a.

A

paired-samples t test was performed to find out whether participation in a

creative writing course results in increased scores of a creativity assessment.

A total of 40 students completed a

pre-test, took the creativity course and then took a post-test. Results

indicated that the mean pret-test score (M = 40.15, SD = 8.30), was lower than

the post-test than score (M = 43.3, SD = 9.60) The mean difference was

statistically significant, t(39) = -2.67, p = .011.

That

is, there is sufficient evidence to conclude that participation in a creative

writing course results in increased scores of a creativity assessment.

An independent sample t test was conducted

to compare pre-test and post-test scores in a creativity assessment. A total of

40 students took a pre-test scores and another sample of 40 students took a

pre-test score. From the Levine’s Test

for equality of variances, significance level is .429. This value is greater than .05 (The

significance level used in the test). This implies that the variances are

equal, we use the Equal variances assumed row in our analysis.The average score for students who took a post-test (M = 43.35, SD = 9.60)

was slightly higher than the average score for students who took a post (M =

40.15, SD = 8.30). However the mean difference was not statistically

significant, t(78) = – 1.595, p = .115. These results suggest that there is not

sufficient evidence to conclude that the average scores were equal between

those students who took a pre-test and the ones that took a post-test.

A

comparison of means test with t test was conducted. The means compared were for the pre-test and

post-test scores in a creativity assessment. The first test assumed paired

samples and results indicated that the mean pret-test score (M = 40.15, SD =

8.30), was lower than the post-test than score (M = 43.3, SD = 9.60) The mean

difference was statistically significant, t(39) =

-2.67, p = .011. The second test assumed independent samples. The

average score for students who took a post-test (M = 43.35, SD = 9.60) was

slightly higher than the average score for students who took a post (M = 40.15,

SD = 8.30). However the mean difference was not statistically significant,

t(78) =

– 1.595, p = .115.

a.

An experiment was

conducted on the comparison of means using independent and dependent tests.

Both tests used the same data set. With the dependent t test the researcher

wanted to find out whether participation in a creative writing leads to

increased scores of a creativity assessment. Participants took a pre-test, took

a creativity course and then too the post-test.

Only 40 participants were involved in the experiment. Independent t

test, using the same results assumed two independent samples. That is, 40

students took the pre-test and another different group of 40 students took the

post test. The aim of the test was to compare the means of the two groups of

students.

The

paired samples t test found statistically significant difference between the

mean pret-test score (M = 40.15, SD = 8.30), and the mean post-test than score

(M = 43.3, SD = 9.60), t(39) = -2.67, p = .011. This showed that

participation in a creative writing course resulted in increased scores of a

creativity assessment. The independent t test results indicated that average

score for students who took a post-test (M = 43.35, SD = 9.60) was not

statistically different from the average score for students who took a

post-test (M = 40.15, SD = 8.30), t(78)

= – 1.595, p = .115. There is no

enough evidence to conclude that the

average scores were equal between those students who took a pre-test and the

ones that took a post-test.

AS expected the findings from the

two tests were not the same. The paired samples t test found a significant

difference between the means, while the independent samples t test results did

not find significance in the mean difference between the two groups that

participated in the creativity assessment. Pairing reduces

variability and hence make more precise comparisons with fewer subjects. When

we use independent samples, the difference between treatment means is compared

to the variability of individual responses within each treatment group. This

might have caused the difference in the results obtained by the two tests. With paired samples, the between-subject

variability will be reduced. Since variability is reduced by using paired

samples, dependent t test may then give more precise results.