II.

INFERENTIAL STATISTICS To construct a 100(1-α ¿% confidence interval for the difference between two population means
A. Estimation
i. Point Estimation (μ1−μ 2), a sample is drawn from each population and the following notation are used.
ii. Interval Estimation Statistics Population 1 Population 2
a. Confidence Interval for One Population Mean Sample Size n1 n2
b. Confidence Interval for Difference between Two Population Means
1. Independent-Sample Sample Mean μ1 μ2
2. Paired-Observations Sample Variance 2 2
B. Hypothesis Testing s1 s2
i. Test for One Population Mean Sample Standard Deviation s1 s2
ii. Test for Difference between Two Population Means
Population Variance 2 2
1. Independent-Sample σ1 σ2
2. Paired-Observations
iii. Test for Several Population Means
Population Standard Deviation σ1 σ2
CONFIDENCE INTERVAL FOR ONE POPULATION MEAN Using the above notation, the following computing formulas are used to construct a confidence
Case 1: When is known: interval for the difference of two population means.

CONFIDENCE INTERVAL FOR DIFFERENCE OF TWO POPULATION MEANS

(INDEPENDENT-SAMPLES)

A frozen food company wishes to know the mean length of corn received in a large Case 1: When and are known:
shipment. A random sample of 20 corns were collected and measured and found to
have a mean length of 8.8 inches. It has been known that the standard deviation of
all corn shipment is 1.5 inches. Find a 95% confidence interval for the mean length
of corn in the large shipment.

Case 2: When is unknown and : Case 2: When and are unknown and and :

A random sample of 40 insomniac patients from a population of 100 was asked to

enumerate their sleeping hours. The mean length of these hours is 4 hours with a The random sample of 45 electric fans of Brand A showed a mean life span of 4.11
standard deviation of 0.25 hours. Compute a 99% confidence interval for the mean years with a standard deviation of 0.55 years, while a random sample of 50 electric
length of sleeping hours of all the patients. fans of Brand B have a mean life span of 3.51 years with a standard deviation of
0.23 years. Construct a 95% confidence interval for the true difference between
Case 3: When is unknown and : means.

Case 3: When = but unknown and and :

with v = n-1 degrees of freedom

An instructor wishes to know the mean score of psychology students in Psy2 exam.
A random sample of 12 students were recorded to have the following scores:

√
24 23 22 17 21 18 1 2
19 23 24 19 22 18 ( n1 −1) s 2 +( n 2−1 ) s 2

Assuming the distribution of exam scores to be normal, construct a 95% confidence sp=
interval for the mean score of psychology students in Psy2 exam. where v with v = n1+n2-2 degrees of freedom
Two groups, A and B, consisting of 10 and 15 individuals respectively, were tested
to two types of sleeping pills. The first group (A) was given a new type of sleeping
pill and second group (B) was given a conventional type. For the patients in group A
the mean number of hours of sleep was 7.82 with a standard deviation of 0.24
hours. For patients in group B, the mean number of hours of sleep was 6.75 with a
standard deviation of 0.30 hours. Find the 95% confidence interval for the
difference in the mean number of hours of sleep induced by the two types of
sleeping pills. Interpret the result.
Assignment:
Case 4: When ¿ but unknown and and :

[ ]
2 2 2 1. The weights in kilograms of five (5) women who took a new pill diet were recorded before and after
s 1 s 2 taking the pill for a 2-week period. Construct a 99% confidence interval for the mean difference in the
+ weights, assuming the distribution of weight to be approximately normal. The data are recorded
n1 n2
v= below.

( ) ( )
2 2 2 2
S 1 S 2 Women 1 2 3 4 5 6

√
2 2
S 1 s 2 n1 n1 Weight Before 58.5 60.3 61.7 69 64 58.5
( x̄ 1− x̄ 2 )±t α / 2 , v + + Weight After 60.0 54.9 58.1 62.1 58.5 60.0
n1 n2 where
n1−1 n 2−1
rounded off to the next higher integer 2. A random sample of 18 girls showed a mean height of 163 cm. with a standard deviation of 20
cm., while a random sample of 25 boys showed a mean height of 170 cm. with a standard deviation
of 2.98 cm. Assuming the heights are approximately normally distributed with unknown population
variances, can we say that the heights are not significantly different at a 0.05 level of significance?
CONFIDENCE INTERVAL FOR DIFFERENCE OF TWO POPULATION MEANS
(PAIRED OBSERVATIONS)

where
d̄=
∑ di
n and
sd =
√
n ( n−1 )
n( ∑ d 2 )−( ∑ d i )2
i

v = n1 - 1
The grade-point averages (GPA) of a random sample of 6 students who joined a cultural
group in the SY 2012-2013 were recorded:

Student 1 2 3 4 5 6
GPA (2nd Sem) 1.8 2.4 2.5 2.0 1.7 2.0
GPA (1st Sem) 2.0 1.9 3.0 2.5 2.4 2.0
Construct and interpret a 90% confidence interval for the mean difference in the GPA
assuming the distribution of the GPAs to be approximately normally distributed. Is there
an evidence of decrease in GPA?
 with v =
n-1
The average length of time for students of LSU to process their clearance is 50 minutes. A new
computerized clearance system is being tried. If a random sample of 12 students had an average
length of time of 42 minutes with a standard deviation of 11.9 minutes under the new system, test
the hypothesis that the average length of time for the students to process their clearance is less than
50 minutes. Use a level of significance of 0.05 and assume the population of times to be normally
HYPOTHESIS TESTING distributed.
Definitions:
1. Statistical Hypothesis – a guess or conjecture about the numerical value of some unknown TEST ABOUT THE DIFFERENCE OF TWO POPULATION MEANS
population parameters. Also called null hypothesis and it is denoted by H0. (INDEPENDENT-SAMPLES)
2. Alternative hypothesis – an alternative range of values of parameters assumed when H0 Test Statistics H1 Critical Region
evidence from the sample dictates that the value, or range of values, as stated in the null (i)
hypothesis is not feasible
3. Test statistics – a rule on deciding how to reject a hypothesis
4. Critical region / rejection region – a part of the set of all possible values of a test statistic for
which H0 is rejected.
5. Type I Error – the error of rejecting a true hypothesis ,
6. Type II Error – the error of accepting a false hypothesis
Steps in Hypothesis Testing: if are known
1. Formulate the null and the appropriate alternative hypothesis. (ii)
2. Specify the level of significance.
3. Select the appropriate test statistic.
4. Establish the critical region.
5. Compute the actual value of the test statistic from the sample.
6. Make a statistical decision. , if
Reject H0 if the computed test statistic is within the critical region.
7. Draw a conclusion.
TEST ABOUT A POPULATION MEAN
H0 Test Statistics H1 Critical Region are unknown and and

(i) = <
A random sample of 45 electric fans of Brand A showed a mean life span of 4.11 years with a
, if is known standard deviation of 0.55 years, while a random sample of 50 electric fans f brand B showed a
>
mean life span of 3.51 years with a standard deviation of 0.23 years. Would it be safe to say
that Brand A is significantly better than Brand B at a 0.05 level of significance?
A frozen food company wishes to know the mean length of corn received in a large shipment. A
random sample of 20 corns were collected and measured and found to have a mean length of 8.8 (iii)
inches. It has been known that the standard deviation of all corn is 1.5 inches. Is there a reason to
believe that the mean length of the corn is not 9.0 inches, at ?

(ii) = < , if
, unknown and
>

are unknown and and

The average weight of 100 randomly selected sacks of rice is 48.54 kilos with a standard deviation of
20 kilos. Test the hypothesis at 0.01 level of significance that the true mean weight is less than 50
kilos. Where

(iii) = <
, unknown and
>
 and v=n-1

A random sample of five (5) male students was selected from a university. Each student was
asked to determine both his left-hand and right-hand grip on an instrument designed to measure
grip in pounds. The data are given below:
In a certain factory, two machines are used to produce metal rods. A random sample of 11 rods
from machine 1 showed a mean length of 5.95 inches with a variance of 0.18 square inches Student 1 2 3 4 5
while random sample of 15 rods from machine 2 showed a mean length of 6.01 inches with a Left Hand (X1) 42 41 29 45 29
variance of 0.2 square inches. Test whether there is a significant difference between the mean Right Hand (X2) 45 42 30 46 37
length of the rods at = 0.05?
(iv) Test the hypothesis that the difference in the mean grip between left and right-hand is greater
( x̄ 1− x̄ 2 )−d 0 than 2 pounds. Use 0.05 level of significance.
t=

√
PROBLEM SET 3
s 2 s 2
1 2
+ Name: _______________________________________________________ Date: ____________
n1 n2 , if
σ1
1. Body temperature was obtained from a sample of 36 identical crabs exposed to air at 26.2 0C. Test
≠
σ2 the hypothesis that the mean body temperature of the species of crab under this condition exceeds
26.20C if the sample had a mean temperature of 26.5 0C with a standard deviation of 0.70C.
are unknown and and
2. The average no. of crimes in Metro Manila for 16 randomly selected monthly records is 165
with a standard deviation of 32. Test the hypothesis at 0.05 level of significance that the true
with mean no. of crime is higher than 150.

3. There is a claim that 90% of Filipinos are not afraid to go home at night. A random sample of 500

[ ]
2
s 2 s 2 showed that 400 are not afraid to go home at night. Would this random sample sustain the claim?
1 2
+ Use a 95% confidence interval to test this claim.
n1 n2
v= 4. A tetrapack company wishes to know the mean content of 50 defectively packed orange juices. A

[ ] [ ]
2 2
s s random sample of 7 packs were recorded to have the following contents, in milliliters:
12 22
240 190 230 220 170 210 180
n1 n2 Assuming the distribution of contents to be normal, construct a 95% confidence interval for the
+ mean content of all orange juices.
n1 −1 n 2−1
rounded off to the next higher 5. A sample of 150 brand X light bulbs showed a mean lifetime of 1400 hours and a standard
integer deviation of 120 hours. A sample of 200 brand Y light bulbs showed a mean lifetime of 1200 hours
and a standard deviation of 80 hours. Test at 0.05 level of significance that there is no difference
of the mean lifetimes of brands X and Y.
TEST ABOUT THE DIFFERENCE OF TWO POPULATION MEANS
(PAIRED-OBSERVATIONS)
H0 Test Statistics H1 Critical Region 6. A research laboratory has recently developed a new paint. A sample of twenty-five gallons is tested
and the mean coverage is found to be 515 sq. ft./gal. Assuming the standard deviation to be 50 sq
μd < d 0
μd =d 0 d̄−d 0 ft/gal, find a 95% confidence interval for the running time of all films, assuming an approximate
t= normal distribution.
sd μd > d 0
√n
where μd ≠d 0

√
2
n ∑ d 2+ (∑ d i )
i
sd =
n