The Unofficial 2014 NFL Player Census Data
Scatter Plot
The two quantitative variables I used were height and age. The scatter plot shows no correlation. The scatter plot does not show an apparent linear relationship between the two variables.
There are no outliers in this data so nothing will be removed.
An appropriate significance level for my data is .05 because it between .1 and .01. That allows for some margin of error without having to be precise such as .01 and having too much error like in .1
My line of best fit will be 0.0627x + 21.561645. The m or slope in the equation is 0.0627 and the b or y intercept will be 21.561645
Simple Linear Regression
Simple linear regression results:
Dependent Variable: AGE Independent Variable: HT AGE = 21.561645 + 0.062728519 HT Sample size: 1689 R (correlation coefficient) = 0.051425411 Rsq = 0.0026445729 Estimate of error standard deviation: 3.2166245 Parameter estimates:
Analysis of variance table for regression model:

My correlation coefficient is .05 indicating a very weak correlation between the age of players and their heights.
My terms are significant because the p value is lower than my significance level which is at .05
The y intercept is about 21.56 years and the slope is about .06 years. The line of best fit shows that for every one inch increase the age of the player will increase by about .06 years.
R squared is about .003 which means that about .3% of the variation in the in the ages of NFL players is accounted for by the bestfit line relating to age and height.
Line of Best Fit
The line of best fit is not a good representation because it hits some points but it is not representative of the other values. There are values not being represented even though they are not outliers.
The data has no correlation.
Extra Credit #1
I do expect my values to follow a normal distribution.
Extra Credit #2
My residual plot implies that the linear model is a good fit because the residual plots are evenly dispersed around the horizontal axis.
Extra Credit #5
Multiple Linear Regression
Multiple linear regression results:
Dependent Variable: AGE Independent Variable(s): HT, WT AGE = 21.319806 + 0.067231098 HT + 0.00036939137 WT Parameter estimates:
Analysis of variance table for multiple regression model:
Summary of fit: Root MSE: 3.2175554 Rsquared: 0.0027 Rsquared (adjusted): 0.0015 
The dependent variable is age and the independent variable is height and weight. The y intercept is 21.319806 and the slope for height is .067 and the slope for weight is 0.00036. R squared is .0027 which is about .003. That means about 3% of variation in the ages of NFL players is accounted for by the line of best fit for ages, height and weight.
Already a member? Sign in.
Nov 13, 2017
Nice report.