Statistics Online T. Bartos & Dr. G. Ströhlein Statistics Online Statistics Online Effect & process evaluation of an Effect & process evaluation of an integrated learning environment integrated learning environment
Jan 03, 2016
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Statistics OnlineStatistics OnlineEffect & process evaluation of an Effect & process evaluation of an
integrated learning environmentintegrated learning environment
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Overview
selection of results
design of the study
concept of the learning environment
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Starting point
integrated learning environment (ILS)
variety of online applications high degree of standardization
concerning
concepts & methods in statistics
study course & tutor marked assignments
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Components of Statistics Online
ILS
text of the study course (concerning statistics) interactive exercises (partly graphic
based) glossary (in german &
english)
CSCL via BSCW
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Evaluation design (sequence)
media-mix offered FINAL TESTConventional Learning and PreprarationCG
PRETEST
3 months
PGCourse
ILS CSCL
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Design (variables)
dep. V media use
ind. V score in final test
interv. V pre-knowledge in statistics computer literacy
ability of selfregulation study motivation attitude towards statistical methods attitude towards computers time spent overall and media speciffic time slice
age gender learning conditions (e.g. computer equipment, disturbances, …
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Selected results
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Drop out ratio in the CG was about 4 times higher than in PG
signed up for final test207 students
participants of evaluation98 students
comparison group109 students
final exams76 students
37%
drop out22 students
10%
final exams61 students
29%
drop out48 students
23%
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Scores in final test were significantly higher in PG vs. CG
1 2 3 4
10%
20%
30%
test score in quartils
Comparison Group
Participants
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Scores in final test (subgroups of PG)
ns *
**
Course N = 17
20
80
120
PGs
ILS N = 26
CSCLN = 34
40
60
100
test score
* p < .05
** p < .01
Statistics OnlineT. Bartos & Dr. G. Ströhlein
ns *ns ns
ns
Analysis on different levels of educational objectives
Course N = 17
PGs
ILS N = 26
CSCLN = 34
recall tasks
10
12
14
16
18
Course N = 17
PGs
ILS N = 26
CSCLN = 34
transfer tasks
10
12
14
16
18
*
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Fun and estimated usefulness of learning with study course
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5
135791113151719212325272931333537394143454749515355
days to final test
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5
135791113151719212325272931333537394143454749515355
days to final test
N = 65
Statistics OnlineT. Bartos & Dr. G. Ströhlein
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5
135791113151719212325272931333537394143454749515355
days to final test
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5
135791113151719212325272931333537394143454749515355
days to final test
Fun and estimated usefulness of learning with the ILS
N = 65
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Summary of results
the trend over the oberseved period of time concerning the related
„fun“ learning with the ILS is positive (no news-effect)
intervening variables did not show any effects (over all learning time, computer literacy, study motivation, etc.)
score in the final test was significantly higher in PG vs. CG
course dropout in the CG was about 4 times higher than in PG;
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Limits of our Evaluation
generalisability of results should be proved within other contexts;
general problems of measuring ‚added value‘ of neo-constructivistic learning by a conventional final test with multiple-choice and open answers;
limits of measuring all important factors influencing score in the final test.
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Essentials
Interactive learning environments are
more efficient in combination with
elements of cooperative learning.
Statistics OnlineT. Bartos & Dr. G. Ströhlein
Links and downloads
www.ao-psychologie.de