IA1
What is a
Hypothesis?
Teaching presentation by
Parnika Reys Gamat
Index
Definition Types Quiz
Characteristics Testing Month, 20YY
Contoso Ltd.
2
Add a footer
Definition
A hypothesis is a formal, testable statement
that proposes an expected relationship
between two or more variables.
Creswell, 1994
Dependent
○ Educated guess Variable
○ Based on limited evidence
Smoking leads to lung cancer
○ Starting point for investigation
Independent
Variable
Contoso Ltd.
3
Add a footer
Characteristics of a Good Hypothesis
Testable Clear & Precise Logically Sound Declarative Limited in Scope
It must be possible to It should clearly state the Statement
Consistent with existing It focuses on a
conduct an experiment or variables involved and the facts and scientific It is an affirmative statement specific phenomenon
gather data to determine expected relationship theories. or relationship rather
that can be confirmed or
if the hypothesis is true or between them. than being overly
refuted through research.
false. broad.
Contoso Ltd.
1
Types of Hypotheses
Null Hypothesis (H₀) Alternative Hypothesis (H₁ or Hₐ)
Sleeping
less does Sleeping
not have an A statement where there is no less has an A statement where there is a
effect on significant relationship or effect on
significant relationship or difference
depression. difference between variables depression.
between variables
If plants Simple Hypothesis If plants
Complex Hypothesis
receive sun receive sun
light, they light and Proposes a relationship between two or
will grow Proposes a relationship water, they
taller. between only two variables will grow
more independent variables and two or
taller. more dependent variables
Non-directional Hypothesis
Students Directional Hypothesis
Study time Suggests there is a relationship who study
affects test longer will
but does not specify the achieve Specifies the direction of the
scores.
direction higher test expected relationship
scores. Contoso Ltd.
5
Add a footer
Hypothesis Testing
Statistical method to determine the probability that a claim or assumption about a population is true
Step 1 Step 2 Step 3
Create Hypothesis Choose significance level Collect & Analyse Data
• Null hypothesis: Average • α (alpha) = chance of • Choose a representative
height of Bangalore being wrong when
sample
residents is equal to 5ft rejecting H₀
• Gather data using
• H₀: μ = 5ft • Usually 5% (0.05)
consistent methods
• Alternate hypothesis: • Means: “We’re 95%
• Organise data using
Average height of confident our result isn’t
descriptive statistics
Bangalore residents is by chance.”
not equal to 5ft • Calculate test statistic
and p-value
• Hₐ: μ ≠ 5ft
Contoso Ltd.
6
Add a footer
Hypothesis Testing
Step 4 Step 5 Step 6
Perform Statistical Test Make a Decision Draw a Conclusion
• Choose the right test: z- • If the p-value is less than
• Explain what this
test, t-test or chi-square the significance level
statistical decision
(α), reject the null
• Calculate the test means for your specific
hypothesis in favour of
statistic: Use formulas or problem or hypothesis
the alternative
software (e.g., “There is enough
hypothesis.
evidence to suggest
• Find the p-value: The p-
• Otherwise, fail to reject that…”).
value helps us decide
the null hypothesis. • Translate the statistical
whether to reject or
keep the null hypothesis. finding into real-world
meaning.
Contoso Ltd.
7
Add a footer
Example of Hypothesis Testing
• Null hypothesis: Average height of
Bangalore residents is equal to 5ft
H₀: μ = 5ft
Alternate hypothesis: Average height of
Bangalore residents is not equal to 5ft
H₁: μ ≠ 5ft
Both tails (left and right) are shaded.
• Left tail (H₁: μ < 5ft) → people are shorter
than expected.
• Right tail (H₁: μ > 5ft) → people are taller
than expected.
The red line at 5.2 ft represents the sample
mean, which falls in the right tail, indicating
evidence to reject H₀.
Contoso Ltd.
8
Add a footer
Quiz
Scan the QR code to
participate in a short quiz
Contoso Ltd.
9
Add a footer
Thank you