Week 6

Hypothesis testing:
One population
Test of hypothesis about one mean

If x is the value of mean of random sample of

size n from a normal population with the
known variance , then our statistical test is,

x
z

n
 In the case of nonnormal population provided
that n is sufficiently large ( n 30), we
may also substitute for s the value of the
sample standard deviation, then our
statistical test is,

x
z
s
n
 Example 2
"The University of Minnesota uses thousands of fluorescent
light bulb each year. The brand of bulb it currently uses
has a mean life of 900 hours. A manufacturer claims that
its new brand of bulbs, which cost the same as the brand
the university currently uses, has a mean life of more than
900 hours. The university has decided to purchase the new
brand if, when tested, the test evidence supports the
manufacturers claim at the 0.05 significance level.
Suppose 64 bulbs were tested with the following results:

x = 920 hours s = 80 hours

Will the University of Minnesota purchase the new brand

of fluorescent bulbs?
 Traditional method
P-value based method
 Exercise 1
A large serving of french fries at a certain fast food
restaurant has an average of 475 calories. The restaurant is
experimenting with the new type of oil for cooking french
fries. It is claimed that servings of fries cooked in the new
oil have less than 475 calories, on average. The following
are the calorie measurements from a sample of serving
cooked in the new oil.

465 459 425 433 466 437 532 411 454 470

436 479 484 496 419 472 512 459 470 399

445 418 379 427 477 423 500 430 457 468

1. Test the claim that the mean number of calories for a all large servings to be cooked in the
new oil is less than 475. Let = 0.05
2. Estimate the mean number of calories for all large servings to be cooked in the new oil.
Construct a 95% confidence interval for this mean.
3. Construct a box-and-whisker plot.
 Small sample condition

When we are dealing with a random sample

from normal population with n < 30.
The test statistic is

x
t
s
n
 Example 3

A new born baby should has length 50 cm. In a

certain maternity unit during a particular
week, 13 babies were born with the following
length.
49, 50, 45, 51, 47, 49, 48, 54, 53, 55, 45, 50, 48
Test whether the population mean is correct
at 5% significant level.
 Statistical power

The power of a statistical test is the

probability of rejecting H0 when H0 is really
false.
Thus power = 1-.
Clearly if the test maximizes power, it
minimizes the probability of Type 2 error, .