ANSWER 1.
a. Independent Sample t-test vs. Paired Sample t-test
• Independent Sample t-test compares the means between two distinct
groups as well as when there is 1 IV and 1 DV where IV must be
categorical and DV must be continuous.
• Example: A brand wishes to determine if a new ad results in better
product recall than an old one. They create a survey with two groups, one
exposed to the new ad and one to the old ad. An independent t-test would
assist in determining whether the two ads have significantly different
recall scores.
Paired Sample t-test is used to compare means of the same group at two
times or under two conditions.
Example: A firm identifies brand awareness before and after campaign
across the same consumers.
b. Validity vs. Reliability
Validity:
• Talks about whether actually what tool measures is what it should
measure, ,measures accuracy.
• Talks about a questionnaire that should have questions which directly
measure loyalty behaviour, not satisfaction or awareness.
Reliability:
• Refers to the steadiness of measurement over time or items, Measures
consistency Are the findings consistent over time?
• Example: Consumers completing a brand image scale similarly today
and a week from now indicate the scale is reliable.
C. Parametric Tests:
• Need normally distributed data and interval/ratio scales.
• Powerful and more exact.
• Example in Marketing: Apply ANOVA to contrast mean purchase
intention scores from three types of ads.
Non-parametric Tests:
• None assumes normality; applied to ordinal/nominal data.
• Less significant but appropriate for skewed data.
Example: Applying Chi-square to see whether gender influences product
preference.
ANSWER 2.
a. Ad Type and Region vs. Brand Perception
• Test: Two-Way ANOVA
• Reason: Two discrete independent variables (ad type: video/print/social;
region: urban/rural), one metric dependent variable (brand perception).
• Purpose: To test main and interaction effects.
b. Preference for Product Feature vs. Expected Distribution
• Test: Chi-Square Goodness-of-Fit Test
• Reason: Cross-tabulating observed frequency (eco-friendly vs. price)
against expected distribution (40%-60%).
• Purpose: To test for conformance to anticipated customer taste.
c. Male vs. Female Median Monthly Spending (Normal Data)
• Test: Independent Sample t-test
• Rationale: Two independent samples (male and female), normally
distributed interval data (monthly spending), comparing means.
Purpose: To determine whether male and female spending patterns vary.
ANSWER 3.
a. Employee Training Program Type and Store Location vs. Customer
Satisfaction
Research Question: Is employee training program type and store location
associated with customer satisfaction?
Hypotheses:
• H0: No significant effect of training or location.
• H1: At least one factor significantly affects satisfaction.
Variables:
• IVs: Training program (A/B), Store location (Urban/Suburban)
• DV: Customer satisfaction (scale 1–10)
Respondents: Customers at various store locations.
Procedure:
• Randomly assign stores to either Program A or B.
• After training, measure customer satisfaction through surveys.
Statistical Analysis: Two-Way ANOVA to evaluate individual and interaction
effects.
b. Packaging Design Effects on Brand Perception and Purchase Intention
Research Question: Does packaging colour, shape, and age group
determine brand perception and purchase intention?
Hypotheses:
• H0: Packaging factors and age have no impact.
• H1: At least one factor influences the results.
Variables:
• IVs: Colour (Bright/Neutral), Shape (Round/Square), Age Group (18-
30/31-45)
• DVs: Brand perception, Purchase intention
Respondents: Target age group participants.
Procedure:
• Present respondents with various packaging combinations.
• Obtain responses via a structured survey.
Statistical Analysis: Multivariate ANOVA (MANOVA) for two DVs and more
than one IV.
c. Loyalty Program vs. Product Preference
• Research Question: Is enrolment in loyalty program linked with preferred
product category?
• Hypotheses:
o H₀: No relationship between loyalty enrolment and product preference.
o H₁: There is a relationship.
• IV: Loyalty program status (Enrolled/Not enrolled)
• DV: Preferred product category (Groceries, Electronics, Clothing)
• Respondents: Supermarket customers
• Procedure: Survey customers for loyalty status and product preferences.
• Statistical Analysis: Chi-square Test of Independence
ANSWER4.
a. Online vs. In-Person Class Exam Scores
i) Hypotheses:
• H0: There is no difference in exam scores between online and in-person
students.
H1: There is a significant difference.
ii) Interpretation:
•t = -2.35, p = 0.021 < 0.05
•Reject H0: Significant difference exists.
iii) Implications:
•Mode of instruction is important in marketing research training.
Attendance to physical classes could result in improved knowledge
assimilation.
b. Productivity Before and After Training
i) Hypotheses:
•H0: No difference in productivity after training.
H1: Productivity increased after training.
ii) Interpretation:
•t = -3.45, p = 0.001 < 0.05
•Reject H0: Significant difference after training.
iii) Implications:
Marketing departments that spend money on training can anticipate
measurable improvements in performance. The training worked.
c. Gender vs Smartphone Brand Preference
• i) Hypotheses:
o H₀: No relationship between gender and brand preference.
o H₁: There is a relationship.
• ii) Interpretation:
• Chi-square = 8.91, p = 0.012 < 0.05
o There is a significant relationship between gender and brand preference.
• iii) Implication: Marketing strategy should segment by gender when
marketing smartphone brands, as brand preference differs based on
gender.
