Abstract
Received 9 July 2009DOI:https://doi.org/10.1103/PhysRevSTPER.5.029902This article is available under the terms of the Creative Commons Attribution 3.0 License. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.
Highlights
In our paper reported previously in this journal, we explored how cluster analysis, a method from data mining used to find natural groupings in data, could be used to categorize the responses given by students on a free-response question about acceleration in two dimensions
We cannot assume that the results of the cluster analysis we reported are an accurate reflection of the information present in the data
The errors in our previous paper arose from two unsupported choices
Summary
In our paper reported previously in this journal, we explored how cluster analysis, a method from data mining used to find natural groupings in data, could be used to categorize the responses given by students on a free-response question about acceleration in two dimensions. In the process of preparing to expand on that work, we discovered that our analysis was both incorrect and incomplete. We were incorrect in that we used default settings in our software package and thereby misidentified the distance measure used. We have since determined that this distance was not appropriate for our data coding. We were incomplete in that we were not sufficiently rigorous in our definition of groups of student responses. We have corrected both of these issues
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have