University Student Used JMP in Research That Won Prize
Apr 29, 2010 9:23 AM
Until recently, Emma Lookabaugh had never even heard of JMP. After using it for the first time this spring in an award-winning research project, she has become a fan.
Emma is a senior at NC State University in Raleigh, North Carolina, not far from JMP offices at SAS headquarters. Hailing from Goldsboro, North Carolina, she will earn a double major in plant biology and biological sciences and a minor in genetics in May when she graduates. But she won't be leaving the area just yet. She begins her graduate studies in plant pathology at NCSU this fall.
Emma used JMP as part of her undergraduate honors research project and presented her research last week in a poster at a symposium at NCSU. Nearly 300 students presented research at the NCSU symposium, and 28 won prizes. Emma's poster won in the Honors Teaching category. Scientific research society Sigma Xi recognized the winners at its annual banquet this week, and she received an Undergraduate Research Award.
I asked Emma to tell me about her winning research project and her impressions of JMP.
Arati: Can you summarize the research you presented on your poster?
Emma: I did an honors teaching project for my honors research requirement. I worked as a teaching assistant for Dr. Tom Wentworth's ecology course in the fall semester. We wanted to measure the effectiveness of supplemental learning materials (self-tests and active learning exercises) combined with attendance and study habits on learning outcomes (as measured by their average exam scores).
We also wanted to measure the potential for a particular ecology course to be converted into a distance education course in the future. We compiled data from survey questions that addressed our four main topics of interest (study habits, completion of active learning exercises, completion of self-tests and attendance). We then used an analysis of variance with average exam score as the dependent response variable and the four topics of interest as our categorical explanatory factors.
After running the model, we discovered that attendance and completion of active learning exercises had no effect of average exam scores, whereas completion of self-tests and time spent studying had statistically significant p-values. Time spent studying and completion of self-tests are both types of "independent learning activities," which are a vital component of distance education. Also, the fact that attendance had no effect on exam scores reinforced the idea that this ecology course would be a great distance education course. We also looked at other variables (like GPA and major codes) to see if we could find any patterns among the students' overall performance.
Arati: Why did you decide to do this research?
Emma: I really wanted to get out of the laboratory and do a teaching project. I met with my mentor to discuss options, and we talked about some different things that could be interesting or provide some sort of concrete outcomes. We spent a lot of time creating a comprehensive website for the students in addition to creating all of the active learning exercises and self-tests. We wanted to figure out if all of the extra time we spent making the course more student-friendly actually had any benefits to the students. A lot of the information we obtained throughout this study will be very valuable in future course planning.
Arati: What is your experience with statistics and statistical software? For instance, have you taken a stats course?
Emma: I have not taken a statistics course since AP statistics in high school. I had virtually no experience with any statistical software. (I have used some of the stats applications in Excel.)
Arati: What did you think of JMP?
Emma: JMP was very user-friendly. I had never heard of JMP until I walked into [statistics professor] Dr. Consuelo Arellano's office for a statistical consulting appointment. She mentioned JMP, so I went home and downloaded one of the licenses through the university.
The next meeting she showed me the basics of the program (how to join tables and transfer data sets from Excel), and then she showed me some of the basic tools like the Fit Y by X tool, Distribution analysis and Fit Model. I basically used the point-and-click method of learning. I didn't have an instruction book, and with no prior experience, I was still able to become comfortable using the software.
I was very impressed with the ease of use and the ability to use more sophisticated tools (again comparing to basic Excel). I also loved how easy it was to join large tables and not have to worry about messing up the matches whenever you wanted to sort rows. I loved the output from JMP. It gave us very nice graphs and images for the poster. I can honestly say that I will no longer use Excel for any of my statistical needs because I find JMP far superior.
Arati: What did you use JMP for in your poster?
Emma: We used JMP to compile all of our data. I used the JMP model for the poster because it seemed easier to understand to a non-statistics person. On the poster, we had the model prediction plot with the effects tests table, and we showed the LS means plots for our model effects. We also used a JMP Distribution histogram.