A sample of 95 healthy men and women was collected to examine the relationship b
ID: 3226279 • Letter: A
Question
A sample of 95 healthy men and women was collected to examine the relationship between cholesterol and other factors including age, weight, height, and gender. The code below gives the output for a linear regression predicting cholesterol based on weight. Based on this model:
> summary(lm(Cholesterol~Weight, temp))
Call:
lm(formula = Cholesterol ~ Weight, data= temp)
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
a) Write down the fitted equation from this model
b) What is the predicted difference in cholesterol between two people who are one pound different in weight?
c) What is the predicted difference in cholesterol between two people who are ten pounds different in weight?
d) Is the heavier person predicted to have higher or lower cholesterol compared to the lighter person, based on this model?
e) Does weight significantly (at alpha=.05) predict cholesterol?
Coefficients:
Estimate
Std Error
t-value
Pr(>|t|)
(Intercept)
205.160
16.773
12.23
<2e-16 ***Weight -0.155 0.105 -1.48 0.14
Explanation / Answer
a) Y = 205.16 - 0.155x
b) if x = 1 => y = 205.005
c) if x = 10 => y = 204.05
d) This model does not belong to Heavy weight and light weight. It concentrates mainly there is significant difference between weight or not.
e) Decision rule : if p value < 0.05, Then there is significant difference between the weight
Actaul decision : No significant difference , Since p value = 0.14 > 0.05