EDET 636 Impact of
Technology on Students
Week 8 Blog
Essential Question: How can data mining assist you in triangulating
your research findings?
Data mining
is a method of research that finds useful types of data and looks at the
patterns in the data to produce useful knowledge. Ma and Capri share that the
techniques of research and interpreting data included in data mining are
“decision trees, neural networks, rule induction, machine learning, and graphic
visualization.” (2014, pg. vii) Martínez and López say that data mining
requires a large amount of data “to extract the relationship between
variables.” (2017, pg.41) With this said, I don’t know if I will be able to use
data mining in my research because I will not be able to produce enough data
and with the research I have done, I can’t find data this is directly connected
to my action research.
Through an
example of research done in Mexico that required the use of data mining
presented by Martínez and López, I can dismiss the variables of student
engagement that are the least explanatory. Then I identify what variable might
affect the student engagement and mess with the variables underneath that and
see what patterns that I find. This should be repeated with every variable at
the higher level. When it is all repeated and finished, then I look at what set
of variables have a higher correlation of a positive impact on student
engagement. (2017, pg.45)
UPDATE: I
have been collecting data through the IXL math program. It provides the amount
of time in each skill and the amount of questions answered correctly and
incorrectly. I have not given an assessment of the topics learned yet but I
will be doing this next week. This will test the retaining of skills being
learned through the program. With the data collected so far, I have seen more
engagement in the Math Review but not as strong of understanding the concepts.
Next week, after the assessment, I will be changing the format of this online
RTI by providing a mini-lesson before and after the time allotted.
Resources:
Ma, X., &
Capri, H. L. (2014). Data Mining : Principles, Applications and
Emerging Challenges. Hauppauge, New York: Nova Science Publishers, Inc.
Martínez Abad,
F., & López, A. C. (2017). Data-mining techniques in detecting factors
linked to academic achievement. School Effectiveness And School
Improvement, 28(1), 39-55. doi:10.1080/09243453.2016.1235591
Mariah,
ReplyDeleteI love finding the patterns in recorded data. I think it helps us see what is working and even more important – what is not working. You have a great realistic view that you are able to evaluate that data mining is not something that you think you can incorporate into your project. I am often too quick to say, yes I can do.
Josie
I know it's difficult to conduct any statistical analysis with a small sample size, but I have read some research that was qualitative in nature that reported findings with SIX students! So you can be assured that you can report your findings and just note that the sample size is very small to be statistically significant, but you can at least go through the "motions" of analysis. I am very lucky to have a good sample size, and a large historical sample size to make some comparisons, otherwise, I would be in the same situation as you! Be strong and make some inferences. :-)
ReplyDeleteLook for attitudes and behaviors as well as work. Students can show what they understand by how they behave when working. Just a thought.
ReplyDeleteGreat points! I agree that I felt like I have too small of a sample to make data mining effective. But after reading other comments, perhaps it isn't so farfetched afterall!
ReplyDelete