In my intro stats class, I spent the last two days covering graphic displays for qualitative data (bar graphs, pie charts) and quantitative data (stem & leaf, histograms). Because I am fortunate enough to teach this class in a computer lab, I devoted today's class to a StatCrunch activity day.
To start off, my students joined a StatCrunch group so that I could share the data set with them. I had a data set consisting of 61 individuals and 13 variables (some qualitative, some quantitative). Once they all had the data, I showed them how to create a bar graph for qualitative data for the class that the individuals were taking (8am or 11am). The students then had to create a bar graph for the birth month of these 61 individuals, and answer a couple of questions that called on them to interpret the graph they created.
We followed the same approach (I show one, they do another and use their graph to answer questions) for relative frequency bar graphs, pie charts, stem and leaf displays, histograms, relative frequency histograms, and creating a frequency distribution from a histogram.
I was really happy with the way it went, and we were able to finish well before the 50-minute class ended. My students learned that working with StatCrunch was really easy, and that StatCrunch was quite powerful. (Wait until they see how truly powerful it is!) I also thought it was a great way to review the different types of graphs, know when they are useful, and how to interpret them.
I'd recommend having the directions typed out (so students have a written record of these procedures). I'd also recommend creating your own data set, so you can not only show students how to download data into StatCrunch, but also making sure the data is relevant to your students. I created mine by posting a StatCrunch survey that my classes filled out. Finally, you should have a set of questions that calls on your students to interpret their graphs.
I have posted both the directions and follow up assignment in a Live Binders binder called Stat Files. If you'd like to join my StatCrunch group so you can access the data, be my guest. All I ask is that you drop me an email through the contact page on my website (georgewoodbury.com) so I can keep visitors and students straight.
George
Showing posts with label statcrunch. Show all posts
Showing posts with label statcrunch. Show all posts
Thursday, January 24, 2013
Thursday, August 23, 2012
Sampling Activity (with StatCrunch)
During the first week of my Statistics course we covered sampling techniques. Some of my students were getting tied up in knots over systematic sampling, specifically applying the formula to find the ith individual to include in the sample. I created an activity that helped my students to understand that the goal is to actually come up with the sample, as well as to help them with StatCrunch.
On the first day of class I had my students fill out a survey I created in StatCrunch, and I ended up with 76 responses. (If you'd like to see the survey or the data, just let me know.) I then printed out a sheet of the numbered responses for gender and age. My students were supposed to select a systematic sample of 10 students to estimate the percentage of female students and the mean age. They computed k (76/10 rounded down to 7), selected a random number p between 1 and k, and obtained their sample by including every 7th student beginning at their value of p.
They then repeated this task using random sampling. They had StatCrunch generate 10 unique values between 1 and 76, and used these numbers to determine which individuals were to be included in the sample. They again used these results to estimate the percentage of females and the mean age.
Finally, I had them load a data set from our StatCrunch group which contained 76 responses for smoking status, iPhone ownership, and this semester's costs for books and supplies. I then gave directions for how to use StatCrunch to sample from these columns, and they selected a sample of 20 students. They used these to estimate the appropriate percentages and mean expenses.
Overall, I was pretty happy with the results. My students seemed to have a better understanding systematic sampling. They saw that working with StatCrunch was pretty easy, and I even got them to create pie charts and compute the mean.
If you'd like a copy of the activity, as well as access to the survey/data, you can reach me through the contact page on my website.
- George
Math Instructor at College of Sequoias (Visalia, CA), Pearson Author
On the first day of class I had my students fill out a survey I created in StatCrunch, and I ended up with 76 responses. (If you'd like to see the survey or the data, just let me know.) I then printed out a sheet of the numbered responses for gender and age. My students were supposed to select a systematic sample of 10 students to estimate the percentage of female students and the mean age. They computed k (76/10 rounded down to 7), selected a random number p between 1 and k, and obtained their sample by including every 7th student beginning at their value of p.
They then repeated this task using random sampling. They had StatCrunch generate 10 unique values between 1 and 76, and used these numbers to determine which individuals were to be included in the sample. They again used these results to estimate the percentage of females and the mean age.
Finally, I had them load a data set from our StatCrunch group which contained 76 responses for smoking status, iPhone ownership, and this semester's costs for books and supplies. I then gave directions for how to use StatCrunch to sample from these columns, and they selected a sample of 20 students. They used these to estimate the appropriate percentages and mean expenses.
Overall, I was pretty happy with the results. My students seemed to have a better understanding systematic sampling. They saw that working with StatCrunch was pretty easy, and I even got them to create pie charts and compute the mean.
If you'd like a copy of the activity, as well as access to the survey/data, you can reach me through the contact page on my website.
- George
Math Instructor at College of Sequoias (Visalia, CA), Pearson Author
Subscribe to:
Posts (Atom)