Sec 7.
4 | MATH 1342
Section 7.4
Confidence Interval for a Population Mean
Recall the formula for a confidence interval is statistic ± margin of error. When we are making
an inference about a population mean, the statistic will be our sample mean, x .
The assumptions for a population mean are:
1. The sample must be an SRS from the population of interest.
2. The data must come from a normally distributed population. If this is not the case or if
we are unsure whether the population is normally distributed, the sampling distribution of
x must be normally distributed. (Recall from section 4.4 that we can assume that the
sampling distribution of x is normal for values of n ³ 30.)
The critical value we use to find the margin of error for our calculation will be based on whether
the population or sample standard deviation is known.
s
To determine our confidence interval we will use x ± z * if the population standard deviation
n
s
is known. We will use x ± t * if the population standard deviation is unknown. (So we’ll be
n
given the sample standard deviation or we’ll be able to find it.)
In this formula, we see a new critical value, t*. Since the statistic, s, is a random variable, it
follows an entirely new and different distribution called the t-distribution.
There is only one z-distribution, but many different t-distributions based on sample size. The
particular form of the t-distribution is determined by its degrees of freedom (df) – the sample
size minus one.
t-distribution vs. Standard Normal Distribution
1
Sec 7.4 | MATH 1342
• The t-distribution is centered at zero, symmetric and bell-shaped, but has heavier tails,
meaning that it is more prone to producing values that fall far from its mean.
• The shape of it depends on the value of n. The higher the n, the more it’ll look like a
standard normal distribution.
æ 1 + CL ö
To find critical values for a t-distribution, the R command is: qt ç , df ÷
è 2 ø
2
Sec 7.4 | MATH 1342
Example: Suppose your class is investigating the weights of Snickers 1-ounce fun-size candy
bars to see if customers are getting full value for their money. Assume that the weights are
normally distributed with standard deviation s = 0.005 ounces (note, this is the population
standard deviation because it uses s ). Several candy bars are randomly selected and weighed
with sensitive balances borrowed from the physics lab.
The weights are: 0.95 1.02 0.98 0.97 1.05 1.01 0.98 1.00
We want to determine a 90% confidence interval for the true mean, µ .
a. What is the sample mean?
b. Determine the critical value
c. Determine the 90% confidence interval.
Confidence Interval:
d. Write a sentence that explains the significance of the confidence interval.
But notice that the population claim, 1 oz, is not even in our confidence interval. This raises a flag! Maybe we
need to question the weight or maybe we just need to take a bigger sample. Or maybe if we use a 95% confidence
interval then the population claim would be in our interval (because the second number is almost 1).
3
Sec 7.4 | MATH 1342
Example: An SRS of 16 seniors from HISD had a mean SAT-math score of 500 and a standard
deviation of 100. We know that the population of SAT-math scores for seniors in the district is
approximately normally distributed.
n= x= s=
a. Find the 90% confidence interval for the mean SAT-math score for the population of all
seniors in the district.
b. Explain the meaning of the above confidence interval.
4
Sec 7.4 | MATH 1342
Like with population proportions, sometimes we are asked to find the minimum sample size
needed to produce a particular margin of error given a certain confidence level. When working
with a one-sample mean, the population standard deviation needs to be known (or estimated) and
s
we can use the formula Maximum ME ³ z * .
n
Whenever n is unknown, always use z * .
Example: An evaluator wishes to make a statement about the emotional maturity of the
freshman population, so she decides to sample the population and administer an emotional
maturity test. Putting aside any concerns about the validity of the test used, and considering
sampling techniques only, how many students should she sample in order to be 95% confident
that her estimate of freshman emotional maturity will be within 6 units of the true mean? (The
test publishers indicate the population variance is 100 units.)
5
Sec 7.4 | MATH 1342
Example: The effect of exercise on the amount of lactic acid in the blood was examined in an
article for an exercise and sport magazine. Eight males were selected at random from those
attending a week-long training camp. Blood lactate levels were measured before and after
playing three games of racquetball, as shown in the accompanying table. Use this data to
estimate the mean increase in blood lactate level using a 95% confidence interval.
Player 1 2 3 4 5 6 7 8
Before 13 20 17 13 13 16 15 16
After 18 37 40 35 30 20 33 19
Since we want to estimate the mean increase, we must first find it.
Player 1 2 3 4 5 6 7 8
Mean Increase
6
Sec 7.4 | MATH 1342
Summary of Formulas to Use
pˆ (1 - pˆ )
• If our statistic is p̂ , then statistic ± margin of error = pˆ ± z * ×
n
• If our statistic is x , then statistic ± margin of error is either
s
x ± z* (if we know the population standard deviation) or
n
s
x ± t* (if we know the sample standard deviation).
n