Looking at my data I have received two very different results. For my English Pre-Test students read “Tar Beach” which is a fiction story at the current 3rd grade level of 3.3. For the Post-Test they read “Tacky the Penguin” which was also at their current AR level of 3.3. The results showed not to be statistically significant. According to the T-Test Confidence interval Pre-Test minus Post-Test equals 4.44. The mean for the English Pre-Test was 88.89 and the mean for the English Post-Test was 84.44
However, the results for the Spanish Pre-Test and Post-Test were considered to be extremely statistically significant. The mean Confidence interval for the Spanish Pre-Test minus the Post-Test was 30.00. The overall mean for the Spanish Pre-Test was 77.78 and for the Post-Test it was 47.78. Looking at my English results I noticed there wasn’t much difference in the reading comprehension test results from the pre-test to the post-test. There was a significant difference in the reading comprehension test results in Spanish; in which students did not do so well in the post-test compared to the pre-test. Looking at the AR test results I noticed that the two Spanish books the students read were at the same AR level of 3.3, however, the AR quiz for the pre-test only had 5 questions and the post-test had 10 questions. I did take a look at the I-Station in English and Spanish and in English students are more interactive with games that have to do with the passage they just read. In Spanish I-Station has less interactive activities for students to do to answer comprehension questions from the passages. I asked my students verbally if they enjoyed using I-Station and why or why not. The responses I received were that they did like the English version of I-Station better than the Spanish version because they got to play more games in the English version.
1 Comment
Helen Blood
12/10/2017 07:23:41 am
I really like your research project, as I use Istation in my classroom, hoping to reach those lower readers. I really do not think that the 5 question-10 question thing should have resulted in such different scores. However, I wonder if the types of questions in AR were significantly different, and what was different about the text that AR decided it needed to have different amount of questions. I do know that we can create our own questions for a book, this might be a way to get more reliable scores. Thank you, your style of writing is very clear, and makes me curious.
Reply
Leave a Reply. |
Archives
January 2018
Categories |