پایان نامه رایگان درمورد improvement، score، rate

دانلود پایان نامه

the experiment.
In order for this comparison to be possible, we needed to calculate and ascribe a numerical value to the improvement made by each individual in the experimental group so that this number could be compared with their score in different subsets of the MI test. The improvement rate for each individual was the difference between their pretest score and posttest score. As the scores given to the trainees’ SI performance was on a 0-100 scale, the difference between the pretest score and posttest score (the improvement rate) is also expressed on the same scale, that is to say, on a percentage basis.
Table 4.15 below presents the improvement rate the experimental subjects demonstrated in the experiment.

Table 4.15 Experimental Subjects’ SI Improvement Rate
Score

Subject
Aver.
Pretest
Score
Aver. Posttest Score
SI Improvement Rate
S1
22
60
38
S2
9
42
33
S3
7
42
35
S4
9
35
26
S5
10
49
39
S6
7
42
35
S7
25
70
45
S8
7
30
23
S9
24
44
20
S10
20
64
44
S11
14
62
48
S12
10
54
44
S13
12
40
28
S14
25
69
44
S15
17
37
20
S16
10
40
30
S17
32
70
38
S18
7
27
20
S19
10
52
42
S20
5
27
22
S21
54
64
10
S22
65
70
5
S23
5
25
20
S24
5
25
20
S25
7
25
18
S26
35
70
35
S27
19
42
23
S28
22
62
40
S29
42
64
22
S30
7
25
18
S31
34
70
36
S32
35
62
27
S33
34
64
30
S34
30
60
30
S35
32
64
32

As can be seen in table 4.15, the mean improvement rate for the experimental subjects was 29, that is to say, on average the experimental participant depicted a 29% improvement in the quality of their SI performance during the experiment period.
What we were looking for was to see whether a statistically significant relation could be established between this set of numbers and the trainees’ scores on each of the subsets of the MI test. The first intelligence tested was linguistic intelligence.

4.4.1 Linguistic Intelligence and SI Improvement Rate
In the test of multiple intelligences utilized in this study, the score in each of the subsets is expressed on a percentage basis – i.e. ranging from zero to one hundred. Table 4.16 shows the scores of the trainees in terms of their verbal-linguistic intelligence.

Table 4.16 Linguistic Intelligence Scores for Experimental Subjects
Subject
Linguistic Intelligence

Subject
Linguistic Intelligence
S1
100

S19
64
S2
46

S20
73
S3
86

S21
60
S4
88

S22
30
S5
62

S23
57
S6
10

S24
40
S7
80

S25
73
S8
73

S26
10
S9
97

S27
73
S10
64

S28
28
S11
46

S29
100
S12
90

S30
91
S13
88

S31
86
S14
24

S32
57
S15
57

S33
48
S16
57

S34
90
S17
100

S35
46
S18
40

The formula for Pearson Product-moment correlation (Hatch & Lazaraton, 1991, p. 434) was made use of in order to find out the correlation coefficient between the two sets of scores.

            r_xy=(N(∑▒〖XY) -(∑▒X)(∑▒Y) 〗)/√([N∑▒X^2 -(∑▒X)^2 ][N∑▒Y^2 -(∑▒Y)^2 ] )

In this formula, r is the symbol for the Pearson correlation coefficient. The subscripts x and y stand for the two variables being compared i.e., the score for improvement rate in SI performance and the score for linguistic intelligence. Below are the calculations.

            r_xy=(35×66089-1040×2234)/√((35×34682-1040×1040)(35×164830-2234×2234) )

            r_xy=(-10245)/√(132270×778294)=(-10245)/320850.3=-0.03

As the calculations showed, the correlation coefficient measured -0.03, which indicates a slight negative correlation between the trainees’ rate of improvement regarding SI performance and their linguistic intelligence. Having consulted the table for critical values for Pearson Product-moment correlation (Hatch & Lazaraton, 1991, p. 604), we find out that this negligible correlation coefficient is much smaller than the critical value required to reject the null hypothesis (0.3494, with df being 30 and p being 0.5). This is indicative of the fact that there is no relation between the participants’ level of verbal-linguistic intelligence and the improvement they made in their SI performance over the experiment period.
This can be better seen and understood when we look at the scatterplot in figure 4.5 below. Each point in the graph represents one subject’s SI improvement rate (on the vertical axis) and the extent of their verbal-linguistic intelligence (on the horizontal axis). As figure 4.5 shows, the points are quite drastically scattered, so much so that it is not possible to imagine any regression line on the plot. This shows a correlation coefficient near zero (-0.03 is really very close to zero). “The most sensible way of interpreting a correlation coefficient is to convert it into overlap between the two measures.” (Hatch & Lazaraton, 1991, p. 440) To do so, we have to square the value of r. When r = -0.03, then r2 = 0.0009, which means that only 0.0009 of the variance in SI improvement rate can be accounted for by the variance in verbal-linguistic intelligence and vice versa. Thus the null hypothesis of no relation between SI improvement and linguistic intelligence, implied in H0.2 of this study, cannot be rejected and is proved right.

Figure 4.5 Scatterplot Diagram for SI Improvement and Linguistic Intelligence

Leave a comment