In: Statistics and Probability
The statistics for the following examples can be computed by hand or using a statistical package. Gina and Ariely (2012) found in a series of studies that creativity was positively related to dishonest behavior, both when people are naturally creative and when they are stimulated to be creative. Suppose you want to know whether creativity is negatively related to integrity in an educational setting. You have students complete a creativity measure and report on their academic honesty (no plagiarism, completing work according to guidelines, no cheating on tests, supporting the institution’s honor code, etc.) The following table presents the data for your sample:
Table: Student ID 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
Creativity 22,45,36,40,28,32,39,18,26,22,30,34,28,17,40
Academic Honesty 21.18.15.18.21.17.10.23.19.15.16.9.13.19.7
a)
graph shows negative relationship between two.
..............
b)
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 457 | 241 | 1023.733333 | 302.9 | -297.47 |
mean | 30.47 | 16.07 | SSxx | SSyy | SSxy |
correlation coefficient , r = Sxy/√(Sx.Sy)
= -0.5342
correlation hypothesis test
Ho: ρ = 0
Ha: ρ > 0
n= 15
alpha,α = 0.05
correlation , r= -0.5342
t-test statistic = r*√(n-2)/√(1-r²) =
-2.278
............
DF=n-2 = 13
p-value = 0.9799
Decison: P value > α, So, Do not reject
Ho
so both are negatively related
............
probability of type I error = 0.05
thanks