Posts
At the beginning of October, Dr. Grabowski gave us an opportunity to look at the ASTRO model we had created. We were asked to pick variables we would like to find and study the relationship. I chose Attitude toward Science and Course, Students’ Performance in Astronomy, and Students’ Degree of Engagement. If I were to study these variables, I would like to find the answers for these questions:
- Is there correlation between Attitude toward Science + Course and Astronomy Performance?
- Is there correlation between Attitude toward Science + Course and Students’ Degree of Engagement?
In addition to the relationship of variables, we were also encouraged to use existing data to run our research questions. I found this activity very helpful and useful for me. This activity gives me a chance to practice analyzing the data. I am hoping to have my first statistic finding before the end of this semester.
In October, my team and I spent most of our time analyzing ASTRO discussion and preparing for the E-Learn conference in November. From trial and error, I have learned a lot about how to use NVivo to do the data mining for qualitative analysis. I have completed the analysis and am waiting anxiously to see what others’ findings would be. One thing I found when running queries on NVivo is that it doesn’t seem like I would be able to run the report in the statistic data, such as comparing the number or percentage of Positive Attitude toward Course and Negative Attitude toward Course. This is probably one of the issues that I am planning to do more research.
Our plan for this analysis is to compare every rater’s findings by using the process called “inter-rater reliability”. From this activity, I expect to learn how this “inter-rater reliability” can be done. I believe in my heart that this will strengthen my research skill and empower and help me when I am doing another, as well as my, research.
On October 27th, 2008, it was the day of the first teaching opportunity. Yu-Hui presented her teaching topic in “Data Screening”. This is, once again, a very interesting and fascinating information. I have learned a lot, in details, about four different concepts of data distribution: (1) normal distribution, (2) Skewness, (3) Kurtosis, and (4) Outliers. More interestingly, Yu-Hui thoroughly delivered the key information of the processes for data screening that researchers should keep in mind: (1) ways to inspect the accuracy of input, (2) check for missing data, and finally (3) how to deal with the missing ones. This will absolutely help me in the future when I am analyzing the data.
