Heights of adults. (7.7, p. 260) Researchers studying anthropometry collected body girth measurements and skeletal diameter measurements, as well as age, weight, height and gender, for 507 physically active individuals. The histogram below shows the sample distribution of heights in centimeters.
library(openintro)
## Please visit openintro.org for free statistics materials
##
## Attaching package: 'openintro'
## The following objects are masked from 'package:datasets':
##
## cars, trees
data(bdims)
par(mar=c(3.7,2.5,0.5,0.5), las=1, mgp=c(2.5,0.7,0), cex.lab = 1.5)
histPlot(bdims$hgt, col = COL[1], xlab = "Height", ylab = "")
The average height is 171.1. The median is 170.3.
summary(bdims$hgt)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 147.2 163.8 170.3 171.1 177.8 198.1
The standard deviation is 9.407205. The IQR is 14.
sd(bdims$hgt)
## [1] 9.407205
IQR(bdims$hgt)
## [1] 14
If the average is 171.1, 180 cm will be considered tall and 155 cm will considered short.
mean<-mean(bdims$hgt)
n<-180
if (n>mean){
print("Tall")
}else if(n<mean){
print("Short")
}
## [1] "Tall"
n<-155
if (n>mean){
print("Tall")
}else if(n<mean){
print("Short")
}
## [1] "Short"
No because another random sample will give us a new set of mean and standard deviation.
# $SD_x = \frac{\sigma}{\sqrt{n}}$)
n<-507
hgt_sd<-sd(bdims$hgt)
hgt_mean<-mean(bdims$hgt)
hgt_sd / sqrt(n)
## [1] 0.4177887
Thanksgiving spending, Part I. The 2009 holiday retail season, which kicked off on November 27, 2009 (the day after Thanksgiving), had been marked by somewhat lower self-reported consumer spending than was seen during the comparable period in 2008. To get an estimate of consumer spending, 436 randomly sampled American adults were surveyed. Daily consumer spending for the six-day period after Thanksgiving, spanning the Black Friday weekend and Cyber Monday, averaged $84.71. A 95% confidence interval based on this sample is ($80.31, $89.11). Determine whether the following statements are true or false, and explain your reasoning.
library(openintro)
data(tgSpending)
par(mar=c(3.7,2.2,0.5,0.5), las=1, mgp=c(2.5,0.7,0), cex.lab = 1.5)
histPlot(tgSpending$spending, col = COL[1], xlab = "Spending", ylab = "")
[1] 80.30173 89.11180 False. We are close though.
n<-436
tg_mean<-mean(tgSpending$spending)
tg_sd<-sd(tgSpending$spending)
lower_vector <- tg_mean - 1.96 * tg_sd / sqrt(n)
upper_vector <- tg_mean + 1.96 * tg_sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 80.30173 89.11180
False. Based on the Central Limit Theorem, if the sample size selected at a random and is greater than 30, the sample mean is normally distributed and its confidence interval is valid.
True.
n<-436
tg_mean<-mean(tgSpending$spending)
tg_sd<-sd(tgSpending$spending)
lower_vector <- tg_mean - 1.96 * tg_sd / sqrt(n)
upper_vector <- tg_mean + 1.96 * tg_sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 80.30173 89.11180
True, 95% CI covers up to 2 standard deviations of the American adults.
True.
n<-436
tg_mean<-mean(tgSpending$spending)
tg_sd<-sd(tgSpending$spending)
lower_vector <- tg_mean - 1.65 * tg_sd / sqrt(n)
upper_vector <- tg_mean + 1.65 * tg_sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 80.99844 88.41509
False.
n=436
1.96*(tg_sd / sqrt(n))
## [1] 4.405038
m=436*3
1.96*(tg_sd / sqrt(m))
## [1] 2.54325
True.
n=436
1.96*(tg_sd / sqrt(n))
## [1] 4.405038
Gifted children, Part I. Researchers investigating characteristics of gifted children col- lected data from schools in a large city on a random sample of thirty-six children who were identified as gifted children soon after they reached the age of four. The following histogram shows the dis- tribution of the ages (in months) at which these children first counted to 10 successfully. Also provided are some sample statistics.
library(openintro)
data(gifted)
par(mar=c(3.7,2.2,0.5,0.5), las=1, mgp=c(2.5,0.7,0), cex.lab = 1.5)
histPlot(gifted$count, col = COL[1],
xlab = "Age child first counted to 10 (in months)", ylab = "",
axes = FALSE)
axis(1)
axis(2, at = c(0,3,6))
Yes, sample is randomly selected and have more than 30 sample size.
Since the significance level is 0.10, the confidence interval will be 0.90. So we will be using +/- 1.645.
n<-36
mean<-30.69
sd<-4.31
lower_vector<- mean - 1.645 * sd / sqrt(n)
upper_vector<- mean + 1.645 * sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 29.50834 31.87166
Our data shows that on 90% confidence that the general average counts to 10 between ages 29.50834 to 31.87166 (months).
Z_score <- (30.69 - 32) / (4.31 / sqrt(36)); Z_score
## [1] -1.823666
p_value <- pnorm(Z_score); p_value
## [1] 0.0341013
The p_value is less than 0.10 so we can reject the hypothesis.
n<-36
mean<-30.69
sd<-4.31
lower_vector<- mean - 1.645 * sd / sqrt(n)
upper_vector<- mean + 1.645 * sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 29.50834 31.87166
Yes, because confidence interval and significance levels correlate. They both total up to 1. For example, if CI= 0.90, 1 minus 0.90= 0.10 for the significance levels.
Gifted children, Part II. Exercise above describes a study on gifted children. In this study, along with variables on the children, the researchers also collected data on the mother’s and father’s IQ of the 36 randomly sampled gifted children. The histogram below shows the distribution of mother’s IQ. Also provided are some sample statistics.
library(openintro)
data(gifted)
par(mar=c(3.7,2.2,0.5,0.5), las=1, mgp=c(2.5,0.7,0), cex.lab = 1.5)
histPlot(gifted$motheriq, col = COL[1],
xlab = "Mother's IQ", ylab = "", axes = FALSE)
axis(1)
axis(2, at = c(0,4,8,12))
n=36
sd=6.5
H0=118.2
HA=100
Z_score <- (H0 - HA) / (sd/ sqrt(n)); Z_score
## [1] 16.8
p_value <- 2 * (pnorm(Z_score, 0, 1, lower.tail = FALSE)); p_value
## [1] 2.44044e-63
n<-36
mean<-118.2
sd<-6.5
lower_vector<- mean - 1.645 * sd / sqrt(n)
upper_vector<- mean + 1.645 * sd / sqrt(n)
c(lower_vector, upper_vector)
## [1] 116.4179 119.9821
Yes they agree. The p value is small so we reject the null hypothesis. In the confidence interval, it does not include the average of 100 IQ for the population.
CLT. Define the term “sampling distribution” of the mean, and describe how the shape, center, and spread of the sampling distribution of the mean change as sample size increases.
Sampling distribution is the randomly selection of samples from a population and calculating the mean, and then creating a distribution from the results of the mean values. As the sample size increases, the spread gets smaller and the shape will get taller and more sharp. The center will stay relatively the same because that is the population mean.
CFLBs. A manufacturer of compact fluorescent light bulbs advertises that the distribution of the lifespans of these light bulbs is nearly normal with a mean of 9,000 hours and a standard deviation of 1,000 hours.
lb_mean<-9000 lb_sd<-1000
H0<-9000 HA<-10500 (a) What is the probability that a randomly chosen light bulb lasts more than 10,500 hours?
pnorm(10500, 9000, 1000, lower.tail = FALSE)
## [1] 0.0668072
sd <- 1000
mean <- 9000
sample_sd <- sd/sqrt(15)
sample_sd
## [1] 258.1989
pnorm(10500, 9000, 1000/sqrt(15), lower.tail = FALSE)
## [1] 3.133452e-09
par(mfrow = c(2, 1))
xpop <- 7000:12000
ypop <- dnorm(xpop,mean=9000,sd=1000)
xsample <- 7000:12000
ysample <- dnorm(xsample,mean=9000,sd=1000/sqrt(15))
plot(xpop,ypop)
plot(xsample, ysample)
It will be hard to estime the probabilities because the sample size is less than 30, it would’nt be a normal distribution based the Central Limit Theorem.
Same observation, different sample size. Suppose you conduct a hypothesis test based on a sample where the sample size is n = 50, and arrive at a p-value of 0.08. You then refer back to your notes and discover that you made a careless mistake, the sample size should have been n = 500. Will your p-value increase, decrease, or stay the same? Explain.
The p-value will decrease as the sample size increases. Sample size increases cause the standard of error to decrease, which lowers the standard deviations from the population mean which lowers the p-value.