JEDM 2014 - Special Issue on Educationla Data Mining with Longitudinal Data Sets

EDUCATIONAL DATA MINING WITH LONGITUDINAL DATA SETS
Special Issue Call for Papers: Journal of Educational Data Mining (JEDM)
Guest Editors
Ryan S. Baker, Teachers College, Columbia University
Alex J. Bowers, Teachers College, Columbia University
Aim of Special Issue: We invite paper submissions for a special issue
of the peer-reviewed Journal of Educational Data Mining that focuses
on using educational data mining methods to study longitudinal data
sets, data sets where a learner (or other individual or entity) is
tracked over a considerable period of time. Educational data mining
and learning analytics have been successful at making predictions
about learners within the course of a semester; a substantially
smaller amount of research has focused on tracking learners over more
extended periods of time, such as multiple years. In this special
issue, we call for research that uses data mining methods to model and
study educational phenomena that take place over the course of years
(work spanning multiple semesters may be considered in cases where
there are significant changes in learning experiences between
semesters). Research must involve a genuinely longitudinal data set
(panel or repeated measures data), with multiple time points involving
the same entity, and the use of data mining methods, to answer
research questions in the domain of education.
Sample topics for this special issue might include, but are not limited to:
? tracking student career goals from middle school to the world of work
? clustering student grade trajectories from elementary school through
high school, and using these trajectories to predict high school
dropout or college attendance
? predicting university donations after graduation, based on admissions records
? studying how course evaluations are influenced by gaining tenure or
promotion, or approaching retirement
? applying EDM techniques to examine longitudinal change at the
teacher, parent, administrator, or school level
? studying how changes in a discipline, or in state-level education
policies, lead to changes in syllabi in subsequent years
? studying the impact on student dropout over several years after a
change in university or district administration
? extending knowledge tracing to predict student preparation for
future learning, by predicting performance in a spring course from
meta-cognitive behaviors in a course the previous fall
Papers should apply accepted or novel educational data mining methods
in rigorous, demonstrably valid ways to study these topics. Articles
should be written in a style that is simultaneously meaningful to
experts in data mining and comprehensible for education researchers
who are unfamiliar with these methods. Data can be drawn from any
educational source (e.g. grades, course evaluations, syllabi,
admissions or enrollment records, interaction logs, questionnaire
instruments, field observations, video or text replays, collaborative
chats, discussion forums) so long as it supports valid inference.
Simulated data is not admissible for this special issue, except in
cases where it complements authentic data. An example of a case where
simulated data would be admissible is in a case where it is used to
show that if a pattern of some kind was in the real data it would have
been detected, and its absence is not simply due to sample size.
All papers must make a contribution to research in the domain studied
and must give full detail on the educational data mining methods used
to derive these contributions; it is not necessary, however, that a
paper make innovations in educational data mining methods although
these are, of course, welcome (so long as they are valid).
Review Process
As stipulated by JEDM reviewing guidelines, each submission will be
peer-reviewed by three experts in the field, including both members of
the JEDM editorial board plus reviewers chosen specifically for this
issue.
Submission Guidelines
We invite submissions of any length. Please see the JEDM submission
guidelines. All submissions should be made through the JEDM article
submission system.
Please note in your cover letter that your article is intended for
this special issue.
http://www.educationaldatamining.org/JEDM/index.ph...
http://www.educationaldatamining.org/JEDM/index.ph...
Deadlines
Please submit your manuscript by August 15, 2014.
You should expect to receive feedback and a decision by approximately
November 15, 2014.
Please direct questions to the guest editors at
ryanshaunbaker-AT-gmail.com and Bowers-AT-tc.edu