Point estimate is the sample mean, 171.1….Median is 170.3
SD is 9.4….IQR is Q3 - Q1 so… 14 <<< Calculations can be seen in R markdown for this answer and future.
For different answers, lets define unusual as within 1 standard deviation. If so then our range of usual is 161.7 - 180.5.
180cm is not unusual.
155cm is unusual.
I would expect the mean and SD of the new sample to be similar, but not the same. If the individuals surveyed were selected with current scientific standards then any sample over 30 of these individuals should resemble a very similar distribution with roughly the same numbers. (of course there is always room for error!)
SE = sSD/sqrt(sampleSize)…0.4174687
FALSE : We are 100% confident that the average spending “OF THESE 436 AMERICAN ADULTS” is between 80-89. Although these 436 Americans are our sample of America, we are making a statement about the sample, making them the population.
FALSE : If the sample size was lower than 30, this would be a problem; however our sample size is 436.
FALSE : The CI is constructed from our sample mean, we can not induce the same interval from a different sample mean.
TRUE : This is what confidence intervals are supposed to tell us in the first place.
TRUE : A 90% confidence interval is narrower, because when we do not need to be as confident in our answer, we can afford to be a bit more specific.
FALSE : We have to remember that the formula calls for the square root of the sample size, therefore 3^2
TRUE : ME = upper tail - lower / 2 … 4.4 = 89.11 - 80.31 / 2
Yes
####(b) Suppose you read online that children ???rst count to 10 successfully when they are 32 months old, on average. Perform a hypothesis test to evaluate if these data provide convincing evidence that the average age at which gifted children ???st count to 10 successfully is less than the general average of 32 months. Use a signi???cance level of 0.10.
Null hypothesis : The true population mean is 32
Alt hypothesis : The true population mean is not 32
n = 36
x = 32
mean = 30.69
sd = 4.31
standardError = sd/sqrt(n)
Z = (mean-x)/standardError
p = pnorm(Z,lower.tail = TRUE)
p
## [1] 0.0341013
A 3% chance of getting the observed mean if the proposed mean was true, MEANS (pun intended) we will reject the null hypothesis in favor of the alternative hypothesis.
#### (c) Interpret the p-value in context of the hypothesis test and the data.
If the null hypothesis were true, then we would have missed a 97% probability that it wasn’t.
####(d) Calculate a 90% con???dence interval for the average age at which gifted children ???rst count to 10 successfully
lowerTail = mean - 1.645 * standardError
upperTail = mean + 1.645 * standardError
lowerTail
## [1] 29.50834
upperTail
## [1] 31.87166
Yes, if the proposed mean was probable, it would not be higher than 31.87, but it is.
null Hypothesis : ?? = 100 alt Hypothesis : ?? != 100
n = 36
mean = 118.2
sd = 6.5
x = 100
standardError = sd/sqrt(n)
Z = (mean-x)/standardError
Z
## [1] 16.8
Thats a big Z, probability around 0. Thats smaller than our significance level. I reject null hypothesis for the alternative hypothesis.
####(b) Calculate a 90% con???dence interval for the average IQ of mothers of gifted children.
lowerTail = mean - 1.645 * standardError
upperTail = mean + 1.645 * standardError
lowerTail
## [1] 116.4179
upperTail
## [1] 119.9821
Yes, the p-value is very small, and the proposed population mean is too far away from our confidence interval.
Sampling distribution of the mean, is when we repeatedly take samples of a population, and distribute the means we find from those samples into its own distribution.
1 - pnorm(10500,mean=9000,sd=1000)
## [1] 0.0668072
Could be anything, sample size is not big enough to predict any outcome; central limit theorom applies, if anything it should be near normal distribution.
####(c) What is the probability that the mean lifespan of 15 randomly chosen light bulbs is more than 10,500 hours?
1-pnorm(10500,mean=9000,sd=258)
## [1] 3.050719e-09
x = 7000:11000
y = dnorm(x,mean=9000,sd=1000)
X = 7000:11000
Y = dnorm(X,mean = 9000,sd=258)
plot(x,y)
plot(X,Y)
####(e) Could you estimate the probabilities from parts (a) and (c) if the lifespans of light bulbs had a skewed distribution Nope, 30+ sample size needed for that.
When you get your point estimates from a bigger sample, the variability, along with the P value gets smaller. So the P-value will decrease, pretty significantly from this mistake.