INFERENTIAL STATISTICS
Concerned with the formulation of conclusions or
generalizations about a population based on a sample
drawn from the population.
TWO AREAS:
Estimation – any process by which a sample is used to
estimate the numerical value of the unknown parameter.
Hypothesis testing – is concerned with determining the
validity of assertions about population parameters
TESTS OF HYPOTHESIS
BASIC CONCEPTS IN
HYPOTHESIS TESTING
Statistical Hypothesis – an assertion, statements or
conjecture concerning one or more unknown
parameters or distribution properties of the
population.
Hypothesis Testing – a statistical procedure in
determining which hypothesis is more acceptable as
true or which hypothesis is more likely to be false
TWO TYPES OF STATISTICAL
HYPOTHESIS
Null Hypothesis – denoted by Ho
statement to be tested
a hypothesis of equality or no difference
Alternative Hypothesis – denoted by Ha or H1
a hypothesis believed to be true when the
null hypothesis rejected
sometimes called the research
hypothesis, i.e. the hypothesis that the
researcher believes in.
EXAMPLES:
1. Manager of a certain bank claims that their non-ATM customers need
to wait, on the average, for at most ten minutes before they are
served.
Ho : Non-ATM customers need to wait an average of at most ten
minutes before they are being served.
In symbol:
Ha : Non-ATM customers need to wait an average of more than
ten minutes before they are being served
In symbol:
EXAMPLES:
2. There is no difference between the average life of brand
A light bulbs and that of brand B light bulbs.
Ho : Average life of brand A and B light bulbs are the
same.
In symbol: µ A = µB
Ha : Average life of brand A and B light bulbs are
different.
In symbol: µA µB
TYPES OF ALTERNATIVE
HYPOTHESIS
Non – directional Hypothesis – an assertion that there is
significant difference between two descriptive measures
Makes use of “not equal to” symbol
Know as two-sided hypothesis
Directional hypothesis – an assertion that one
descriptive measure is less than or greater than another
descriptive measure.
Makes use of “less than (<)” or “greater than (>)” symbols
Sometimes called one sided hypothesis
GUIDELINES:
1. Read and understand the problem carefully. After which, decide on what you
want to test.
2. If simple direction such as more than, less than, superior to, inferior to, and so
on is suggested, then alternative hypothesis will be stated using the
inequality symbols.
3. If compound direction such as at least, equal to or greater than, at most, no
more than, and so on is suggested, then this entire compound direction (< or
>) is expressed as null hypothesis, but using only the equality sign, and
alternative hypothesis is given the opposite direction.
4. If no direction is suggested, then alternative hypothesis is stated using the
not equal symbol.
LEVEL OF SIGNIFICANCE
The probability that the null hypothesis will be rejected in
error when it is true.
Denoted by
Commonly used are 0.1, 0.05 and 0.01
“significant” means probably true (not due to chance); a
research finding may be true without being important
“highly significant” means it is very probably true which does
not necessarily mean it is highly important
FOUR POSSIBILITIES: (WHENEVER HO IS
“REJECTED” OR “FAIL TO REJECT” ON THE BASIS OF A
SAMPLE)
Decisions/Fact Ho is TRUE Ho is FALSE
Fail to reject Ho Correct decision Type II error
Reject Ho Type I error Correct decision
TESTS OF SIGNIFICANCE
One-tailed Test – when states a direction
Examples:
1. The mean income of men is greater than the mean
income of women. In symbol:
2.The mean income of men is less than the mean
income of women. In symbol:
TESTS OF SIGNIFICANCE
Two-tailed Test – when does not state a direction
Examples:
The mean income of women is not equal to the mean
income of men.
CRITICAL AND ACCEPTANCE REGIONS
Critical Region – is the region in which the null
hypothesis is rejected if the test statistic falls
within such region.
Critical value – serves as the basis whether the
hypothesis is rejected or failed to reject
the location of the critical region is determined by
the nature of the alternative hypothesis.
WHEN TO REJECT
If computed test statistic
absolute value is greater
than the test statistic
tabulated value
CRITICAL REGIONS FOR Z-TEST
STEPS IN HYPOTHESIS TESTING
1. State the null and alternative hypotheses.
2. Choose a level of significance of size α
3. Select the appropriate test statistics.
4. Establish the critical region (Decision Rule)
5. Compute the value of the test statistic from the sample data.
6. Provide statistical decision.
7. Make an interpretation and if, if possible, draw a conclusion.
TEST CONCERNING MEANS AND
PROPORTIONS
Test of Single Mean for Large Samples
Test of Single Mean for Small Samples
Test of Means for Two Independent Samples
Test of Two Means for Dependent Samples
Test of Single Proportion
Test of Two Proportions
TEST OF SINGLE MEAN FOR LARGE
SAMPLES
The z –.
When the population standard deviation is known, regardless of the
sample sizes
When the population standard deviation is not known but the
sample size is large ()
=
SAMPLE PROBLEM 1:
A pediatrician claimed that the average time for a drug to
take effect is 15 minutes with a standard deviation of eight
minutes. In a sample of 49 trials, the average time was 18 minutes.
Test the claim of a pediatrician that the average time is not equal
to 15 minutes. Assume that the average time for a drug to take
effect is approximately normally distributed.
SOLUTION:
: The average time a drug to take effect is 15 minutes.
: The average time a drug to take effect is not 15 minutes.
2. Level of significance: α =0.05
3. Test statistic: z – test for one population mean
4. Decision Rule: Reject if ;
SOLUTION:
5. Computation:
6. Decision: Reject since the computed value, = 2.63 falls within the critical
region.
7. Interpretation/Conclusion: We conclude that the average time a drug to
take an effect is not 15 minutes. Thus the claim of pediatrician is true.
SAMPLE PROBLEM 2:
Director of MSU-TCTO /science High School noted from
the last year results of the MSU-TCTO Science High School
Scholarship Examination that students of MSU-TCTO
Laboratory Elementary School has a mean score of 95 with
standard deviation of 14. For this academic year, there were 38
examinees that obtained a mean score of 93. Can the director
claim that this year’s performance is inferior to the previous
year? Assume that the mean score is approximately normally
distributed.
SOLUTION:
: Pupils of MSU-TCTO Laboratory Elementary School have mean
score of 95
: Pupils of MSU-TCTO Laboratory Elementary School have mean
score lower than 95
2. Level of significance: α =0.05
3. Test statistic: z – test for one population mean
4. Decision Rule: Reject if
SOLUTION:
5. Computation:
6. Decision: Do not reject the null hypothesis because the computed z-
statistic falls within the acceptance region
7. Interpretation/Conclusion: We have no enough reason to believe
that the performance of the examinees for this year is inferior to the
previous year.
TEST OF SINGLE MEAN FOR SMALL SAMPLES
The t-statistic or t-test is appropriate when the population
standard deviation is unknown and the sample size is
small .
SAMPLE PROBLEM 3:
Prof. Respicto administered an achievement
test, which as he remembers from years of
experience, yielded a population mean score of 85.
If his class of 25 students obtains a mean score of
89 with standard deviation score of 14, can the
professor claim that his class is superior class?
Assume that the mean score is approximately
normally distributed.
SOLUTION:
: Mean score of the students in the achievement test is 85.
: Mean score of the students in the achievement test is
greater than 85.
2. Level of significance: α =0.05 ; df = n – 1 = 24
3. Test statistic: t – test for one population mean
4. Decision Rule: Reject if t
SOLUTION:
5. Computation:
6. Decision: Do not reject the null hypothesis because the computed z-
statistic falls within the acceptance region
7. Interpretation/Conclusion: We have no sufficient reason to say that
the professor is justified in assuming that his class is a superior class.
ACTIVITY 1:
The average length of time for students to register
in the second semester at a certain university has been 55
minutes. A new registration procedure is being tested. If
random samples of 16 students have an average of 50
minutes with standard deviation of 12 minutes under the
system, can you conclude that the new system is faster
than the old one? Assume that the average length of
time is normally distributed. Use .
TEST OF MEANS FOR TWO INDEPENDENT SAMPLES
Comparing the difference of two groups
Independent Samples implies that the selection of samples
in one group does not in any way affect the selection of
samples in the other group
Z-test is used when the population standard deviation of
both groups is known or unknown but the size of the
samples from group is large.
= sample mean from population 1
= sample mean from population 2
= population standard deviation from
population 1
=population standard deviation from
population 2
= total no. of samples in population 1
= total no. of samples in population 2
SAMPLE PROBLEM 4:
During the SJC Foundation Day, the Director of the
Physical Education recorded the mean height of 37 male
students who showed average participation was 70.3
inches with inches, while 32 male students who showed no
interest in participation had a mean height of 69.1 inches
with inches. Is the director justified that male students
who showed average participation during the foundation
day are taller than the other male students who showed no
interest in participation. Assume that the height is
normally distributed.
SOLUTION:
or = 0 : There is no significant difference on the mean height between the
male students who showed average participation and
that of the male students who showed no interest in
participation
: Male students who showed average participation are taller than the
male students who showed no interest in participation
2. Level of significance: α =0.05
3. Test statistic: z – test for two independent samples
4. Decision Rule: Reject if
SOLUTION:
5. Computation:
6. Decision: Reject since the computed z is within the critical/rejection region.
7. Interpretation/Conclusion: The mean height of male students who showed
average participation is taller than those who showed no interest in participation.
The director was justified in his assumption.
ACTIVITY NO. 2:
Dr. Tejano wishes to determine which of the two methods of teaching is better in
teaching Physics subject. In a class of 38 students, he used lecture with laboratory and in
another class of 39 students, he used lecture without laboratory. The doctor administered
the same set of test and obtained the following:
Lecture Method Means Standard Deviation
With laboratory 85.9 5
Without laboratory 83 6
Is the doctor correct in assuming that there is significant difference between the two
methods of teaching Physics. Assume that the method of teaching is approximately
normally distributed
T-TEST FOR INDEPENDENT SAMPLES
df = 15 + 12 – 2 = 25
Reject Ho if
df = n – 1 = 10 – 1 = 9
t-test for two dependent samples
𝟏
𝒅= =𝟎 . 𝟏
𝟏𝟎
TEST OF SINGLE PROPORTION
Where:
Z-test for one population
proportion
𝒒 𝟎=𝟏 − 𝟎 .𝟓=𝟎 . 𝟓
SAMPLE PROBLEM FOR TWO PROPORTIONS
SOLUTION
SOLUTION:
3) Test Statistic: z –test for Two Proportions
4) Decision Rule: Reject Ho if
5) Computation:
SOLUTION: