Research Methodology II
Module 2
Overview of Hypothesis Testing
•A method used to determine if there is enough evidence to reject a null hypothesis (H₀)
•Compares observed data against expected outcomes under the null hypothesis
Hypothesis
• (H₀) – Null Hypothesis (Dull Hypothesis)/ Hypothesis of Equality or no
change or no relationship
• (H₁) – Alternate Hypothesis – Difference, relationship
Key Terms
• Margin of Uncertainty/ Level of Significance - P-value (Probability of
Type 1 error): Probability of observing data as extreme as the sample, under
H₀
• Significance Level (α): Commonly set at 0.05 as per different fields (0.01 in
medical sciences and 0.10 in agricultural sciences)
• If p value<0.5 reject Null Hypothesis or else don’t reject it.
Blackstone ratio
• The Blackstone Ratio (by William Blackstone ) is a principle in law that
emphasizes the importance of protecting the innocent, even at the risk of
allowing some guilty individuals to go free.
• "It is better that ten guilty persons escape than that one innocent
suffer."
• The Blackstone Ratio is based on the belief that it is more morally acceptable
for guilty people to avoid punishment than for an innocent person to be
wrongfully convicted.
Remember it
• If P Value is Low, Null will go
• If P value is high, Null will fly
Comparison with existing value/ benchmark
•Scenario: Testing if the average exam score of a class (sample) differs from the
national average of 70
•Null Hypothesis (H₀): The average score is 70 (μ = 70)
•Alternative Hypothesis (H₁): The average score is different from 70 (μ ≠ 70)
Hypothesis Testing for a Single Population
• Example Scenario -
• Test whether the average score of a group is significantly different from a
known value (e.g., test if the average IQ of a group differs from 100)
SPSS Test
•Hypotheses:
•H₀: μ = μ₀ (no difference from the known population mean)
•H₁: μ ≠ μ₀ (the mean is different)
•SPSS Test Used: One-Sample T-Test
Steps for One-Sample T-Test in SPSS
• Step 1: Go to "Analyze" > "Compare Means" > "One-Sample T-Test"
• Step 2: Select the scale variable to test
• Step 3: Enter the test value (e.g. –> 50000 in Employeedata.sav)
• Step 4: Run the test and interpret the p-value
• Interpretation:
• If p-value < 0.05, reject H₀; otherwise, fail to reject H₀
Employee Data
• For employeedata.sav
• Steps:
• Go to "Analyze" > "Compare Means" > “Independent-Sample T-Test“
(Test Variable is Current Salary and Grouping variable is Gender)
• We need to Define Groups – Group 1 (Female) and Group 2 (Male)
• Continue -> OK
Hypothesis Testing for Two Populations
• What is a Two-Population Test?
• Compares means or proportions between two independent groups or samples
• Common Applications:
• Comparing average scores between male and female students
• Testing differences in income between two regions
• Types of Tests:
• Independent-Samples T-Test (for comparing means)
Steps for Independent-Samples T-Test in SPSS
• Step 1: Go to "Analyze" > "Compare Means" > "Independent-Samples T-Test"
• Step 2: Select the test variable (e.g., test scores) and grouping variable (e.g., gender)
• Step 3: Define groups (e.g., Male = 1, Female = 2)
• Step 4: Run the test and interpret the output
• Interpretation:
• Look at the Levene’s test for equality of variances
• If p-value < 0.05, reject H₀; otherwise, fail to reject H₀
Hypothesis Testing through Chi-Square
• Only when two categorical variables (C for Categorical and C for Chi-Square)
• Eg – Gender and genre of movie choice
Step-by-Step Chi-Square Calculation
• State the Hypotheses:
• H₀ (Null Hypothesis): There is no association between gender and preferred movie
genre (gender and preference are independent).
• H₁ (Alternative Hypothesis): There is an association between gender and preferred
movie genre.
• Step 2: Set Up the Cross Tabulation
Click on "Analyze" in the top menu.
• Select "Descriptive Statistics" and then "Crosstabs" from the drop-down menu.
• In the dialog box:
• Move Gender to the Rows box.
• Move Movie Genre to the Columns box.
• Click on the "Statistics" button in the Crosstabs dialog box.
• Step 3: Perform the Chi-Square Test
• In the Statistics dialog box: Check the box next to "Chi-Square" to select the Chi-Square test.
Then Continue to return to the Crosstabs dialog box.
• Optionally, click on the "Cells" button to customize how the data is displayed:
• Check the "Observed" and "Expected" boxes to show observed and expected frequencies in the output.
• Click Continue to return to the main Crosstabs dialog box.
• Step 4: Run the Test
• Click OK to run the Chi-Square test.
Paired Sample T -Test
• Before and After Comparison
• Eg – Change in blood sugar level of patient before and after food.
• File- Open- Data- C- Program- IBM- SPSS Statistics- Samples – English –
dietstudy.sav (after selecting all files).
• Select weight and final weight - OK
Correlation
• Analyze – Correlate – Bivariate – Select Sales and Fuel Efficiency
(Carsales.sav)
• +1 to -1 it shows positive to negative correlation
Useful Resource
• https://statisticsbyjim.com/
