Does Ict Improve The Efficiency of Learning?: Sciencedirect

Available online at www.sciencedirect.
com
ScienceDirect
Procedia - Social and Behavioral Sciences 177 (2015) 306 – 311
Global Conference on Contemporary Issues in Education, GLOBE-EDU 2014, 12-14 July 2014,
Las Vegas, USA
Does ICT Improve the Efficiency of Learning?

Martin Misuta*, Milan Pokornyb
a
Department of Applied Informatics, Faculty of Economic Informatics, University of Economics in Bratislava, Dolnozemska cesta 1/b, 852 35
Bratislava, Slovakia
b
Department of Mathematics and Informatics, Faculty of Education, Trnava University, Priemyselna 4, 91843 Trnava, Slovakia
Abstract
This paper deals with the efficiency of blended learning in selected courses, within the mathematics and computer science
curriculum at Trnava University, Faculty of Education. The experiment, as part of a quality project, lasted for two years, when
the subjects were taught by a combination of e-learning and traditional method. The complex solution covers a model of blended
teaching, which includes the on-line learning environment, assessment with the support of technology, verification of the model
by implementation in selected courses, adaptation of the teaching model and learning environments based on the results of
experiments, as well as development of recommendations and application for other subjects of study at Trnava University in
Trnava. We present a comparison of the results of pre- and final tests in both years for three selected courses, as well as the
comparison of students’ overall performance, keeping in mind the improvement of teaching efficiency and assurance of quality.
©©2015
2014TheTheAuthors.
Authors. Published
Published by by Elsevier
Elsevier Ltd.Ltd.
This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the Scientific Committee of GLOBE-EDU 2014.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: ICT; e-learning; blended learning; efficiency
1. Introduction
E-learning as a term refers to a variety of different forms of technology-supported learning, usually characterized
as the application of knowledge, information and educational technology to link people to each other and / or with
educational resources, for the purpose of education (formal or informal). (Ehlers & Hilera, 2012) Blended learning
can be defined as the combination of face-to-face formats and web-based formats. Furthermore, the use of e-learning
* Martin Misut. Tel.: +421- 2 6729 5859; fax: + 421-2 -6729 5182
E-mail address: martin.misut@euba.sk
1877-0428 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
doi:10.1016/j.sbspro.2015.02.346
Martin Misut and Milan Pokorny / Procedia - Social and Behavioral Sciences 177 (2015) 306 – 311 307
as an important element of blended learning provides the appropriate tools to support students´ generic skills.
(Kashefi, Ismail, Yusof, & Rahman, 2012) It is believed that utilization of e-learning courses in educational process
in the form of blended learning improves the study results of students. This idea was examined in several studies
focused on e-learning and blended learning, for example (Bargagliotti, Botelho, Gleason, Haddock, & Windsor,
2012; Kashefi et al., 2012; Misutova, 2009; Pokorny, 2012; Pyzdrowski, Butler, Walker, Pyzdrowski, & Mays,
2011; Rajini, Tiwary, & Ganapathy, 2011; Ramakrisnan, Yahya, Hasrol, & Aziz, 2012).
Partial results of the project, ”Rationalization of Education at Trnava University in Trnava" are described in this
article. The whole project focuses on development and pilot testing of methods and tools continually helping to
improve the quality of teaching and learning, (Misut, Pribilova, Orolinova, & Kotulakova, 2013) while education is
considered in various combinations of traditional and technology-supported learning assuming a decisive share of
blended learning i.e. combination of traditional teaching with e-learning. A number of activities, procedures and
steps, directly or indirectly affecting the quality of education and covering not only education process, but also all
related areas, such as university-related legislation, financing, organization and management of education, technical,
human and other information assumptions, has been tested during the project.
The influence of the teaching environment improvement in blended learning as well as students´ attitude and
proposal for learning environment improvement are presented further.
The rest of the paper is organized as follows: research context and method is presented in the second section,
gained results are in the third section. The fourth section interprets results.
2. Research context and method
The main goal of the conducted experiments was to find answers on the following questions: (a) have the
improvements of learning environments had any influence on students´ performance, (b) if improvements of learning
environments caused a change in students’ performance, is this change the same or similar in all courses?
To find relevant answers, we need to compare the quality of students enrolled in courses in subsequent years, as
well. It was necessary to fully understand the reasons for the change in students´ performance, if any occur.
2.1. Participants
The participants in this study were 67 students enrolled in the project in the 2012/2013 academic year and 79
students enrolled in the project in the 2013/2014 academic year. Students’ results in three courses: Discrete
Mathematics, Database systems, and Operating systems and Computer Architecture in subsequent academic years
are discussed further.
2.2. Tests
Students took the entrance tests at the beginning of the course and the final tests at the end of the course to
provide a standardized pre- and posttest measure of performance. Entrance and final tests tested the same knowledge
and skills within the course in both academic years. Tests contain a different number of items for each course, but
were identical for the courses in subsequent years.
2.3. Data Collection
All students enrolled in Discrete Mathematics, Database systems, and Operating systems and Computer
Architecture, were given the test twice. The scores from final tests were used to evaluate students’ points for the
course, scores from pre-tests served only as a base for comparison of students’ advancement. The first administration
of the test was given to all students before they started the course. The second administration of the test was given to
students after they finished the course.
308 Martin Misut and Milan Pokorny / Procedia - Social and Behavioral Sciences 177 (2015) 306 – 311
2.4. Procedure
The extent of technology support offered to students depends on the particular course in the experiment. Courses
originally developed online were prepared and used for the first time in the 2012/2013 academic year. At the end of
the course the opinion and proposals for improvement of students, as well as teachers, was collected. Then courses
and entire learning environment together with organization of education process were adapted accordingly to
students and teachers proposals. Course content, although adapted in 2013/2014, was made available to all students
in much the same way in both academic years. Students received the educational content through learning
management system integrated with wiki system and virtual conferencing system, and the university teachers
provided face-to-face lectures supplementing the material as needed through the teaching period. There were two to
four primary course components, depending on the course, associated with helping students learn the content in the
courses in addition to the lectures: homework, labs, quizzes, and assignments.
3. Results
Collected data was analyzed to determine the reasons and to provide answers to research questions. Results are
organized according to courses.
3.1. Discrete Mathematics (DM)
The average score of the students in the academic year 2012/2013 in the entrance test was 11.52%. Using the
Shapiro Wilk normality test we reject the hypothesis that the score in the entrance test is normally distributed
(W=0.9207, critical value is 0.9419).
The average score of the students in the academic year 2012/2013 in the final test was 62.71%. Using the Shapiro
Wilk normality test we accept the hypothesis that the score in the final test is normally distributed (W=0.9436,
critical value is 0.9419). It is clear that the score of each student in the final test is much higher than the score in the
entrance test. The correlation coefficient between the results of the entrance and the final test is 0.27 and its 95%
confidence interval is (-0.047;0.537), from which follows that the dependence between the results of the tests is very
small.
Shapiro Wilk normality test we reject the hypothesis that the score in the entrance test is normally distributed
(W=0.9042, critical value is 0.943). The average score of the students in the academic year 2013/2014 in the final
test was 57.91%. Using the Shapiro Wilk normality test we accept the hypothesis that the score in the final test is
normally distributed (W=0.984, critical value is 0.946). It is clear that the score of each student in the final test is
much higher than the score in the entrance test. The correlation coefficient between the results of the entrance and
the final test is 0.557 and its 95% confidence interval is (0.263;0.757), from which follows that there is a small
dependence between the results of the tests.
Using the Mann-Whitney U test we reject the hypothesis that there is no significant difference between the results
of the entrance tests (u=4.088, critical value is 1.96), from which follows that at the beginning of the term the
students in the academic year 2012/2013 were better than the students in the following year. However, using
unpaired two-sample t-test we accept the null hypothesis ‘There is no significant difference between the results of
the final tests in the years 2012/2013 and 2013/2014’ (the probability of Type I error is 9.89%). Moreover, using F-
test we accept hypothesis ‘There is no significant difference between the variances of the final test results in the
years 2012/2013 and 2013/2014’ (F=1.58, critical value is 1.91).
To fully understand the meaning of the gained results, we compared the results of students in other courses, which
were taught by the classical way of teaching. The weighted arithmetic mean of all their courses results was used to
gain a better image of overall students´ study quality. The average of the weighted arithmetic means of students that
studied discrete mathematics in 2012/2013 was 2.62 and the average of the weighted arithmetic means of students
that studied discrete mathematics in 2013/2014 was 2.57. Using unpaired two-sample t-test we accept the null
hypothesis ‘There is no significant difference between the results of students which studied discrete mathematics in
2012/2013 and 2013/2014’ (the probability of Type I error is 66.98%).
3.2. Database systems (DBS)
Shapiro Wilk normality test we accept the hypothesis that the score in the entrance test is normally distributed
(W=0.9378, critical value is 0.889). The average score of the students in the academic year 2012/2013 in the final
test was 71.71%. Using the Shapiro Wilk normality test we accept the hypothesis that the score in the final test is
the final test is 0.165 and its 95% confidence interval is (-0.485;0.698), from which follows that the dependence
between the results of the tests is very small.
(W=0.939, critical value is 0.928). The average score of the students in the academic year 2013/2014 in the final test
was 69.49%. Using the Shapiro Wilk normality test we accept the hypothesis that the score in the final test is
the final test is 0.377 and its 95% confidence interval is (-0.068;0.697), from which follows that the dependence
Using the Mann-Whitney U test we accept the hypothesis that there is no significant difference between the
results of the entrance tests (u=1.592, critical value is 1.96). Using unpaired two-sample t-test we accept the null
hypothesis ‘There is no significant difference between the results of the final tests in the years 2012/2013 and
2013/2014’ (the probability of Type I error is 55.4%). Moreover, using F-test we accept the hypothesis ‘There is no
significant difference between the variances of the final test results in the years 2012/2013 and 2013/2014’ (F=1.06,
critical value is 2.60).
We also compared the results of these students in other subjects, which were taught by the classical way of
teaching. We used weighted arithmetic mean of all their subjects. The average of the weighted arithmetic means of
students that studied database systems in 2012/2013 was 2.37 and the average of the weighted arithmetic means of
students that studied database systems in 2013/2014 was 2.53. Using unpaired two-sample t-test we accept the null
hypothesis ‘There is no significant difference between the results of students which studied database systems in
2012/2013 and 2013/2014’ (the probability of Type I error is 27.99%).
3.3. Operating systems and Computer Architecture (OSaCA)
the final test is 0.335 and its 95% confidence interval is (-0.423;0.817), from which follows that there is a small
dependence between the results of the tests.
310 Martin Misut and Milan Pokorny / Procedia - Social and Behavioral Sciences 177 (2015) 306 – 311
the final test is -0.091 and its 95% confidence interval is (-0.551;0.410), from which follows that the dependence
Using unpaired two-sample t-test we accept the hypothesis that there is no significant difference between the
results of the entrance tests (the probability of Type I error is 16.31%). Using the same test we reject the null
hypothesis ‘There is no significant difference between the results of the final tests in the years 2012/2013 and
2013/2014’ (the probability of Type I error is 3.56%)), from which follows that at the end of the term the students in
the academic year 2012/2013 were better than the students in the following year.
We also compared the results of these students in other subjects, which were taught by the classical way of
teaching. We used weighted arithmetic mean of all their subjects. The average of the weighted arithmetic means of
students that studied operating systems and computer architecture in 2012/2013 was 2.34 and the average of the
weighted arithmetic means of students that studied operating systems and computer architecture in 2013/2014 was
2.55. Using unpaired two-sample t-test we accept the null hypothesis ‘There is no significant difference between the
results of students which studied operating systems and computer architecture in 2012/2013 and 2013/2014’ (the
probability of Type I error is 19.64%).
4. Conclusions and areas for further investigation
Comparing students’ results of subsequent years in the discrete mathematics course, we observe that in general
there were no statistically significant differences with one exception – entrance test. The fact, that students’ results in
entrance test were significantly worse and performance in final tests was similar with previous academic year
students helps us conclude that the improvement of the learning environment could have a positive influence on
students’ performance.
The same situation is in DBS and OSaCA courses. There is no statistically significant difference in performance
between 12/13 and 13/14 academic year students in final tests, although 12/13 students slightly outperformed 13/14
academic year students. These findings were unexpected because of the adaptation of learning environments
according to the students’ proposals. To clear the situation, additional data was taken into account. Weighted
arithmetic mean of study results were calculated for each student and then the mean of the whole course group was
computed. Results are summarized in Table 1, where F represents the final test group score and M denote the
weighted arithmetic mean of students results for the given academic year in percentage.
Table 1. Comparison of students´ final tests and weighted arithmetic mean of study results
DM DBS OSaCA
year F M F M F M
12/13 62.71 57.51 71.71 62.49 72.22 63.20
13/14 57.91 58.37 69.49 59.39 64.74 58.94
From Table 1 it is obvious that the 12/13 academic year students had better results in general. The only one
exception is course DM where the 13/14 academic year students had slightly better overall results but they were
outperformed with statistically significant difference in entrance tests. These results could indicate that the 13/14
academic year students had slightly worse assumptions to reach better score in final tests.
When final test scores are compared to the weighted arithmetic mean of study results we observe better results in
experimental blended courses than in traditional courses. There is only one exception – DM course in the 13/14 year.
Since all exceptions coming from this course we can assume that the reason lie in the change of teacher. This course
was taught with different teachers in subsequent years 12/13 and 13/14 so we did not analyze this course’s 13/14
year results further. Data was tested using the Shapiro Wilk normality test and we accept in all cases the hypothesis
that the difference in scores is normally distributed (results are in columns W and crit in Table 2). Then we followed
with paired samples t-test and we reject in all cases the hypothesis that there is no significant difference between the
results of the final test and the weighted arithmetic mean (the probability of Type I error is in columns E in Table 2).
Table 2. Shapiro Wilk normality test and paired samples t-test results
DM DBS OSaCA
year W crit E W crit E W crit E
12/13 0.960 0.950 0.9% 0.952 0.889 0.7% 0.935 0.883 0.6%
13/14 N/A N/A N/A 0.965 0.928 0.1% 0.970 0.917 10.1%
As it is obvious from gained results of additional analysis, there is strong evidence that students’ performance in
blended courses was higher than in traditional ones. Moreover, the probability of the significant difference was more
than 99% (only in OSaCA 13/14 it was 89.9%). Those results proved the usefulness of the realized project, but need
confirmation by further research along with the search for causes of drop in scores in subsequent years’ tests.
Acknowledgements
This research was financially supported by the EU OPV-2010/1.2/02-SORO, ITMS26110230039 and partially by
KEGA (Grant No. 010TTU-4/2012).
References
Bargagliotti, A., Botelho, F., Gleason, J., Haddock, J., & Windsor, A. (2012). The Effectiveness of Blended Instruction in Core Postsecondary
Mathematics Courses. The International Journal for Technology in Mathematics Education, 19(3), 83-94.
Ehlers, U. D., & Hilera, J. R. (2012). Special Issue on quality in e-learning. Journal of Computer Assisted Learning, 28(1), 1-3. doi:
10.1111/j.1365-2729.2011.00448.x
Kashefi, H., Ismail, Z., Yusof, Y. M., & Rahman, R. A. (2012). Supporting Students Mathematical Thinking in the Learning of Two-Variable
Functions Through Blended Learning. Procedia - Social and Behavioral Sciences, 46, 3689-3695. doi: 10.1016/j.sbspro.2012.06.128
Misut, M., Pribilova, K., Orolinova, M., & Kotulakova, K. (2013). Quality of Education Improvement through Adapted Kirkpatrick Evaluation
Model. Advances in Education Research, 44, 109-114.
Misutova, M. (2009). New Education Technology In Mathematics (Vol. 1 and 2). Olomouc: Palacky University, 510-513
Pokorny, M. (2012). Efficiency of Blended Learning in Teaching Logic, Sets and Binary Relations. New York: IEEE, 301-305.
Pyzdrowski, L. J., Butler, M. B., Walker, V. L., Pyzdrowski, A. S., & Mays, M. E. (2011). Exploring the Feasibility of Dual-Credit Mathematics
Courses in High School via a Web-Enhanced, Blended Model. The Journal of General Education, 60(1), 43-60.
Rajini, R., Tiwary, A., & Ganapathy, R. (2011). A Blended Training Model to Improve Learning Outcome of IT Professionals in Scripting
Languages. Paper presented at the Proceedings of the 2011 IEEE International Conference on Technology for Education.
Ramakrisnan, P., Yahya, Y. B., Hasrol, M. N. H., & Aziz, A. A. (2012). Blended Learning: A Suitable Framework For E-Learning In Higher
Education. Procedia - Social and Behavioral Sciences, 67, 513-526. doi: 10.1016/j.sbspro.2012.11.356

Does Ict Improve The Efficiency of Learning?: Sciencedirect

Uploaded by

Copyright:

Available Formats

Does Ict Improve The Efficiency of Learning?: Sciencedirect

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Does Ict Improve The Efficiency of Learning?: Sciencedirect

Uploaded by

Copyright:

Available Formats

Available online at www.sciencedirect.

Does ICT Improve the Efficiency of Learning?

2. Research context and method

2.3. Data Collection

3.1. Discrete Mathematics (DM)

3.2. Database systems (DBS)

3.3. Operating systems and Computer Architecture (OSaCA)

4. Conclusions and areas for further investigation

You might also like