In: Statistics and Probability
A company that manufactures small lathes is interested in establishing standards for employees. A random sample of 18 employees is selected in order to develop the standards. The data collected is below. A manager at the firm feels that assembly time is related to intelligence. She feels that employees who did well in high school (as measured by high school averages) should be able to do the job faster. Does the data support her hunch?
high school
average
Time to assemble lathe
(minutes)
50 52
62 53
65 62
68 70
71 73
73 78
75 80
79 82
80 85
82 86
83 90
85 94
88 95
90 106
90 111
94 120
94 139
100 145
x | y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
50 | 52 | 863.7068 | 1448.2253 | 1118.4105 |
62 | 53 | 302.3735 | 1373.1142 | 644.3549 |
65 | 62 | 207.0401 | 787.1142 | 403.6883 |
68 | 70 | 129.7068 | 402.2253 | 228.4105 |
71 | 73 | 70.3735 | 290.8920 | 143.0772 |
73 | 78 | 40.8179 | 145.3364 | 77.0216 |
75 | 80 | 19.2623 | 101.1142 | 44.1327 |
79 | 82 | 0.1512 | 64.8920 | 3.1327 |
80 | 85 | 0.3735 | 25.5586 | -3.0895 |
82 | 86 | 6.8179 | 16.4475 | -10.5895 |
83 | 90 | 13.04 | 0.00 | -0.20 |
85 | 94 | 31.48 | 15.56 | 22.13 |
88 | 95 | 74.15 | 24.45 | 42.58 |
90 | 106 | 112.60 | 254.23 | 169.19 |
90 | 111 | 112.60 | 438.67 | 222.24 |
94 | 120 | 213.48 | 896.67 | 437.52 |
94 | 139 | 213.48 | 2395.56 | 715.13 |
100 | 145 | 424.82 | 3018.89 | 1132.47 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 1429.00 | 1621.00 | 2836.28 | 11698.94 | 5389.61 |
mean | 79.39 | 90.06 | SSxx | SSyy | SSxy |
sample size , n = 18
here, x̅ = Σx / n= 79.389 ,
ȳ = Σy/n = 90.056
SSxx = Σ(x-x̅)² = 2836.2778
SSxy= Σ(x-x̅)(y-ȳ) = 5389.6
estimated slope , ß1 = SSxy/SSxx = 5389.6
/ 2836.278 = 1.90024
intercept, ß0 = y̅-ß1* x̄ =
-60.80246
so, regression line is Ŷ =
-60.802 + 1.900
*x
SSE= (SSxx * SSyy - SS²xy)/SSxx =
1457.3848
std error ,Se = √(SSE/(n-2)) =
9.544
correlation coefficient , r = Sxy/√(Sx.Sy)
= 0.9356
===============
Ho: ß1= 0
H1: ß1 > 0
n= 18
alpha = 0.05
estimated std error of slope =Se(ß1) = Se/√Sxx =
9.544 /√ 2836.28 =
0.1792
t stat = estimated slope/std error =ß1 /Se(ß1) =
1.9002 / 0.1792 =
10.6037
t-critical value= 1.7459 [Excel function:
=T.INV(α,df) ]
Degree of freedom ,df = n-2= 16
p-value = 0.00
decison : p-value<α , reject Ho
reject Ho and conclude that linear relations
exists between X and y