Chemistry Reference
In-Depth Information
making attribute for a portion of qualitative data (Miles & Huberman,
1994
).
Patterns in the codes were reported after the three coders reached consensus via
discussion.
Finally, once validity and reliability evidence supported score interpretation,
students
attitude scores from KU were compared with scores from WU and SE
students, using a standard effect size method (Cohen,
1988
). Cohen
'
s
d
effect size
can be used to quantify the difference in the attitude score. Values for Cohen
'
s
d
of
0.2-0.3 are generally considered a “small” effect, of around 0.5 a “medium” effect,
and of 0.8 or above a “large” effect. A medium effect size reflects a difference that
would be noticeable to a careful observer.
'
4 Results
4.1 Descriptive Statistics for Item Scores for KU
Descriptive statistics are shown in Table
1
for each item (four reverse ordered items
were reverse coded for interpretation). High scores mean that students feel chem-
istry is intellectually accessible and emotionally satisfying. The average scores
range from 3.37 to 4.54, and standard deviations range from 1.39 to 1.92. No item
was found to have skewness or kurtosis greater than 1.15, which suggests good
normality of the item scores. From the items (6 and 8) with extreme scores in
Table
1
, students feel chemistry is organized and is challenging. This pattern is very
similar with that observed for data from WU and SE (Brandriet et al.,
2012
;Xu
et al.,
2012
; Xu & Lewis,
2011
).
The box-and-whisker plot for each survey item for KU students is presented in
Fig.
1
. The star near the middle of box represents the mean score for each item,
ranging from 3.37 for item 6 to 4.54 for item 8. The line near the middle of the box
represents the median score for each item. Five items have a median at the middle
point of 4, while items 5 and 8 have the median toward the positive side, with item
6 toward the negative side. The left and right of the box represent the lower and
upper quartiles of the item score. For most items, students tend to pick between
3 and 5. The ends of the whiskers represent the minimum and maximum of all
responses for each item. Note that the students used the full range; the whiskers
extend from 1 to 7. From Fig.
1
, most students tend to pick up the neutral attitude
around 4, and there is a lot of overlap for the item scores.
Because the instrument has the internal structure of two subscales, we proceeded
to examine the evidence based on the intended test design and interpret on the
subscale level rather than on the item level.
Search WWH ::
Custom Search