Introduction to testing of

hypothesis
Somia Bakhtiar Lone
PhD Statistics*
Lecturer Biostatistics
Inferential Statistics
• This branch of statistics deals with techniques of making conclusions
about the population.
• Inferential statistics builds upon descriptive statistics.
• Inferential statistics enables us to make confident decisions in the
face of uncertainty.
For Example: Antibiotics reduce the duration of viral throat infections
by 1-2 days. To make a confidence decision about this statement based
on sample data inferential statistics will help.
What is hypothesis testing?
1. Testing of hypothesis is defined as the formal procedures used by
researchers to accept or reject statistical hypotheses
2. Making an assumption, called hypothesis, about a population parameter.
3. Using the sample statistic to evaluate the hypothesis (how likely is it that
our hypothesized parameter is correct.)
4. To test the validity of our assumption we determine the difference
between the hypothesized parameter value and the sample value.
Key Point
 Rejection of Hypothesis: The rejection of a hypothesis is to declare it false.
 Acceptance of Hypothesis: Acceptance of hypothesis is to conclude that
there is not sufficient evidence to reject it. Acceptance does not necessarily
mean that the hypothesis is true

3
Type of statistical hypothesis
• A statistical hypothesis is an assumption about a population
parameter. This assumption may or may not be true.
• Stating a Hypothesis
Null hypothesis Alternative hypothesis
A statistical hypothesis that contains a A statement of strict inequality such as
statement of equality such as ≤, =, or ≥. >, ≠, or <.
Denoted by H0 read as “H naught.” Must be true if H0 is false.
Denoted by Ha or H1 read as “H sub-
a”or “H-alternative”

4
NULL HYPOTHESIS ALTERNATIVE HYPOTHESIS
Any hypothesis which is to be tested An alternative hypothesis is any other
for possible rejection under the hypothesis which we accept when the null
assumption that it is true. A null hypothesis is rejected .
hypothesis should always be precise. Example:
Example: In the clinical trial of a new drug, the
In a clinical trial of a new drug, the alternative hypothesis might be that the
null hypothesis might be that the new drug has a different effect, on average,
new drug is no better, on average, compared to that of the current drug. We
than the current drug. Null would write
hypothesis for the given statement H1: the two drugs have different effects, on
will be like average. or
H0: There is no difference between H1: the new drug is better than the current
the two drugs on an average. drug, on average.

5
Concept of one tail and two tail test
Three types of hypothesis tests
 One-tailed test: A one-sided test is a
statistical hypothesis test in rejection
region falls on a single tail of the
probability distribution.
 Left-tailed test
 Right-tailed test
 Two-tailed test: A two-sided test divides
the region of rejection into two portions
for each tail.
These regions are indicated by the
alternative hypothesis.
6
 Left-tailed Test
Left tailed test will reject the null hypothesis if the sample mean is significantly lower
than the hypothesized mean. The alternative hypothesis Ha contains the less-than
inequality symbol (<).
H0: μ ≥ μ0
Example: A researcher wants to test if a
Ha: μ < μ0 new drug reduces blood pressure more
than a standard treatment.
Critical Region is the Solutions:
area to the left of the
standardized test H0: μ ≥ μ0 (mean BP reduction is greater
statistic. than or equal to standard treatment)
z
H1: μ < μ0 (mean BP reduction is less
–3 –2 –1 0
Critical value
1 2 3
than standard treatment)
Critical values: The value(s) that separates the critical region
from the acceptance region, is called the critical value(s).
7
Right-tailed Test
Right tailed test will reject the null hypothesis if the sample mean is significantly
higher than the hypothesized mean. The alternative hypothesis Ha contains the
greater-than inequality symbol (>).
H0: μ ≤ μ0 Example: Suppose a clinical trial is
Ha: μ > μ0 assessing whether a new medication
increases the average heart rate more
than an established medication.
Critical Region is
the area to the right Solutions:
of the standardized
test statistic. H0: μ ≤ μ0 (mean HR increase is less
than or equal to the established
medication)
z
–3 –2 –1 0 1 2 3 H1: μ > μ0 (mean HR increase is greater
Critical than the established medication)
value 8
 Two-tailed Test
Two tailed test will reject the null hypothesis if the sample mean is significantly higher
or lower than the hypothesized mean. The alternative hypothesis Ha contains the not-
equal-to symbol (≠). Each tail has an area of ½ critical Region.
H0: μ = μ0 & Ha: μ ≠ μ0

Critical Region is twice the Critical region is twice the

z
–3 –2 –1 0 1 2 3
Critical value Critical value
Example: A study aims to determine if a new vaccine has an effect on the
incidence of a disease compared to a placebo.
H0: μ = μ₀ (mean incidence rate of disease is the same for both groups)
H1: μ ≠ μ₀ (mean incidence rate of disease is different between groups)9
Procedure for Hypothesis testing
Set up the Null Hypothesis and
Alternative hypothesis

Identify level of significance

Test Statistic

Decision on calculated p-value /

Conclusion

10
Writing Hypotheses
Write the claim about the population. State the null and alternative
hypothesis.
Example: A hospital claims its ambulance response time is less than 10
minutes.
H0:µ ≥10 min
Ha:µ < 10 min (Claim)
Example: Study claims that average weight of male is greater than female.
H0: μMale = μFemale
Ha: μMale > μFemale
H0: There is no difference between the weight of male and female
Ha: The weight of male is greater than female.

11
Level of Significance
The probability of rejecting true null hypothesis is called level of significance. The significant
level of a test is the probability used as a standard for rejecting a null hypothesis H0 when H0
is assumed to be true. It is denoted by α. The value of α is also known as the size of the
critical region. For α = 5% we mean that we are 95% confident in making the correct decision.

Though any level of significance can be adopted, in practice we either take 5% or 1% level of
significance. When we take 5% level of significance(α= .05), then there are about 5 chances
out of 100 that we would reject the null hypothesis when in fact it was true. In other words
out of 100, 95% chances are there that the null hypothesis will be accepted when it is true i.e.
we are about 95% confident that we have made the right decision.

12
Test Statistic
A sample statistic which provides a basis for testing a null hypothesis, is
called a test statistic. In other words, it is a numerical value computed
from sample data.

Determination of suitable test statistic: For example Z, student’s t, Chi-

13
p-value
• A p-value ,or probability value, is a statistical measurement used to
validate a hypothesis against observed data.
• A p value is the probability that the computed value of a test statistic is at
least as extreme as a specified value of the test statistic when the null
hypothesis is true.
• The p-value in hypothesis testing uses the calculated probability to
determine whether there is evidence to reject the null hypothesis.
• In simpler terms, it tells you whether your results are statistically significant
or not.
• The lower the p-value, the greater the statistical significance of the
observed difference.

14
15
Conclusion
• A low p-value suggests that the observed results are unlikely under
the null hypothesis and thus provides evidence to reject the null
hypothesis in favor of the alternative hypothesis.
• It means that if p-value of 0.05 or lower is generally considered
statistically significant.
• The value 0.05 is level of significance, α. Which is generally take 5% or
1%.
Decision Rule Conclusion
p-value ≤ α Reject H0
p-value > α Accept H0
16