7.24 Nutrition at Starbucks, Part I. The scatterplot below shows the relationship between the number of calories and amount of carbohydrates (in grams) Starbucks food menu items contain. 21 Since Starbucks only lists the number of calories on the display items, we are interested in predicting the amount of carbs a menu item has based on its calorie content
There are a moderate to weak positive relationship between the number of calories adn the amount of carbohydrates. It appears from the graph, that the fit is better for data point representing lower number of calories.
The explanatory variable is Calories on the x axis, and the response variable is Carbohydrates on the y axis.
The regression line helps to predict the amount of carbohydrate (in grams) a starbucks menu food items contains by looking at the number of calories in the item..
Linearity: The data should show a linear trend. The scatterplot suggests that the data show a moderate to weak linear relationship.
Nearly normal residual: From the histogram for residual, the residuals in this case have a slightly left skewed.
Constant variability: The residual plot suggests that there is no constant variability for residuals. The data fit the linear model much better for lower number of calories than for higher as shown by much larger residuals value.
Independent observation: Its assumed that each menu item is independent of the other, but they are all Starbucks menu items.
7.26 Body measurements, Part III. Exercise 7.15 introduces data on shoulder girth and height of a group of individuals. The mean shoulder girth is 107.20 cm with a standard deviation of 10.37 cm. The mean height is 171.14 cm with a standard deviation of 9.41 cm. The correlation between height and shoulder girth is 0.67.
# The general equation for the regression line: y = B0 + B1*x where B0 and B1 represent two model parameters.
# X is the explanatory or predictor variable and y is the response.
# B1 is the slope, which also equals: (sample y standard deviation) / (sample x standard deviation) * R.
# R is the correlation between the two variables.
# R ranges from -1 to 1, with -1 being completely negative correlation and +1 being a completely positive correlation.
# Shoulder girth is the explanatory variable
shoulder.mean <- 107.20 # in cm
shoulder.SD <- 10.37 # in cm
# Height is the response
height.mean <- 171.14 # in cn
height.SD <- 9.41 # in cm
# R for correlation
R <- 0.67
# Calculate the slope (or otherwise known as B1)
B1 <- R * (height.SD/shoulder.SD)
# Now to calculate for B0, we will use the values (x,y) = (107.20, 171.14). They are also the mean values, and they lie along the regression line.
# Now to rearrange the equation to solve for B0. B0 = y - B1*x
B0 <- 171.14 - B1 * 107.20
B1;B0
## [1] 0.6079749
## [1] 105.9651
So the general equation of the regression line is
\(height\quad =\quad 105.9651\quad +\quad 0.6079749\quad \times \quad shoulder\_ girth\)
The intercept (105.9651) is the height in centimeters at girth of 0 cm. The slope (0.6079749) is the number of centimeters increase in height for each increase in shoulder girth. Not all values make sense when we plug them into this linear regresion equation. For example, a person with a shoulder width of 0 cm (which can’t happen) would indicate a height of 105.965 cm. This does not make sense.
R.squared <- R^2
paste("R squared: ", round(R.squared,3))
## [1] "R squared: 0.449"
This means that 44.9% of the variation found in this data is explained by the linear model i.e. explained by the shoulder girth width.
student.shoulder <- 100 # in cm
student.height <- B0 + B1 * student.shoulder
paste("According to the model, the estimated height of a student with a shoulder girth of 100 cm is: ", round(student.height,3), "cm.")
## [1] "According to the model, the estimated height of a student with a shoulder girth of 100 cm is: 166.763 cm."
# The residual = time(actual) - time(expected)
residual <- 160 - student.height
paste("The residual is: ", round(residual,3))
## [1] "The residual is: -6.763"
As per excercise 7.15, The original data set had a response variable values between ~80 and 140 cm. A measure of 56 is outside the sample and we would require extrapolation and would not be appropriate. There may be many confounding factors such as malnourishment, developmental delays, congenital diseases, etc. that may not be accounted for in this current linear regression model.
7.30 Cats, Part I. The following regression output is for predicting the heart weight (in g) of cats from their body weight (in kg). The coefficients are estimated using a dataset of 144 domestic cats.
(a). Write out the linear model.
As stated in Question 7.26 for the linear model equation y = B0 + B1 * x, where B0 is the height of the intercept and B1 is the slope. Therefore, by looking at the table, the linear regression model here is:
Heart Weight (g) = -0.357 + 4.034 * `Body Weight (kg)
Expected heart weight in cats with 0 kg body weight is -0.357 g. This is not a meaningful value, it just serves to adjust the height of the regression line.
For each additional kg increase in body weight, we expect an additional 4.034 grams in the heart weight.
Body weight (in kg) explains 64.66% of the variability in the heart weight (in g) of the cat.
paste("Correlation coefficient: ", round(sqrt(.6466),3))
## [1] "Correlation coefficient: 0.804"
7.40 Rate my professor. Many college courses conclude by giving students the opportunity to evaluate the course and the instructor anonymously. However, the use of these student evaluations as an indicator of course quality and teaching e???ectiveness is often criticized because these measures may reflect the influence of non-teaching related characteristics, such as the physical appearance of the instructor. Researchers at University of Texas, Austin collected data on teaching evaluation score (higher score means better) and standardized beauty score (a score of 0 means average, negative score means below average, and a positive score means above average) for a sample of 463 professors.24 The scatterplot below shows the relationship between these variables, and also provided is a regression output for predicting teaching evaluation score from beauty score.
**Answer Slope can be calculated from the linear model equation y = B0 + B1 x Here, they provide enough information so that we can solve B1 (the slope) We can substitute x with the average standardized beauty score (explanatory variable) = -0.0883 and we can substitute y with the average teaching evaluation score of 3.9983 (response). We know that the averages lie on the linear regression model.We also know the Y intercept. It is on the table. Y intercept = 4.010 Therefore, the equation is now: 3.9983 = 4.010 + B1 * (-0.0883) Now solve for B1 to find the slope.
prof.B1 <- (3.9983 - 4.010)/(-0.0883)
paste("B1 or the slope: ", round(prof.B1, 3))
## [1] "B1 or the slope: 0.133"
**Answer* While on visual inspection, if anything, it looks like there may be a near zero slope level. However, the data from the summary chart does suggest that there is some correlation between beauty and teaching evaluation. The P values in the summary chart demonstrates p’s that are ~0, which suggests that we should reject the null hypothesis (null hypothesis being that there is NO correlation).
Answer
There are four criteria that needs to be satisfied prior to using the least squares linear regresion method.
Linearity : There does appear to be some linearity if we take a horizontal slope and place a horizontal line roughly right through the data points
Nearly normal residuals : The residuals do appear to be nearly normal, by looking at its distribution on the histogram.
Constant variability : By looking at the residual plots, there appears to be constant variability throughout
Independent observations : Looking at the order of data collection, there doesn’t seem to an obvious pattern that would suggest that these observations were dependent on each other. It is likely that these were all independent observations.
Hence, we can perform the linear regression.