Homework Chapter 10.1
Dr. Paul Pearson
2024-11-01
Define helper functions
make_graphs <- function(df, explanatory, response, title_string, xlab_string, ylab_string, filename) {
formula_string <- paste(substitute(response), "~", substitute(explanatory))
# winner: https://win-vector.com/2019/07/06/programming-over-lm-in-r/
# https://stackoverflow.com/questions/27261232/passing-argument-to-lm-in-r-within-function
model <- eval(bquote( lm( as.formula(formula_string), data = df) ))
print(model)
# ggplot scatterplot without line
sp <- ggplot(df, aes( x = {{ explanatory }}, y = {{ response }} )) +
geom_point(size = 4) +
ggtitle( title_string ) +
xlab( xlab_string ) +
ylab( ylab_string ) +
theme_bw(base_size = 18)
print(sp) # generates both graph and VI(ts) automatically
# ggplot scatterplot with line
sp <- sp + geom_smooth(method='lm', se=FALSE, size=2, color = "black")
print(sp) # generates both graph and VI(ts) automatically
MakeAccessibleSVG.ggplot(sp, file = paste0("reader-", filename))
# tactile scatterplot with no line
brl_begin(paste0( 'tactile-', filename, '-scatterplot.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
# to get the variable values, use data[[substitute(varname)]] instead of data$varname
plot(df[[substitute(explanatory)]],
df[[substitute(response)]],
main = title_string,
xlab = xlab_string,
ylab = ylab_string
)
brl_end()
# tactile scatterplot with line
brl_begin(paste0( 'tactile-', filename, '-scatterplot-and-line.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
# to get the variable values, use data[[substitute(varname)]] instead of data$varname
plot(df[[substitute(explanatory)]],
df[[substitute(response)]],
main = title_string,
xlab = xlab_string,
ylab = ylab_string
)
abline( model, col = "black", lwd = 3 )
brl_end()
# tactile scatterplot with line and ellipses and boxplots
brl_begin(paste0( 'tactile-', filename, '-scatterplot-and-line-and-ellipses-and-boxplots.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
scatterplot( as.formula(formula_string), data = df,
main = title_string,
xlab = xlab_string,
ylab = ylab_string,
pch = 16, cex = 2, col = "black",
smooth = FALSE, grid = FALSE, frame = TRUE,
ellipse = list(levels=c(.25, .75), robust=TRUE, fill=FALSE, fill.alpha=0, lwd = 3) )
brl_end()
}HW 10.1.1 hiking distance vs time
distance_mi time_min
0.61 20.32
0.91 30.36
1.01 40.99
1.51 60.30
1.51 45.41
1.80 60.29
2.11 60.43
2.21 90.19
2.56 89.99
2.51 60.30
2.81 89.99
3.01 89.99
3.21 90.12
3.45 90.16
3.80 119.81
4.01 149.91
4.01 120.15
4.56 120.11
4.61 149.91
5.01 149.91
5.51 180.01
5.57 149.91
5.80 180.01
6.01 180.01
7.01 179.88
7.42 419.12
9.52 299.56make_graphs( df = hike,
explanatory = distance_mi,
response = time_min,
title = "HW 10.1 #1 hiking distance vs time",
xlab_string = "distance (miles)",
ylab_string = "time (minutes)",
filename = "HW_10.1.1"
)
Call:
lm(formula = as.formula(formula_string), data = df)
Coefficients:
(Intercept) distance_mi
-8.32 34.70 
This scatter plot has title ‘Hiking distance versus time’ with x-axis ‘distance (miles)’ that has values from 0 to 10, and y-axis ‘time (minutes)’ with values from 0 to 400. There are 27 big solid circle points. The points are concentrated tightly around a straight path (except for one outlier) that is increasing at a constant rate of about 30 minutes for every mile.
This chart has title 'HW 10.1 #1 hiking distance vs time'.
It has x-axis 'distance (miles)' with labels 2.5, 5.0 and 7.5.
It has y-axis 'time (minutes)' with labels 100, 200, 300 and 400.
The chart is a set of 27 big solid circle points of which about 100% can be seen.
It has size set to 4.This chart has title 'HW 10.1 #1 hiking distance vs time'.
It has x-axis 'distance (miles)' with labels 2.5, 5.0 and 7.5.
It has y-axis 'time (minutes)' with labels 0, 100, 200, 300 and 400.
It has 2 layers.
Layer 1 is a set of 27 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.This chart has title 'HW 10.1 #1 hiking distance vs time'.
It has x-axis 'distance (miles)' with labels 2.5, 5.0 and 7.5.
It has y-axis 'time (minutes)' with labels 0, 100, 200, 300 and 400.
It has 2 layers.
Layer 1 is a set of 27 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.
This scatter plot has title ‘Hiking distance versus time’ with x-axis ‘distance (miles)’ that has values from 0 to 10, and y-axis ‘time (minutes)’ with values from 0 to 400. There are 27 big solid circle points. The points are concentrated tightly around a straight path (except for one outlier) that is increasing at a constant rate of about 30 minutes for every mile.
png
2 HW 10.1.3 about glycemic loads and mean rating
foods <- read.table(text=
"MeanRating GlycemicLoad
4.01 22
3.73 14
3.73 12
3.71 7
3.68 14
3.6 21
3.51 17
3.29 16
3.26 24
3.22 0
3.03 0
2.97 7
2.73 15
2.64 26
2.56 22
2.57 22
2.54 0
2.5 29
2.47 3
2.18 0
2.16 0
2.13 15
2.07 11
1.94 0
1.93 10
1.88 1
1.87 8
1.84 0
1.77 12
1.74 0
1.74 20
1.66 4
1.63 7
1.6 2
1.53 0", header=TRUE, check.names=FALSE)
Call:
lm(formula = as.formula(formula_string), data = df)
Coefficients:
(Intercept) GlycemicLoad
2.1944 0.0349 
This scatter plot has title ‘Glycemic load versus mean rating’ with x-axis ‘glycemic load’ that has values from 0 to 30, and y-axis ‘mean rating’ with values from 1.5 to 4. There are 35 big solid circle points. The points are quite spread out horizontally and vertically but they do show a general slight increase in mean rating as glycemic load increases.
This chart has title 'HW 10.1 #3 glycemic load vs mean rating'.
It has x-axis 'glycemic load' with labels 0, 10, 20 and 30.
It has y-axis 'mean rating' with labels 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0.
The chart is a set of 35 big solid circle points of which about 97% can be seen.
It has size set to 4.This chart has title 'HW 10.1 #3 glycemic load vs mean rating'.
It has x-axis 'glycemic load' with labels 0, 10, 20 and 30.
It has y-axis 'mean rating' with labels 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0.
It has 2 layers.
Layer 1 is a set of 35 big solid circle points of which about 97% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.This chart has title 'HW 10.1 #3 glycemic load vs mean rating'.
It has x-axis 'glycemic load' with labels 0, 10, 20 and 30.
It has y-axis 'mean rating' with labels 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0.
It has 2 layers.
Layer 1 is a set of 35 big solid circle points of which about 94% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.
This scatter plot has title ‘Glycemic load versus mean rating’ with x-axis ‘glycemic load’ that has values from 0 to 30, and y-axis ‘mean rating’ with values from 1.5 to 4. There are 35 big solid circle points. The points are quite spread out horizontally and vertically but they do show a general slight increase in mean rating as glycemic load increases.
png
2 HW 10.1.4 on housing sq ft vs prices by location
houses <- read.table(text=
"sqft price location
900.4245641 121.4827342 notlakefront
1332.911007 202.2002959 notlakefront
1664.055598 182.3960118 notlakefront
2236.790998 241.6676939 notlakefront
2272.71276 213.7819329 notlakefront
2407.247765 328.0687403 notlakefront
3165.875711 314.1584852 notlakefront
1597.275542 412.0484081 lakefront
1736.981082 532.1023415 lakefront
1904.070236 497.9171293 lakefront
2695.003193 643.0897783 lakefront
3258.721474 754.0413121 lakefront
3351.031211 901.4073519 lakefront", header=TRUE, check.names=FALSE)
notlakefront <- subset(houses, location == "notlakefront")
lakefront <- subset(houses, location == "lakefront")
print(houses, row.names=FALSE) sqft price location
900.4 121.5 notlakefront
1332.9 202.2 notlakefront
1664.1 182.4 notlakefront
2236.8 241.7 notlakefront
2272.7 213.8 notlakefront
2407.2 328.1 notlakefront
3165.9 314.2 notlakefront
1597.3 412.0 lakefront
1737.0 532.1 lakefront
1904.1 497.9 lakefront
2695.0 643.1 lakefront
3258.7 754.0 lakefront
3351.0 901.4 lakefrontHouse sq ft vs price
Call:
lm(formula = as.formula(formula_string), data = df)
Coefficients:
(Intercept) sqft
-57.346 0.214 
This scatter plot has title ‘Housing price versus square footage of home’ with x-axis ‘square footage’ that has values from 500 to 3500, and y-axis ‘price / 1000’ with values from 100 to 900 (in units thousands of dollars). There are 12 big solid circle points. The points are spread out but do show a general increase in price as square footage increases.
This chart has title 'HW 10.1 #4 house sq ft vs price'.
It has x-axis 'square feet' with labels 1000, 1500, 2000, 2500 and 3000.
It has y-axis 'price/1000' with labels 250, 500 and 750.
The chart is a set of 13 big solid circle points of which about 100% can be seen.
It has size set to 4.This chart has title 'HW 10.1 #4 house sq ft vs price'.
It has x-axis 'square feet' with labels 1000, 1500, 2000, 2500 and 3000.
It has y-axis 'price/1000' with labels 250, 500 and 750.
It has 2 layers.
Layer 1 is a set of 13 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.This chart has title 'HW 10.1 #4 house sq ft vs price'.
It has x-axis 'square feet' with labels 1000, 1500, 2000, 2500 and 3000.
It has y-axis 'price/1000' with labels 250, 500 and 750.
It has 2 layers.
Layer 1 is a set of 13 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.
This scatter plot has title ‘Housing price versus square footage of home’ with x-axis ‘square footage’ that has values from 500 to 3500, and y-axis ‘price / 1000’ with values from 100 to 900 (in units thousands of dollars). There are 12 big solid circle points. The points are spread out but do show a general increase in price as square footage increases.
png
2 House sq ft vs price by location
This scatter plot has title ‘Housing price versus square footage of home’ with x-axis ‘square footage’ that has values from 500 to 3500, and y-axis ‘price / 1000’ with values from 100 to 900 (in units thousands of dollars). There are 12 big solid circle points. The points are spread out but do show a general increase in price as square footage increases.
This chart has title 'HW 10.1 #4: House price versus sq ft'.
It has x-axis 'sq ft' with labels 1000, 1500, 2000, 2500 and 3000.
It has y-axis 'Price/1000' with labels 250, 500 and 750.
It has 3 layers.
Layer 1 is a set of 13 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.
Layer 3 is a 'lm' smoothed curve.Layer 3 has colour set to black.
Layer 3 has size set to 2.Plot with two lines (one line per group)
model_lakefront <- lm( price ~ sqft, data = lakefront)
print(model_lakefront)
Call:
lm(formula = price ~ sqft, data = lakefront)
Coefficients:
(Intercept) sqft
88.112 0.221 model_notlakefront <- lm( price ~ sqft, data = notlakefront)
print(model_notlakefront)
Call:
lm(formula = price ~ sqft, data = notlakefront)
Coefficients:
(Intercept) sqft
59.7405 0.0848 # tactile scatterplot with line
brl_begin(paste0( 'tactile-HW_10.1.4-scatterplot-and-two-lines.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
plot(houses$sqft,
houses$price,
main = "HW 10.1 #4: House price versus sq ft by location",
xlab = "sq ft",
ylab = "Price/1000"
)
abline( model_lakefront, col = "black", lwd = 3 )
abline( model_notlakefront, col = "black", lwd = 3 )
brl_end()png
2 HW 10.1.5 cricket chirps per minute
temperature_Celsius chirps_per_second cricket_type
1 12.72 0.9278 Karschi
2 13.01 0.9808 Karschi
3 13.10 1.0724 Karschi
4 14.28 1.0733 Karschi
5 14.43 1.0317 Karschi
6 15.42 1.0322 Karschi
7 18.50 1.2667 Karschi
8 18.80 1.2546 Karschi
9 19.01 1.6871 Karschi
10 18.87 1.2326 Karschi
11 19.10 1.5040 Karschi
12 19.30 1.3845 Karschi
13 19.44 1.8043 Karschi
14 19.72 1.6675 Karschi
15 19.74 1.8048 Karschi
16 19.81 1.8973 Karschi
17 19.87 1.3919 Karschi
18 20.00 1.2751 Karschi
19 20.03 1.4375 Karschi
20 20.20 1.6409 Karschi
21 20.52 1.5753 Karschi
22 20.64 1.5340 Karschi
23 20.81 1.7587 Karschi
24 20.95 1.5773 Karschi
25 21.16 1.7868 Karschi
26 21.42 1.5563 Karschi
27 21.45 1.8113 Karschi
28 21.61 1.6488 Karschi
29 21.81 1.4840 Karschi
30 22.32 1.8118 Karschi
31 22.32 1.7198 Karschi
32 22.41 1.5825 Karschi
33 22.41 1.3741 Karschi
34 22.51 2.0395 Karschi
35 22.93 1.7424 Karschi
36 22.93 1.6292 Karschi
37 23.03 1.6983 Karschi
38 23.12 1.7911 Karschi
39 23.36 1.7002 Karschi
40 23.86 1.7044 Karschi
41 24.09 1.7263 Karschi
42 24.31 1.8179 Karschi
43 24.40 1.9804 Karschi
44 24.92 1.8651 Karschi
45 25.53 2.3069 Karschi
46 25.72 2.4914 Karschi
47 26.01 2.3269 Karschi
48 10.50 1.3241 Fultoni
49 10.50 1.1803 Fultoni
50 11.13 1.3405 Fultoni
51 11.60 1.3438 Fultoni
52 12.04 1.3608 Fultoni
53 13.01 1.4211 Fultoni
54 13.50 1.5067 Fultoni
55 13.49 1.4380 Fultoni
56 14.01 1.6015 Fultoni
57 14.01 1.4801 Fultoni
58 14.53 1.6581 Fultoni
59 14.50 1.5634 Fultoni
60 15.03 1.6689 Fultoni
61 14.99 1.5903 Fultoni
62 15.02 1.5285 Fultoni
63 15.51 1.7220 Fultoni
64 15.51 1.6315 Fultoni
65 15.51 1.5272 Fultoni
66 15.99 1.7917 Fultoni
67 16.52 1.7145 Fultoni
68 16.50 1.5680 Fultoni
69 17.01 1.8540 Fultoni
70 17.04 1.7673 Fultoni
71 17.01 1.6825 Fultoni
72 18.02 1.9160 Fultoni
73 18.02 1.8408 Fultoni
74 18.02 1.7813 Fultoni
75 20.04 2.0893 Fultoni
76 21.02 2.1567 Fultoni
77 23.03 2.3977 Fultoni
78 23.51 2.4162 Fultoni
79 15.51 1.5625 Fultoni
80 16.01 1.7255 Fultoni
81 16.01 1.6848 Fultoni
82 16.01 1.6114 Fultoni
83 16.01 1.5666 Fultoni
84 16.51 1.7745 Fultoni
85 16.51 1.8152 Fultoni
86 16.51 1.5951 Fultoni
87 17.01 1.7296 Fultoni
88 17.01 1.8152 Fultoni
89 14.01 1.5095 FultoniCrickets temperature vs chirps/sec
Call:
lm(formula = as.formula(formula_string), data = df)
Coefficients:
(Intercept) temperature_Celsius
0.7512 0.0482 
This scatter plot has title ‘African crickets: temperature versus chirps per second’ with x-axis ‘temperature (Celsius)’ that has values from 10 to 30, and y-axis ‘chirps per second’ with values from 0.5 to 2.5. There are 89 big solid circle points. The points are somewhat spread out but do show a clear increase in chirps per second as temperature increases.
This chart has title 'HW 10.1 #5 crickets temp vs chirps/second'.
It has x-axis 'temperature (Celsius)' with labels 10, 15, 20 and 25.
It has y-axis 'chirps per second' with labels 1.0, 1.5, 2.0 and 2.5.
The chart is a set of 89 big solid circle points of which about 100% can be seen.
It has size set to 4.This chart has title 'HW 10.1 #5 crickets temp vs chirps/second'.
It has x-axis 'temperature (Celsius)' with labels 10, 15, 20 and 25.
It has y-axis 'chirps per second' with labels 1.0, 1.5, 2.0 and 2.5.
It has 2 layers.
Layer 1 is a set of 89 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.This chart has title 'HW 10.1 #5 crickets temp vs chirps/second'.
It has x-axis 'temperature (Celsius)' with labels 10, 15, 20 and 25.
It has y-axis 'chirps per second' with labels 1.0, 1.5, 2.0 and 2.5.
It has 2 layers.
Layer 1 is a set of 89 big solid circle points of which about 100% can be seen.
Layer 1 has size set to 4.
Layer 2 is a 'lm' smoothed curve.Layer 2 has colour set to black.
Layer 2 has size set to 2.
This scatter plot has title ‘African crickets: temperature versus chirps per second’ with x-axis ‘temperature (Celsius)’ that has values from 10 to 30, and y-axis ‘chirps per second’ with values from 0.5 to 2.5. There are 89 big solid circle points. The points are somewhat spread out but do show a clear increase in chirps per second as temperature increases.
png
2 Plot with two lines (one line per group)
model_karschi <- lm( chirps_per_second ~ temperature_Celsius, data = karschi)
print(model_karschi)
Call:
lm(formula = chirps_per_second ~ temperature_Celsius, data = karschi)
Coefficients:
(Intercept) temperature_Celsius
-0.2192 0.0882 model_fultoni <- lm( chirps_per_second ~ temperature_Celsius, data = fultoni)
print(model_fultoni)
Call:
lm(formula = chirps_per_second ~ temperature_Celsius, data = fultoni)
Coefficients:
(Intercept) temperature_Celsius
0.2904 0.0875 # tactile scatterplot with line
brl_begin(paste0( 'tactile-HW_10.1.5-scatterplot-and-line-karschi-only.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
plot(karschi$temperature_Celsius,
karschi$chirps_per_second,
main = "HW 10.1 #5: Karschi",
xlab = "temperature (Celsius)",
ylab = "chirps per second"
)
abline( model_karschi, col = "black", lwd = 3 )
brl_end()png
2 # tactile scatterplot with line
brl_begin(paste0( 'tactile-HW_10.1.5-scatterplot-and-line-fultoni-only.pdf'),
pt=12, paper='special', font='SWL') #, font='BRL' or 'SWL')
plot(fultoni$temperature_Celsius,
fultoni$chirps_per_second,
main = "HW 10.1 #5: Fultoni",
xlab = "temperature (Celsius)",
ylab = "chirps per second"
)
abline( model_fultoni, col = "black", lwd = 3 )
brl_end()png
2