112
time each student started and finished the reading tasks
.
The reading tests were
administered in two separate sessions
:
4
.
4
.
3
.
1
.
Reading with Access to a CD-ROM Dictionary
In the first session, and before the students were allowed to enter the test lab, I
activated the MS in all the computers
.
The CD application was also loaded and its
“Close button” was successfully disabled
.
The students read a text and answered
comprehension questions with access to a CD (Figure 16)
.
The MS recorded the
students’ look-ups without them being aware that they were being monitored for their
lookup behavior (See Appendix A for full test)
.
Figure 16
.
Students reading in CD-ROM dictionary condition
113
4
.
4
.
3
.
2
.
Reading with Access to a Printed Dictionary
In the second session (after a week), the students read another text with access
to a PD (Figure 17)
.
However this time, they were requested to insert the words they
had looked up in a box imbedded into the text (See Appendix B for full test)
.
Figure 17
.
Students reading in printed dictionary condition
To account for the problem of test wiseness, the subjects did the reading in the
CD condition first, since it would have been impossible for them to find out that the
tests were also about their lookup behavior and reading time, and not only about their
reading scores
.
114
However, if the subjects did the reading in the PD condition first, they would
have at least realized that the tests were also about their lookups, since I would have
asked them to report on their lookups in the PD
.
This would have eventually made the
subjects feel suspicious and they might have acted differently the next time they did
the reading in the CD condition
.
4
.
4
.
4
.
Procedure to Test Vocabulary Retention
In order to test students’ vocabulary retention, I administered pretests and
posttests to the subjects in both PD and CD lookup conditions
.
The tests were
administered only on paper since the only variable tested was memory for words
.
The
time taken to find the target words in the dictionaries was not considered for it is not a
variable of interest
.
To account for the problem of test wiseness, I assigned half of the sample -22
students- to the PD lookup condition and the other half to the CD lookup condition
.
I
did not assign all the subjects to both conditions because it would have been
impractical if the students did the pretest and posttest in one condition, and then did a
similar pretest and posttest in the other
.
I could speculate that after taking the first
series of tests in either condition, the subjects would have absolutely realized that they
would be tested on their lookups the next time the pretest in the second series of tests
is carried out
.
If this had happened, they might have copied the target words for later
115
review and further lookup and consolidation of meaning at home, which would have
definitely invalidated the tests’ results
.
4
.
4
.
4
.
1
.
Vocabulary Pretest and Posttest
In the pretest, the students were instructed to supply the definitions for the
words in bold in the sentences using either a PD or a CD (See Appendix G)
.
However, they were not informed that a vocabulary retention test would be given
regarding the words looked up in the pretest, and were assured that what really
mattered were their scores
.
I also made sure the students did not take any notes
regarding vocabulary outside the test room in that they were neither allowed to use
mobile phones nor rough papers
.
After a week lapse, the students were given the posttest to measure their
retention of the words they had looked up a week earlier (See Appendix H)
.
Yet, this
time, the students were not allowed to use any dictionary type, but were encouraged to
rely solely on their memory
.
4
.
5
.
Method of Data Analysis
It should be noted that I designed and administered the entire reading and
vocabulary tests in such a way that the students would not be aware of their real
objectives, except in the PD condition reading task, in which I could by no means find
a trick to make the subjects unaware of the ‘lookup frequency’ variable
.
Before I
116
proceed to analyze the results of the study (next chapter), it is worth mentioning how
the results of the tests were calculated in the first place
.
Since the study consisted of one independent variable, i
.
e
.
dictionary type, and
four dependent variables, i
.
e
.
reading time, reading scores, lookup frequency, and
vocabulary retention, comparison between pretests and posttests was made using
paired-samples t-tests, except for the last dependent variable (i
.
e
.
vocabulary
retention) where an independent-samples t-test was used since each half of the
subjects did only one test type (either a CD or a PD)
.
Pearson’s correlation coefficient
was conducted to examine whether there was any kind of correlation between the
subjects’ lookup frequencies and their reading scores in both conditions
.
4.6. Statistical Procedures Used in the Study
T-tests are often used to compare the means from two different groups of data
.
They can help us find out if the means are significantly different from one another or
if they are relatively the same
.
If the means are significantly different, we can say that
the variable being manipulated (independent variable) had an effect on the variable
being measured (dependent variable)
.
Paired Samples T-Tests
These types of tests are used to compare groups that are related in some way
.
There are so many ways that participants in two groups can be related
.
One way is
Dostları ilə paylaş: |