In: Statistics and Probability
We wish to gather some information relating to the average time
a hospitalized Eagle Flu patient remains in the hospital. We
randomly sample 16 cured formerly hospitalized Eagle Flu patients
and these were the number of days spent in the hospital:
6.09, 8.04, 6.43, 8.58, 10.34, 5.40, 9.87, 9.24, 9.18, 5.06, 5.39,
5.89, 11.91, 8.94, 5.68, 9.85
Many surveys of flu hospital stays show that hospitalization length
is normally distributed. So, we assume that our sample comes from a
normal population with unknown mean of μ days and an
unknown standard deviation of σ days. We would like to
test whether the average hospital stay length is less than 8.1
days..
The null hypothesis is thus
H0:μ=8.1
. We will test this against the alternative
Ha
.
We want to test at the 8% level.
Let x = the sample mean and s = the sample standard
deviation.
a) What should the alternative hypothesis,
Ha
, be?
Ha:μ<8.1
Ha:μ=8%
Ha:μ=8.1
Ha:μ≠8.1
Ha:μ>8.1
b) What is the formula for your test statistic?
T =
x -8% |
s |
T =
x-8.1 | ||||
|
T =
x-8.1 | ||
|
T =
x-8.1 | ||
|
T =
x-8.1 | ||||
|
c) What value does your test statistic,T, take on with the sample
data?
d) What type of probability distribution does your test
statistic,T, have?
Cauchy binomial normal t Chi-Squared
e) How many degrees of freedom does T have?
(a)
Since ; So, it's a One Tailed ( Left Tailed ) Hypothesis Test.
(b) To test the we use the t - Statistic
NOTE; Since it's the LEFT TAILED HYPOTHESIS TEST; So, we shouldn't take MODULOUS in the Numrator of the formula.
Where = Sample Mean
Where = Sample Standared Deviation
n = Sample Size
(c)
x | |
6.09 | |
8.04 | |
6.43 | |
8.58 | |
10.34 | |
5.4 | |
9.87 | |
9.24 | |
9.18 | |
5.06 | |
5.39 | |
5.89 | |
11.91 | |
8.94 | |
5.68 | |
9.85 | |
MEAN | 7.8681 |
S.D | 2.0875 |
(e) Degrees of freedom (d.f)
We have n = 16
Degress of freedom = n-1
Therefore d.f = 16-1 = 15
Therefore d.f = 15.
Given = 8% = 0.08
we reject Ho
(d) we follow t distribution since n is a small sample,