Hypothesis Testing in Psychological Research
1. What is Hypothesis Testing?
Definition:
Hypothesis testing is a statistical method used by researchers to make decisions or inferences about
population parameters based on sample data. It involves testing an assumption (hypothesis) about a
population using probability theory and sample statistics.
APA Definition: “A method of drawing inferences or conclusions about a population from sample data by
assessing the probability that an observed effect occurred by chance.” (APA Dictionary of Psychology,
2024)
Purpose in Psychology:
In psychological research, hypothesis testing helps determine whether observed effects (like differences
between groups or relationships between variables) are statistically significant or could have happened by
random chance.
2. Key Terms in Hypothesis Testing
Term Definition Example
All college students in
Population The entire group about which we want to draw conclusions
India
Sample A subset of the population used in the study 200 psychology students
Average IQ of all Indian
Parameter A numerical characteristic of a population
students
Mean IQ of sample
Statistic A numerical value derived from a sample
students
Hypothesis A testable prediction about the relationship between variables CBT reduces anxiety
Significance Level The threshold for rejecting the null hypothesis (commonly 5% chance of Type I
(α) 0.05) error
The probability of obtaining results as extreme as the observed p = 0.02 means 2%
p-value
results under the null hypothesis chance
Test Statistic A value computed from sample data to test hypotheses t, z, F, χ² values
3. The Hypothesis Testing Process (Step-by-Step)
🔹 Step 1: Formulating Hypotheses
A. Null Hypothesis (H₀):
Represents a statement of no effect, no difference, or status quo.
Assumes that any observed difference in data is due to random chance.
Example:
H₀: There is no difference in depression levels between individuals receiving therapy and those who don’t.
B. Alternative Hypothesis (H₁ or Hₐ):
Represents a statement of effect, difference, or relationship.
This is what the researcher wants to prove or support.
Example:
H₁: Individuals receiving therapy will have significantly lower depression scores.
✅ Tip: Always state both hypotheses clearly, as they guide the direction of statistical testing.
🔹 Step 2: Set the Level of Significance (α)
Definition:
The significance level (alpha, α) is the probability of rejecting the null hypothesis when it is actually
true (i.e., Type I Error).
Common values:
o α = 0.05 (5%) → Most widely used in psychology
o α = 0.01 (1%) → For stricter evidence
o α = 0.10 (10%) → Sometimes used in exploratory studies
Why it's important:
It defines the critical region in which you would reject H₀ and accept H₁.
🔹 Step 3: Select the Appropriate Statistical Test
Depending on the type of data and research design, different statistical tests are used:
A. Types of Tests:
Test Use Case Example
t-test Comparing means of two groups CBT vs Medication on depression
ANOVA (F-test) Comparing means of 3 or more groups Comparing 3 therapy types
Chi-Square Test Frequency data, categorical variables Gender vs Voting behavior
Correlation (Pearson’s r) Relationship between variables Stress vs Sleep Quality
Regression Predicting values Predicting exam scores from study hours
B. Test Selection Depends On:
Scale of measurement (nominal, ordinal, interval, ratio)
Independent vs repeated measures design
Number of groups or conditions
🔹 Step 4: Collect and Analyze Data
A. Data Collection
Data must be valid, reliable, and unbiased.
Tools: Surveys, experiments, interviews, scales (like BDI, STAI)
B. Statistical Analysis
Use statistical software to compute the test statistic and p-value:
SPSS
JASP
R
Python (SciPy, Statsmodels)
🔹 Step 5: Compare p-value to α and Make a Decision
🔹 Decision Rule:
Condition Decision Interpretation
p≤α Reject H₀ Result is statistically significant
Fail to reject
p>α No significant effect found
H₀
Example:
You found p = 0.03 and α = 0.05
Since 0.03 < 0.05 → Reject H₀
➤ Therapy significantly reduces depression
4. Visual Aids
🔸 A. Flowchart of Hypothesis Testing Process
[Start]
↓
[State Hypotheses: H₀ and H₁]
↓
[Select Significance Level (α)]
↓
[Choose Appropriate Statistical Test]
↓
[Collect Data and Calculate p-value]
↓
[Compare p-value with α]
↓ ↓
p ≤ α p > α
Reject H₀ Fail to Reject H₀
↓ ↓
Statistical Significance No Significance
🔸 B. Normal Distribution Diagram for Hypothesis Testing
Critical Region Critical Region
⬅️α/2 (0.025) α/2 (0.025)➡️
| |
--------|-------------------|--------
-1.96 +1.96 (z-values)
H₀ Accepted in this region
For two-tailed tests with α = 0.05, you reject H₀ if z < -1.96 or z > 1.96.
❗ 5. Types of Errors in Hypothesis Testing
Error Explanation Consequence Example
Type I Error (α) Rejecting a true H₀ False positive Concluding CBT works when it doesn't
Failing to reject a false
Type II Error (β) False negative Missing a real effect of CBT
H₀
✔️Minimizing Errors: Use larger sample sizes and control extraneous variables.
6. Real-Life Psychological Example
Research Question:
Does mindfulness meditation reduce anxiety levels in college students?
Element Description
H₀ Mindfulness has no effect on anxiety
H₁ Mindfulness reduces anxiety
α 0.05
Test Independent Samples t-test
Sample 60 students (30 experimental, 30 control)
Result t(58) = 2.31, p = 0.024
Decision p < 0.05 → Reject H₀
Conclusion Mindfulness significantly reduces anxiety
7. APA Reporting Style
According to APA 7th Edition, you must report:
Test statistic (e.g., t, F, χ²)
Degrees of freedom (df)
p-value (e.g., p = .034, not p < .05)
Effect size (Cohen’s d, η²)
Confidence interval (e.g., 95% CI [2.4, 5.8])
🧾 Example:
t(58) = 2.31, p = .024, Cohen’s d = 0.6, 95% CI [2.1, 4.5]
8. References (APA Style)
1. Gravetter, F. J., & Wallnau, L. B. (2021). Statistics for the Behavioral Sciences (11th ed.). Cengage
Learning.
2. American Psychological Association. (2020). Publication Manual of the American Psychological
Association (7th ed.).
3. Coolican, H. (2018). Research Methods and Statistics in Psychology (7th ed.). Routledge.
4. Field, A. (2018). Discovering Statistics Using IBM SPSS Statistics (5th ed.). Sage Publications.
5.