They learned it for the test, and promptly forgot it. Now I am forcing them to actually go back and relearn everything we have covered the entire year. They are cursing me, but it is working. I have had a lot of “aha’s” because they didn’t learn it well the first time, and now they are being forced to figure it out. It is working.

I wonder if I can’t teach the material this way to begin with? Can’t I give them a data set and say “Go”. We can figure out how to handle it together?

Makes a person go hmmm.

May 042011
 

The FRQ review that I did last class was fairly successful. The learners worked their problems, thought about how to teach that problem, and gave good presentations to the class. The one thing I was very disappointed in was the lack of questions on how to do the problems the class was teaching. They all said, “yup, I understand how to do it, that makes sense.”

Come on, as a teacher, I know that is b.s. They were just humoring their classmates giving the presentation on the problem.

Today I am giving them another MC practice exam. I am still mixing it up though. I numbered them off by 1,2’s. Now they all have an odd or an even number. The ones went to one side of the room, the evens to the other.  I gave them 5 minutes to work on as many problems as they can in pairs, then called “switch”. Then they worked more problems for 5 minutes before switching again. And again. And again.

When the class was half way over, I required them do a “switch different”. Now they were paired up 1-2 instead of 1-1, and they were teaching each other the problems they already did.

It is a modification of a think-pair-share but for AP Stats review.

First period went off terrifically! Second period also! At great day of reviewing multiple choice problems where everyone was engaged and focused on the problems at hand.

Apr 282011
 

This week is review. All review. It is as interesting and as fun as watching paint dry. It is also necessary. I have 90 minute blocks, so I can give them the full exam in two periods. The first period I gave them the multiple choice, ran it through the scanner, and handed them their results, answer sheets and the worked out solutions as they headed out the door.

Let’s just say they didn’t realize what 90 minutes of sustained effort on a Stats exam felt like until then.

The next time I saw them, I gave them the free response. Now I have to grade them all, so it will take the weekend for me to give them their results back, the effort will be worth it (I hope).

But … there is no way I can justify giving them another set of exams like that. Just working problems in a silent room will not be the best way to learn.

For the next class period’s review I am going to give them a card when the walk in. That card will group them into one of 6 groups. It will be random, so the question of favoritism can not come into play, and after the class starts I will assign the groups one of the 6 FRQ’s from 2009. I will try to give #6 to the strongest group. Each group will have 30 minutes to create a presentation to the class on what the correct answer is. 15 of the minutes will be without the answers, so they will have to actually work the FRQ honestly.  I will give them the answer packet after 15 minutes, so they can check their answer.

Finally, they will have to come up with (in the 30 minutes) a presentation to the class to TEACH the correct answer, including the language, graphs, and anything else needed.

My goal is to have the learners discuss the correct and incorrect answers, answer questions and become experts on the questions and then model the expert behavior for a more successful exam scenario in two weeks.

Let’ s see if it works. I will let you know. Monday will be the day I do this. It has to work better then me standing in front of the room telling them. How long will they retain that?