Information Technology Reference
In-Depth Information
points. Completely correct answers were given 3
points and completely incorrect answers received
0 points. Answers that were neither entirely cor-
rect nor completely incorrect received 2 points.
As such, possible scores on near-transfer learning
test have a possible range of 0 to 15.
Complex or far-transfer tasks are often used
to measure meaningful and far-transfer learning
(Davis & Bostrom, 1993). Simon et al. (1996)
suggest that far-transfer tasks comprise two or
more near-transfer tasks, and it is left to the
trainee to determine which and how the near-
transfer tasks should be combined to perform the
given far-transfer task. Thus, far-transfer learn-
ing was measured by a four-item learning test.
Similar to the tasks used by Davis and Bostrom
(1993), far-transfer tasks included on the learning
performance test were: (1) creating a multilevel
directory structure, (2) deleting a non-empty
subdirectory, (3) copying a file from one direc-
tory to another directory in a different level in
the directory structure, and (4) renaming files in
another directory. Since performing a far-transfer
task requires more mental and physical effort than
performing a near-transfer task, each question
was worth 6 points. Scoring of the far-transfer
learning test followed the same procedure used in
the near-transfer learning test with proportional
points for partially correct answers.
The two learning tests were graded by two
independent graders. Each grader was given
written instructions to follow in grading the tests
and assigning grades. The correlation between
the grades obtained from the two graders was
0.94 (p < 0.001) for the near-transfer learning
test and 0.91 (p <.001) for the far-transfer learn-
ing test. The average of the two scores awarded
by the two graders for each test was used in the
data analysis.
results
Although this study used previously validated in-
struments, a confirmatory analysis was performed
on all multi-item measures. All items exhibited
high loading on their intended constructs and all
measures demonstrated high internal consistency
reliability (coefficient alpha) as can be seen in
Appendix 1.
The means, standard deviations, and correla-
tions among the study variables are presented
in Table 1. As can be seen in Table 1, although
measured on different scales, general CSE beliefs
are noticeably higher than application (Unix) CSE
beliefs. This was expected because Unix is not a
common technology that students interact with
very often or use on a regular basis. The correla-
Table 1. Means, standard deviations, and correlations
Variable
Mean
SD
1
2
3
4
5
1. General CSE
40.63
11.62
--
2. Software-specific CSE
28.43
9.32
0.62**
--
3. Perceived ease of use
15.26
4.64
0.49**
0.34**
--
4. Computer anxiety
13.68
6.44
-0.56**
-0.55**
-0.60**
--
5. Near-transfer learning
10.82
2.74
0.20
0.56**
-0.02
-0.19
--
6. Far-transfer learning
15.79
5.45
0.53**
0.58**
0.30**
-0.34**
0.36**
** p < 0.01
Search WWH ::
Custom Search