129
range of 9 min 40 s, compared to the middle-half of the subjects’ reading times in the
CD
condition, which were rather scattered across a time range of 16 min 43 s
.
In addition, the condensed size of the third quartile section in the PD box plot
indicates that the subjects’ in this group (25 % of the sample) had extremely similar
reading times, all falling in a time interval of 2 min 28 s
.
On the other hand, the third
quartile section of the CD box plot indicates that the subjects in that group were more
variable in their reading times which were spread across a range of 10 min 09 s
.
Information from the box plots also suggests that the subjects generally read
the text faster in the CD than when they did in the PD, with a median of 41 min 40 s
and 55 min 23 s respectively
.
Interestingly, there is an observation of a subject who
spent far longer reading than all the subjects did in PD or
CD conditions, spending 1 h
14 min 48 s (small circle in boxplot for PD reading = subject 30)
.
Considering the reading times for each subject in Table 2, we notice that the
vast majority of subjects took longer to read the text with access to the PD than with
the CD- 37 subjects in total
.
Moreover, and as shown in Table 3 on next page, the
subjects spent 54 min 16 s on average reading the text in the PD condition, with a
minimum time of 40 min 08 s and a maximum time of 74 min 48 s
.
However, in the
CD condition, reading time mean was reduced to 43 min 29 s, with shortest and
longest time values of 24 min 23 s and 64
.
5
minutes respectively
.
130
Table 3 Descriptive Statistics for Reading Time in PD
and CD Conditions
Paired samples statistics
Variable
Total
observations
Mean
Std
.
deviation
Minimum
Median
Maximum
PD
CD
44
44
0
:
54
:
16
0
:
43
:
29
0
:
06
:
55
0
:
10
:
12
0
:
40
:
08
0
:
24
:
23
0
:
55
:
23
0
:
41
:
40
1
:
14
:
48
1
:
04
:
30
At first glance, the aforementioned results might suggest that the PD affected
the subjects’ reading rate negatively compared to the CD
.
Yet, we cannot come to a
valid conclusion about this assumption looking at the subjects’ reading times and time
means only
.
Thus, a paired-samples t-test was conducted to find out whether the
reading time difference between the two conditions is statistically significant
.
The
results for reading time analysis are presented in Table 4
.
Table 4 T-test Results for Reading Time in
PD and CD Conditions
Paired samples test
Variable
Paired differences
t
df
Sig
.
(2-tailed)
Mean
Std
.
deviation
Std
.
error
mean
95% CI
Lower
Upper
PD-CD
0
:
10
:
46
0
:
09
:
27
0
:
01
:
26 0
:
07
:
54 0
:
13
:
39 7
.
56 43
.
000
Note
.
PD= Printed Dictionary, CD= CD-ROM Dictionary, CI= Confidence Interval,
df = Degree of
Freedom
131
The results indicate that there was a highly significant difference in the
reading time for PD (M = 0
:
54
:
16, SD = 0
:
06
:
55) and CD (M = 0
:
43
:
29, SD =
0
:
10
:
12) conditions; t (43) = 7
.
56, p =
.
000, which is less than
0
.
05
.
These results
suggest that using the CD did have an effect on reducing the subjects’ reading time
.
Specifically, the results suggest that when the subjects used the CD, they managed to
finish the reading task in significantly less time than when they used the PD
.
These
results reflect the longer reading and dictionary consultation time needed by the
subjects to read the passage using the PD, compared to the CD which seemed to have
eased and accelerated the reading tasks
considerably
.
These results were expected and are consistent with the results of previous
studies which investigated the effect of using electronic dictionaries on reading time
.
Consulting definitions in a conventional dictionary is a time-consuming task which
increases
reading time significantly
.
The hypothesis which stated that students would spend less time reading with
access to a CD than with a PD is safely validated
.
But, does the higher reading time in
the PD condition reflect greater number of lookups? In other words, did the subjects
look up far more words in the PD than in the CD? Findings related to these questions
will be discussed
later in this chapter
.