In: Statistics and Probability
A study was conducted to determine whether the final grade of a student in an introductory psychology course is linearly related to his or her performance on the verbal ability test administered before college entrance. The verbal scores and final grades for 10 students are shown in the table below.
Student / Verbal Score x / Final Grade y
1 / 42 / 88
2 / 68 / 65
3 / 31 / 83
4 / 68 / 64
5 / 42 / 82
6 / 65 / 65
7 / 72 / 68
8 / 35 / 89
9 / 76 / 76
10 / 42 / 68
Find the following:
(a) The correlation coefficient: r= -0.7231
(b) The least squares line: y^=
(c) Calculate the residual for the ninth student=
Solution:
Correlation coefficient can be calculated as
Correlation Coefficient = (n*Summation(XY) -
Summation(X)*Summation(Y))/sqrt(((n*Summation(X^2)) -
(Summation(X))^2))*((n*Summation(Y^2)) -
(Summation(Y))^2)))
Student |
Verbal Score(X) |
Final Grade(Y) |
X^2 |
Y^2 |
XY |
1 |
42 |
88 |
1764 |
7744 |
3696 |
2 |
68 |
65 |
4624 |
4225 |
4420 |
3 |
31 |
83 |
961 |
6889 |
2573 |
4 |
68 |
64 |
4624 |
4096 |
4352 |
5 |
42 |
82 |
1764 |
6724 |
3444 |
6 |
65 |
65 |
4225 |
4225 |
4225 |
7 |
72 |
68 |
5184 |
4624 |
4896 |
8 |
35 |
89 |
1225 |
7921 |
3115 |
9 |
76 |
76 |
5776 |
5776 |
5776 |
10 |
42 |
68 |
1764 |
4624 |
2856 |
541 |
748 |
31911 |
56848 |
39353 |
Correlation coefficient = ((10*39353) - (541*748))/sqrt(((10*31911
- 541*541)*(10*56848 - 748*748)) = -11138/sqrt(26429*8976) =
-0.7231
Solution(b)
Regression equation can be calculated as
Y = a+bx
Here a is intercept of line and b is slope of line
Slope = (n*Summation(XY) -
Summation(X)*Summation(Y))/((n*Summation(X^2)) - (Summation(X))^2))
= ((10*39353) - (541*748))/(10*31911 - 541*541) = -0.4214
Intercept can be calculated as
Intercept = (Summation(Y) - Slope*Summation(X))/n = (748 +
0.4214*541)/10 = 97.6
So regression equation is
Y = 97.6 - 0.42*X
Solution(c)
At X = 76, Y = 97.6 - 0.42*76 = 65.57
So residual can be calculated as
Residual = Predicted value - Observed value = 65.57 - 76 =
-10.43