Вы находитесь на странице: 1из 5

SKITTLES PROJECT-PART 3

CORRELATION AND REGRESSION

-I do not think height has anything to do with predicting the number of candies in a bag. I am
5’6”, my friend could be 5’4” and another person could be 5’8”and this numbers will not
change anything. I think it is just a chance of how many candies you can get in a bag regardless
of height.

- The explanatory variable would be the height of a person –X

-The response variable would be the number of candies-Y


Row red orange yellow green purple total height
1 8 10 11 12 15 56 72
2 8 8 10 14 17 57 67
3 11 9 16 9 12 57 60
4 10 22 13 8 5 58 66
5 9 12 13 11 13 58 66
6 12 6 15 12 13 58 64
7 14 10 12 13 9 58 62
8 11 13 10 14 10 58 62
9 10 10 14 9 15 58 67
10 6 10 11 16 16 59 65
11 11 15 11 10 12 59 68
12 9 7 12 16 15 59 68
13 16 11 12 7 13 59 63
14 12 9 18 10 11 60 60
15 14 12 10 13 11 60 58
16 11 13 8 17 11 60 64
17 15 7 13 14 12 61 64
18 7 15 16 13 10 61 67
19 10 17 9 13 12 61 66
20 9 15 17 11 9 61 66
21 14 13 11 13 11 62 67
22 9 20 9 11 13 62 66
23 13 10 11 13 15 62 64
24 13 15 9 9 16 62 66
25 11 10 11 17 13 62 62
26 9 10 15 15 14 63 75
27
28
Simple linear regression results:
Dependent Variable: total
Independent Variable: height
total = 56.064493 + 0.055057924 height
Sample size: 26
R (correlation coefficient) = 0.10307475
R-sq = 0.010624405
Estimate of error standard deviation: 1.9462399

-The value of the correlation coefficient is .103, which is below the critical value. There is not
a significant relationship between the two variables.

Parameter estimates:
Parameter Estimate Std. Err. Alternative DF T-Stat P-value
Intercept 56.064493 7.0806118 ≠ 0 24 7.9180295 <0.0001
Slope 0.0550579240.10845323 ≠ 0 240.50766512 0.6163

Analysis of variance table for regression model:


SourceDF SS MS F-stat P-value
Model 10.976219350.976219350.25772388 0.6163
Error 24 90.908396 3.7878498
Total 25 91.884615

-Regression equation would be:

Y=.0550x+56.065

-How many candies wuld be expected to be in a bag purchased by someone who is 63.5 inches tall?

Y=.0550(63.5)+56.065

Y=59.55

Y=around 60 candies would be expected in the bag purchased.

-It was not appropiate to use this equation because the is not linear relation between the explanatory
variable and the response variable.

-R2=.01062
Meaning there is a 0.1 %variation in candies explained by the height.

-It still would not be appropiate to predict the number of candies per bag purchased by Yao Ming who is
90 inches tall, because it is outside the scope of the model.

-Smaller Data Set:

Regression Equation

Y=.973x+1.361

Correlation Coefficient

.9972

-The critical value for correlation coefficient shows that there is not a significant linear relationship
between X and Y for this smaller data set.