HW 6
Lisa Coletti
5/14/2022
1) Compare Two Means
1.1) Create two sample sets
Create two samples from the gcookbook::heightweight data set , each of size 15, but one Female only, and the other Male only.
# Create initial data set from gcookbook
pop <-gcookbook::heightweight
# Set seed and sample number 'N'
set.seed(1515)
N <- 15
# Female sample of 15 students
samp_fem <- pop %>%
filter(sex == "f")
samp_fem <- (samp_fem) %>%
sample_n(N, replace=FALSE)
# Male sample of 15 students
samp_mal <- pop %>%
filter(sex == "m")
samp_mal <- samp_mal %>%
sample_n(N, replace=FALSE)1.2) Run a two sample t-test
For the t-test parameters, I left the default arguments:
alternative = two.sided because we are performing a two sided t-test
mu = 0 because we are assuming the mean of the two samples are the same
var.equal = FALSE to approximate the degrees of freedom
conf.level = 0.95 because the sample size of each data frame is less than 30
# Two sample t-test for weight of both male and female samples
#The null hypothesis is that there is no statistically significant difference in weight between males and females. If p-value is greater than confidence interval, then we accept the null.
#The alternative hypothesis is that on average, men's weight and females weights are different
ttest <- t.test(samp_fem$weightLb, samp_mal$weightLb, mu = 0, var.equal = FALSE)
ttest##
## Welch Two Sample t-test
##
## data: samp_fem$weightLb and samp_mal$weightLb
## t = 0.65267, df = 27.755, p-value = 0.5193
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -8.131051 15.731051
## sample estimates:
## mean of x mean of y
## 97.43333 93.63333#t should be greater than x to be able to reject the null hypothesis
#t = .65267 >
#p value should be less than the confidence interval
#p value = .5193 > than .05 based on a 95% confidence interval
#we cannot reject the null hypothesis, which means there is not evidence of a statistically significant different between male and female weight
# Assign variable for error bars in plot below
se <- ttest$stderrThe sample sets and plot are good examples of how random samples might be contrary to our general assumption of the data. I assumed that the sample sets would show that males have a higher mean weight than females, but this isn’t the case. The mean weight of females in this sample set is 97.43 and is 93.63 for males. I also thought that the error bars looked too big, but it makes sense with such a small sample size resulting in a standard error of 5.82.
1.3) Combine samples and plot
# Combine sample data frames
samp_combined <- bind_rows(samp_fem, samp_mal)
# Find the mean of each sex and add to new data frame
samp_means <- samp_combined %>%
group_by(sex) %>%
mutate(mean = mean(weightLb))
# Plot data
jitter <- ggplot(samp_means, aes(x=sex, y=weightLb, color=factor(sex))) +
geom_jitter(
width = 0.1,
size = 2,
alpha = .8) +
geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=0.5) +
theme_bw() +
theme(
legend.position = "none") +
scale_x_discrete(labels = c("f" = "Female", "m" = "Male")) +
labs(
x = "Gender",
y = "Age (years)",
title = "Comparing female and male sample sets",
subtitle = paste0("Sample size of ", N, " female students and ", N, " male students")
)
jitter
# dont forget labels
#jitterMy sample sets and plot are good examples of how random samples might be contrary to your general assumption of the data. I assumed that the sample sets would show that males have a higher mean weight than females, but this isn’t the case. The mean weight of females in this sample set is 97.43 and is 93.63 for males. I also thought that the error bars looked too big, but it makes sense with such a small sample size resulting in a standard error of 5.82.
2) Plot by Category
2.1) Perform a 2-means (M vs. F Height) t-test for grades 5-10
As suggested, I created a grade column in my original pop data frame. I then created 10 new data frames, filtered by sex and grade, and ran a t-test for all grades. I used the default arguments for all t-tests, meaning the confidence level is 95% and the alpha is 0.05.
# Add grade variable to pop data frame
ttest$estimate## mean of x mean of y
## 97.43333 93.63333ttest$stderr## [1] 5.822234# pop$grade <- code here
pop$grade <- as.numeric(round(pop$ageYear)-6)
pop$grade## [1] 6 7 7 7 10 8 9 6 7 6 6 9 7 6 6 6 10 6 6 9 6 6 8 9 9
## [26] 8 9 9 12 6 8 8 6 6 10 10 8 9 9 9 8 8 8 8 6 9 6 6 6 9
## [51] 9 10 6 6 6 6 7 8 9 6 9 9 6 6 7 7 6 7 7 10 10 6 6 10 6
## [76] 9 8 7 6 6 9 8 8 10 9 6 9 10 9 9 6 8 8 7 6 7 8 6 6 12
## [101] 8 8 9 6 10 6 8 10 8 9 9 8 7 6 6 6 6 10 9 6 6 7 7 8 7
## [126] 6 6 8 7 11 9 9 6 9 9 6 9 6 7 7 6 7 9 10 8 8 10 6 9 9
## [151] 9 8 8 8 8 6 7 6 6 6 10 8 8 6 10 7 8 6 7 7 7 9 7 6 9
## [176] 8 6 6 10 11 6 10 10 11 7 8 8 6 6 8 8 9 8 7 8 9 9 9 8 6
## [201] 6 10 8 8 8 6 8 6 6 6 7 6 11 7 6 10 10 7 6 6 10 7 7 10 8
## [226] 7 9 7 7 9 6 8 10 8 8 7# Assign variables for t-test to use for later results table
alpha = 0.05
CL = 1 - alpha
# Grade 6
fem_pop_g6 <- pop %>%
filter(sex == "f", grade == 6)
mal_pop_g6 <- pop %>%
filter(sex == "m", grade == 6)
fem_pop_g6## sex ageYear ageMonth heightIn weightLb grade
## 1 f 11.92 143 56.3 85.0 6
## 2 f 11.83 142 56.5 69.0 6
## 3 f 11.67 140 53.8 68.5 6
## 4 f 11.58 139 61.5 104.0 6
## 5 f 12.42 149 58.3 93.0 6
## 6 f 11.92 143 51.3 50.5 6
## 7 f 12.08 145 58.8 89.0 6
## 8 f 12.50 150 59.5 78.5 6
## 9 f 12.25 147 61.3 115.0 6
## 10 f 11.75 141 61.8 85.0 6
## 11 f 11.67 140 53.5 81.0 6
## 12 f 12.17 146 56.3 83.5 6
## 13 f 12.42 149 64.3 110.5 6
## 14 f 11.58 139 57.5 96.0 6
## 15 f 12.50 150 61.3 94.0 6
## 16 f 11.58 139 52.8 63.5 6
## 17 f 12.25 147 59.8 84.5 6
## 18 f 12.00 144 59.5 93.5 6
## 19 f 12.17 146 60.0 109.0 6
## 20 f 12.08 145 59.0 91.5 6
## 21 f 12.25 147 55.8 75.0 6
## 22 f 12.08 145 57.8 84.0 6
## 23 f 11.92 143 55.5 84.0 6
## 24 f 12.33 148 56.3 77.0 6
## 25 f 12.25 147 58.3 111.5 6
## 26 f 12.00 144 55.8 73.5 6
## 27 f 11.67 140 60.0 77.0 6
## 28 f 12.33 148 60.5 84.5 6
## 29 f 11.92 143 58.3 77.5 6
## 30 f 12.25 147 59.5 101.0 6
## 31 f 12.33 148 59.0 95.0 6
## 32 f 11.92 143 61.5 116.5 6
## 33 f 11.83 142 56.0 72.5 6
## 34 f 12.50 150 54.5 74.0 6
## 35 f 11.75 141 56.0 72.5 6
## 36 f 12.25 147 51.5 64.0 6
## 37 f 12.00 144 61.0 92.0 6
## 38 f 11.75 141 61.3 85.0 6mal_pop_g6## sex ageYear ageMonth heightIn weightLb grade
## 1 m 12.00 144 57.3 76.5 6
## 2 m 12.50 150 59.5 84.0 6
## 3 m 12.50 150 60.8 128.0 6
## 4 m 11.58 139 60.5 87.0 6
## 5 m 12.25 147 50.5 79.0 6
## 6 m 12.17 146 57.5 90.0 6
## 7 m 11.75 141 53.3 84.0 6
## 8 m 12.50 150 59.0 99.5 6
## 9 m 11.67 140 59.5 94.5 6
## 10 m 12.17 146 57.3 83.0 6
## 11 m 11.67 140 56.5 84.0 6
## 12 m 12.00 144 62.8 94.0 6
## 13 m 11.92 143 57.5 75.0 6
## 14 m 12.00 144 59.5 88.0 6
## 15 m 12.42 149 57.0 92.0 6
## 16 m 12.00 144 60.0 117.5 6
## 17 m 12.25 147 57.0 84.0 6
## 18 m 12.50 150 59.5 84.0 6
## 19 m 11.67 140 58.5 86.5 6
## 20 m 12.00 144 57.0 84.0 6
## 21 m 12.42 149 52.5 81.0 6
## 22 m 11.83 142 55.0 70.0 6
## 23 m 11.83 142 58.8 84.0 6
## 24 m 12.08 145 56.5 91.0 6
## 25 m 11.92 143 57.5 101.0 6
## 26 m 12.50 150 59.0 98.0 6
## 27 m 12.50 150 61.8 118.0 6
## 28 m 11.75 141 57.5 85.0 6
## 29 m 11.83 142 56.0 87.5 6
## 30 m 12.33 148 60.5 118.0 6
## 31 m 11.67 140 56.8 83.5 6
## 32 m 12.00 144 60.0 89.0 6
## 33 m 12.42 149 56.3 72.0 6
## 34 m 12.17 146 55.0 71.5 6
## 35 m 11.58 139 55.0 73.5 6
## 36 m 11.83 142 55.0 76.0 6ttest_g6 <- t.test(fem_pop_g6$weightLb, mal_pop_g6$weightLb, conf.level=CL)
ttest_g6 # pvalue = .3914, t = -.86 ##
## Welch Two Sample t-test
##
## data: fem_pop_g6$weightLb and mal_pop_g6$weightLb
## t = -0.86224, df = 71.85, p-value = 0.3914
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -9.580244 3.795156
## sample estimates:
## mean of x mean of y
## 85.81579 88.70833# Grade 7
fem_pop_g7 <- pop %>%
filter(sex == "f", grade == 7)
mal_pop_g7 <- pop %>%
filter(sex == "m", grade == 7)
fem_pop_g7## sex ageYear ageMonth heightIn weightLb grade
## 1 f 12.92 155 62.3 105.0 7
## 2 f 12.75 153 63.3 108.0 7
## 3 f 13.42 161 59.0 92.0 7
## 4 f 13.33 160 62.0 94.5 7
## 5 f 13.08 157 64.5 123.5 7
## 6 f 12.92 155 61.3 107.0 7
## 7 f 12.83 154 60.0 114.0 7
## 8 f 13.00 156 54.5 75.0 7
## 9 f 12.83 154 62.8 93.5 7
## 10 f 12.67 152 60.5 105.0 7
## 11 f 13.08 157 60.5 112.0 7
## 12 f 12.83 154 61.0 122.5 7
## 13 f 12.92 155 66.0 144.5 7mal_pop_g7## sex ageYear ageMonth heightIn weightLb grade
## 1 m 13.08 157 60.5 105.0 7
## 2 m 13.33 160 60.5 84.0 7
## 3 m 13.00 156 61.8 112.0 7
## 4 m 12.58 151 66.3 117.0 7
## 5 m 12.75 153 60.0 84.0 7
## 6 m 12.58 151 58.3 86.0 7
## 7 m 12.58 151 61.0 81.0 7
## 8 m 13.33 160 59.3 78.5 7
## 9 m 13.00 156 66.3 106.0 7
## 10 m 13.08 157 58.0 80.5 7
## 11 m 13.00 156 58.3 92.5 7
## 12 m 13.00 156 61.5 108.5 7
## 13 m 13.17 158 65.0 121.0 7
## 14 m 13.00 156 68.5 114.0 7
## 15 m 12.67 152 59.5 105.0 7
## 16 m 12.92 155 61.8 91.5 7
## 17 m 13.33 160 64.0 116.0 7
## 18 m 13.25 159 63.3 112.0 7
## 19 m 12.67 152 60.8 97.0 7
## 20 m 13.42 161 56.8 75.0 7
## 21 m 12.75 153 64.8 128.0 7
## 22 m 13.25 159 62.8 99.0 7
## 23 m 12.75 153 57.8 79.5 7
## 24 m 12.92 155 57.3 80.5 7
## 25 m 12.58 151 59.3 87.0 7ttest_g7 <- t.test(fem_pop_g7$weightLb, mal_pop_g7$weightLb, conf.level=CL)
ttest_g7 #p-value .101, t = 1.71, cannot reject null ##
## Welch Two Sample t-test
##
## data: fem_pop_g7$weightLb and mal_pop_g7$weightLb
## t = 1.7103, df = 22.434, p-value = 0.101
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.070286 21.676440
## sample estimates:
## mean of x mean of y
## 107.4231 97.6200# Grade 8
fem_pop_g8 <- pop %>%
filter(sex == "f", grade == 8)
mal_pop_g8 <- pop %>%
filter(sex == "m", grade == 8)
fem_pop_g8## sex ageYear ageMonth heightIn weightLb grade
## 1 f 14.25 171 62.5 112.0 8
## 2 f 13.67 164 58.0 83.5 8
## 3 f 13.83 166 61.5 103.5 8
## 4 f 14.17 170 64.3 90.0 8
## 5 f 13.50 162 58.0 84.0 8
## 6 f 14.08 169 62.3 99.5 8
## 7 f 14.42 173 62.8 102.5 8
## 8 f 13.83 166 59.3 89.5 8
## 9 f 14.00 168 61.5 95.0 8
## 10 f 14.08 169 62.0 98.5 8
## 11 f 13.92 167 62.3 92.5 8
## 12 f 13.67 164 65.3 98.0 8
## 13 f 14.25 171 61.5 91.0 8
## 14 f 14.33 172 64.8 142.0 8
## 15 f 13.75 165 61.3 106.5 8
## 16 f 13.75 165 55.5 67.0 8
## 17 f 13.58 163 56.5 84.0 8
## 18 f 14.25 171 63.0 84.0 8
## 19 f 13.92 167 61.0 93.5 8
## 20 f 13.67 164 63.3 108.0 8
## 21 f 14.08 169 61.5 85.0 8mal_pop_g8## sex ageYear ageMonth heightIn weightLb grade
## 1 m 13.75 165 64.8 98.0 8
## 2 m 14.42 173 61.3 93.0 8
## 3 m 13.67 164 57.8 95.0 8
## 4 m 13.50 162 64.5 119.0 8
## 5 m 13.67 164 60.5 95.0 8
## 6 m 14.42 173 69.0 112.5 8
## 7 m 14.17 170 63.8 112.5 8
## 8 m 14.50 174 66.0 108.0 8
## 9 m 13.67 164 63.5 108.0 8
## 10 m 14.08 169 62.0 100.0 8
## 11 m 14.33 172 65.0 112.0 8
## 12 m 14.00 168 60.0 93.5 8
## 13 m 14.00 168 66.5 111.5 8
## 14 m 14.50 174 69.8 119.5 8
## 15 m 13.83 166 62.5 84.0 8
## 16 m 13.58 163 65.3 117.5 8
## 17 m 13.83 166 67.3 121.0 8
## 18 m 14.42 173 66.0 112.0 8
## 19 m 13.50 162 60.0 105.0 8
## 20 m 13.83 166 62.0 91.0 8
## 21 m 13.58 163 66.0 112.0 8
## 22 m 14.25 171 61.8 112.0 8
## 23 m 13.50 162 63.0 91.0 8
## 24 m 14.50 174 63.0 112.0 8
## 25 m 13.67 164 58.0 84.0 8
## 26 m 13.67 164 66.5 112.0 8
## 27 m 13.67 164 61.5 140.0 8
## 28 m 13.92 167 62.0 107.5 8ttest_g8 <- t.test(fem_pop_g8$weightLb, mal_pop_g8$weightLb, conf.level=CL)
ttest_g8 #p-value = .011 t = -2.655, reject null, significant weight difference. ##
## Welch Two Sample t-test
##
## data: fem_pop_g8$weightLb and mal_pop_g8$weightLb
## t = -2.6553, df = 38.985, p-value = 0.01142
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -18.823615 -2.545433
## sample estimates:
## mean of x mean of y
## 95.69048 106.37500# Grade 9
fem_pop_g9 <- pop %>%
filter(sex == "f", grade == 9)
mal_pop_g9 <- pop %>%
filter(sex == "m", grade == 9)
fem_pop_g9## sex ageYear ageMonth heightIn weightLb grade
## 1 f 15.42 185 59.0 104.0 9
## 2 f 14.83 178 61.5 103.5 9
## 3 f 15.00 180 63.3 114.0 9
## 4 f 14.67 176 61.3 112.0 9
## 5 f 15.42 185 63.3 101.0 9
## 6 f 14.58 175 60.8 93.5 9
## 7 f 15.00 180 59.0 112.0 9
## 8 f 14.75 177 61.8 142.5 9
## 9 f 15.42 185 65.3 118.0 9
## 10 f 15.17 182 58.3 104.5 9
## 11 f 15.33 184 62.3 108.0 9
## 12 f 14.75 177 61.3 112.0 9
## 13 f 14.83 178 63.5 148.5 9
## 14 f 15.25 183 64.3 109.5 9
## 15 f 15.25 183 64.5 102.5 9
## 16 f 15.42 185 60.0 106.0 9
## 17 f 14.83 178 66.5 117.5 9
## 18 f 14.75 177 61.3 81.0 9
## 19 f 15.25 183 66.5 112.0 9
## 20 f 14.92 179 63.0 98.5 9
## 21 f 15.17 182 65.5 133.0 9
## 22 f 15.17 182 62.0 91.5 9
## 23 f 15.17 182 64.0 111.5 9
## 24 f 14.58 175 60.3 86.0 9
## 25 f 15.00 180 61.3 110.5 9mal_pop_g9## sex ageYear ageMonth heightIn weightLb grade
## 1 m 15.25 183 64.8 111.0 9
## 2 m 14.67 176 63.8 98.5 9
## 3 m 14.67 176 65.0 118.5 9
## 4 m 15.42 185 66.0 105.0 9
## 5 m 15.00 180 61.8 104.0 9
## 6 m 15.25 183 66.0 105.5 9
## 7 m 14.83 178 67.3 119.5 9
## 8 m 14.58 175 64.0 92.0 9
## 9 m 14.58 175 68.0 112.0 9
## 10 m 14.58 175 63.5 98.5 9
## 11 m 15.33 184 66.5 112.0 9
## 12 m 14.67 176 61.5 81.0 9
## 13 m 15.17 182 67.0 133.0 9
## 14 m 14.75 177 63.0 111.0 9
## 15 m 14.75 177 60.5 112.0 9
## 16 m 14.58 175 65.5 114.0 9
## 17 m 14.83 178 63.8 112.0 9
## 18 m 14.83 178 63.5 102.5 9ttest_g9 <- t.test(fem_pop_g9$weightLb, mal_pop_g9$weightLb, conf.level=CL)
ttest_g9 #p-value .72 , t = 0.35, cannot reject null ##
## Welch Two Sample t-test
##
## data: fem_pop_g9$weightLb and mal_pop_g9$weightLb
## t = 0.35093, df = 40.932, p-value = 0.7274
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -6.805071 9.667293
## sample estimates:
## mean of x mean of y
## 109.3200 107.8889# Grade 10
fem_pop_g10 <- pop %>%
filter(sex == "f", grade == 10)
mal_pop_g10 <- pop %>%
filter(sex == "m", grade == 10)
fem_pop_g10## sex ageYear ageMonth heightIn weightLb grade
## 1 f 15.92 191 62.5 112.5 10
## 2 f 15.92 191 65.3 107.0 10
## 3 f 15.50 186 57.8 95.0 10
## 4 f 16.42 197 61.5 121.0 10
## 5 f 16.42 197 64.8 112.0 10
## 6 f 15.92 191 63.3 113.5 10
## 7 f 15.83 190 66.8 140.0 10
## 8 f 15.75 189 64.3 113.5 10
## 9 f 15.83 190 56.8 98.5 10
## 10 f 15.50 186 57.0 83.5 10
## 11 f 16.08 193 59.8 115.0 10
## 12 f 15.50 186 63.5 108.0 10mal_pop_g10## sex ageYear ageMonth heightIn weightLb grade
## 1 m 15.75 189 67.0 128.0 10
## 2 m 16.08 193 66.3 133.0 10
## 3 m 15.50 186 66.0 112.0 10
## 4 m 15.67 188 67.3 112.0 10
## 5 m 16.08 193 67.8 127.5 10
## 6 m 15.67 188 71.0 140.0 10
## 7 m 15.75 189 66.3 112.0 10
## 8 m 15.67 188 65.8 150.5 10
## 9 m 15.67 188 63.3 115.5 10
## 10 m 16.08 193 72.0 150.0 10
## 11 m 16.17 194 65.3 134.5 10
## 12 m 15.50 186 66.5 112.0 10
## 13 m 16.33 196 64.5 98.0 10
## 14 m 15.75 189 65.0 114.0 10ttest_g10 <- t.test(fem_pop_g10$weightLb, mal_pop_g10$weightLb, conf.level=CL)
ttest_g10 # Pvalue is .022 < .05, t = -2.43 > -2.262, reject null, significant weight difference. ##
## Welch Two Sample t-test
##
## data: fem_pop_g10$weightLb and mal_pop_g10$weightLb
## t = -2.4335, df = 23.972, p-value = 0.02278
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -26.347653 -2.164252
## sample estimates:
## mean of x mean of y
## 109.9583 124.21432.2) The null hypothesis
The null hypothesis is that the there is no significant correlation between sex and height for each grade. We’ll know if the null hypothesis is true by looking at the p-value from each t-test. If the p-value is greater than 0.05, we can accept the null hypothesis. If the p-value is less than 0.05, we can reject the null hypothesis.
Results:
Grade 6
Is 0.391 > 0.05? TRUE
Grade 7
Is 0.101 > 0.05? TRUE
Grade 8
Is 0.0114 > 0.05? FALSE
Grade 9
Is 0.727 > 0.05? TRUE
Grade 10
Is 0.0228 > 0.05? FALSE
For grades 6, 7, and 9 the null hypothesis is TRUE. This means we’re 95% confident that there’s no statistical correlation between sex and weight.
For grades 8 and 10, the null hypothesis is FALSE. This means we can’t confidently rule out a statistical correlation between sex and height. In other words, its possible that a reason for weight difference is gender.
2.3) Optional/Unfinished
Overall, growth trajectories vary year by year. Some years, we are able to observe a statistical relationship between weight and gender, and other years we are 95% confident that there is not a difference.
I added the means and the p-value as labels to the visuals below and it made it easier for me to see the takeaways. In 6 grade, men and women on average, are close in weight. We are 95% confident that there is not a statistical difference based on gender. In 7th grade, while the sample of women weighted more than men, on average, we did not find statistical correlation since the p-value was less than the confidence interval. Similar to 6th grade, the difference may have been due to chance. In 8th grade, on average, men weighted more than women. It’s possible that there is a difference based on gender, since the p-value was less than the confidence interval. In 9th grade, we were 95% confident that there is no difference between the genders. However in 10th grade, we can’t confidently rule it out.
# Combine data frames created for each grade above
# Grade 6
tf6 <- t.test(fem_pop_g6$weightLb)
tm6 <- t.test(mal_pop_g6$weightLb)
g6 <- bind_rows(fem_pop_g6, mal_pop_g6)
g6plot <- ggplot(g6, aes(x= weightLb, color=factor(sex))) +
geom_density() +
geom_vline(aes(xintercept=tf6$estimate), color="#F8766D", size=1.25) +
geom_vline(aes(xintercept=tf6$conf.int[1]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tf6$conf.int[2]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm6$estimate), color="#00BFC4", size=1.25) +
geom_vline(aes(xintercept=tm6$conf.int[1]), color="#00BFC4", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm6$conf.int[2]), color="#00BFC4", alpha=.75, linetype="dashed") +
theme(
legend.position = "none"
) +
labs(title="Grade 6")+
geom_label(data=mal_pop_g6, aes(x=mean(mal_pop_g6$weightLb), y=.02), label= paste0("Mean:", round(mean(mal_pop_g6$weightLb),)))+
geom_label(data=fem_pop_g6, aes(x=mean(fem_pop_g6$weightLb), y=.04), label= paste0("Mean:", round(mean(fem_pop_g6$weightLb),)))+
geom_label(aes(x=75, y=.04), color = "black", label= paste0("Pvalue:", round(ttest_g6$p.value, digits = 2)))
g6plot## Warning: Use of `mal_pop_g6$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g6$weightLb` is discouraged. Use `weightLb` instead.
# Grade 7
tf7 <- t.test(fem_pop_g7$weightLb)
tm7 <- t.test(mal_pop_g7$weightLb)
g7 <- bind_rows(fem_pop_g7, mal_pop_g7)
g7plot <- ggplot(g7, aes(x= weightLb, color=factor(sex))) +
geom_density() +
geom_vline(aes(xintercept=tf7$estimate), color="#F8766D", size=1.25) +
geom_vline(aes(xintercept=tf7$conf.int[1]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tf7$conf.int[2]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm7$estimate), color="#00BFC4", size=1.25) +
geom_vline(aes(xintercept=tm7$conf.int[1]), color="#00BFC4", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm7$conf.int[2]), color="#00BFC4", alpha=.75, linetype="dashed") +
theme(
legend.position = "none"
) +
labs(title="Grade 7")+
geom_label(data=mal_pop_g7, aes(x=mean(mal_pop_g7$weightLb), y=.02), label= paste0("Mean:", round(mean(mal_pop_g7$weightLb),)))+
geom_label(data=fem_pop_g7, aes(x=mean(fem_pop_g7$weightLb), y=.04), label= paste0("Mean:", round(mean(fem_pop_g7$weightLb),)))+
geom_label(aes(x=130, y=.04), color = "black", label= paste0("Pvalue:", round(ttest_g7$p.value, digits = 2)))
g7plot## Warning: Use of `mal_pop_g7$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g7$weightLb` is discouraged. Use `weightLb` instead.
# Grade 8
tf8 <- t.test(fem_pop_g8$weightLb)
tm8 <- t.test(mal_pop_g8$weightLb)
g8 <- bind_rows(fem_pop_g8, mal_pop_g8)
g8plot <- ggplot(g8, aes(x= weightLb, color=factor(sex))) +
geom_density() +
geom_vline(aes(xintercept=tf8$estimate), color="#F8766D", size=1.25) +
geom_vline(aes(xintercept=tf8$conf.int[1]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tf8$conf.int[2]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm8$estimate), color="#00BFC4", size=1.25) +
geom_vline(aes(xintercept=tm8$conf.int[1]), color="#00BFC4", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm8$conf.int[2]), color="#00BFC4", alpha=.75, linetype="dashed") +
theme(
legend.position = "none"
) +
labs(title="Grade 8")+
geom_label(data=mal_pop_g8, aes(x=mean(mal_pop_g8$weightLb), y=.02), label= paste0("Mean:", round(mean(mal_pop_g8$weightLb),)))+
geom_label(data=fem_pop_g8, aes(x=mean(fem_pop_g8$weightLb), y=.04), label= paste0("Mean:", round(mean(fem_pop_g8$weightLb),)))+
geom_label(aes(x=130, y=.04), color = "black", label= paste0("Pvalue:", round(ttest_g8$p.value, digits = 2)))
g8plot## Warning: Use of `mal_pop_g8$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g8$weightLb` is discouraged. Use `weightLb` instead.
# Grade 9
tf9 <- t.test(fem_pop_g9$weightLb)
tm9 <- t.test(mal_pop_g9$weightLb)
g9 <- bind_rows(fem_pop_g9, mal_pop_g9)
g9plot <- ggplot(g9, aes(x= weightLb, color=factor(sex))) +
geom_density() +
geom_vline(aes(xintercept=tf9$estimate), color="#F8766D", size=1.25) +
geom_vline(aes(xintercept=tf9$conf.int[1]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tf9$conf.int[2]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm9$estimate), color="#00BFC4", size=1.25) +
geom_vline(aes(xintercept=tm9$conf.int[1]), color="#00BFC4", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm9$conf.int[2]), color="#00BFC4", alpha=.75, linetype="dashed") +
theme(
legend.position = "none"
) +
labs(title="Grade 9")+
geom_label(data=mal_pop_g9, aes(x=mean(mal_pop_g9$weightLb), y=.02), label= paste0("Mean:", round(mean(mal_pop_g9$weightLb),)))+
geom_label(data=fem_pop_g9, aes(x=mean(fem_pop_g9$weightLb), y=.04), label= paste0("Mean:", round(mean(fem_pop_g9$weightLb),)))+
geom_label(aes(x=130, y=.04), color = "black", label= paste0("Pvalue:", round(ttest_g9$p.value, digits = 2)))
g9plot## Warning: Use of `mal_pop_g9$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g9$weightLb` is discouraged. Use `weightLb` instead.
# Grade 10
tf10 <- t.test(fem_pop_g10$weightLb)
tm10 <- t.test(mal_pop_g10$weightLb)
g10 <- bind_rows(fem_pop_g10, mal_pop_g10)
g10plot <- ggplot(g10, aes(x= weightLb, color=factor(sex))) +
geom_density() +
geom_vline(aes(xintercept=tf10$estimate), color="#F8766D", size=1.25) +
geom_vline(aes(xintercept=tf10$conf.int[1]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tf10$conf.int[2]), color="#F8766D", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm10$estimate), color="#00BFC4", size=1.25) +
geom_vline(aes(xintercept=tm10$conf.int[1]), color="#00BFC4", alpha=.75, linetype="dashed") +
geom_vline(aes(xintercept=tm10$conf.int[2]), color="#00BFC4", alpha=.75, linetype="dashed") +
theme(
legend.position = "none"
) +
labs(title="Grade 10")+
geom_label(data=mal_pop_g10, aes(x=mean(mal_pop_g10$weightLb), y=.02), label= paste0("Mean:", round(mean(mal_pop_g10$weightLb),)))+
geom_label(data=fem_pop_g10, aes(x=mean(fem_pop_g10$weightLb), y=.04), label= paste0("Mean:", round(mean(fem_pop_g10$weightLb),)))+
geom_label(aes(x=130, y=.04), color = "black", label= paste0("Pvalue:", round(ttest_g10$p.value, digits = 2)))
g10plot## Warning: Use of `mal_pop_g10$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g10$weightLb` is discouraged. Use `weightLb` instead.
all_plots <- plot_grid(
g6plot,
g7plot,
g8plot,
g9plot,
g10plot,
align = c("hv"),
nrow = 3
)## Warning: Use of `mal_pop_g6$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g6$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `mal_pop_g7$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g7$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `mal_pop_g8$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g8$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `mal_pop_g9$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g9$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `mal_pop_g10$weightLb` is discouraged. Use `weightLb` instead.## Warning: Use of `fem_pop_g10$weightLb` is discouraged. Use `weightLb` instead.all_plots