 |
Asia-Pacific Forum
on Science Learning and Teaching, Volume 10, Issue 1, Article 2
(June, 2009) |
Under this title, the results of descriptive and inferential statistics analyses will be presented. For the analysis, total scores on the subtests of both tests were used for comparisons.
Table 3 The Result of Independent t-test for pre-test scores of the participants at each group on the logical thinking and the creative thinking test
Pre-tests
Groups
n
Mean
SD
t
p
Creative Thinking Test
Experimental
44
49.14
14.75
.66
.51
Control
46
47.24
12.56
Logical Thinking Test
Experimental
44
10.82
3.39
.56
.58
Control
46
11.24
3.75
As seen in Table 3, there is no statistically significant difference between pre-test scores of the sophomores on creative thinking and logical thinking tests (t=.66, p<0.05, t=.56, p<0.05). This result shows that the groups of the study are equivalent in terms of creative and logical thinking parameters before the implementation.
Table 4 Experimental Design (MANOVA)
Independent Variable
Dependent Variables
Groups
Creative and Critical Thinking Based Laboratory Applications and Traditional Lab Applications (Method)
Logical Thinking Skill
Experimental
Control
Creative Thinking Skill
Experimental
Control
Table 4 shows that the design of the study includes one independent variable and two dependent variables within the two groups. The normality assumption was investigated by looking at skewness and kurtosis values for each cell. The values for skewness and kurtosis range from -1 to +1. Therefore, the assumption was accepted to be provided. The assumption of equality of variances was tested by considering Levene’s Test results. The test showed that the assumption was provided with the statistically non-significant results for the equality of error variances.
Table 5 The Results for the Test of Equality of Covariance Matrices
Variable
F
df1
df2
p
Logical Thinking Skill
.02
1
88
.90
Creative Thinking Skill
1.21
1
88
.27
Table 6 The Results for the Test of Equality of Covariance Matrices
Box’s M
5.36
F
1.74
df1
3
df2
1490495
Sig.
0.16
Table 6, shows Box’s value and the significance shows it is one of the most important assumptions of the MANOVA analysis. Equality of covariance is not violated in the study (Box’s M= 5.36, p<0.05).
Table 7 The mean, standard deviation and frequency values of the participants of each group
Dependent Variables
Groups
n
Mean
SD
Creative Thinking Skill
Experimental
44
57.61
9.64
Control
46
51.52
10.36
Logical Thinking Skill
Experimental
44
13.75
2.93
Control
46
10.43
4.12
Table 7 presents means and standard deviations of each group on the post-test scores. The experimental group developed higher scores on the creative and logical thinking abilities tests than the control group.
Table 8 The result for one-way MANOVA
Source of Variance
Multivariate Test
Value
df1
df2
F
Multivariate η2
p
Group
Wilks’ Lambda
.75
2
87
14.64
.25
.00
A one-way multivariate analysis of variance (MANOVA) was conducted to determine the effects of creative and critical thinking based laboratory applications and traditional laboratory applications on two dependent variables, the scores of the students on creative and logical thinking tests. Significant differences were found between the creative and critical thinking based laboratory applications and traditional laboratory applications on the dependent measures, Wilks’s Λ= .75, F (2, 87) =14.64, p<0.05. The multivariate η2 based on Wilks’s Λ was moderately strong at .25. Table 7 contains the mean, standard deviations and frequencies on the dependent variables for each group. The mean of scores on the creative and logical thinking tests demonstrate that the creative and critical thinking based laboratory applications given to the experimental group were found to be more effective than the traditional laboratory activities in improving both the logical and creative thinking abilities.
