Marketing Research
Ninth Edition
Chapter 12
Using Descriptive
Analysis, Performing
Population Estimates,
and Testing Hypotheses
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Learning Objectives (1 of 2)
In this chapter you will learn
12.1 What the different types of statistical analyses used in
marketing research are
12.2 What descriptive analysis is and how to do it
12.3 When to use a particular descriptive analysis measure
12.4 How to perform descriptive analyses with SPSS
12.5 How to report descriptive statistics to clients
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Learning Objectives (2 of 2)
In this chapter you will learn
12.6 The difference between sample statistics and population
parameters
12.7 How to estimate the population percentage or mean with a
confidence interval
12.8 How to obtain confidence intervals with SPSS
12.9 How to communicate confidence intervals insights to clients
12.10 What hypothesis tests are and how to perform them
12.11 How to communicate hypothesis tests insights to clients
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Where We Are
1. Establish the need for marketing research.
2. Define the problem.
3. Establish research objectives.
4. Determine research design.
5. Identify information types and sources.
6. Determine methods of accessing data.
7. Design data collection forms.
8. Determine the sample plan and size.
9. Collect data.
10. Analyze data.
11. Communicate insights.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Types of Statistical Analyses Used in Marketing
Research
❖ Marketing research uses a variety of statistical
techniques to collect, summarize, and interpret data about
customers, markets, and business performance.
❖ These methods help researchers make informed
decisions, identify trends, and test hypotheses.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
1. Descriptive Analysis
Purpose: To summarize and describe data characteristics.
Use: Provides a clear picture of what is happening in the
market.
Common Techniques:
❖ Frequency distribution
❖Mean, median, mode
❖Percentage, variance, and standard deviation
❖Cross-tabulation (for comparing categories)
Example: Calculating the average customer satisfaction
score or market share percentage.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
2. Inferential Analysis
Purpose: To make generalizations or predictions about a
population based on a sample.
Use: Helps determine whether observed differences or
relationships are statistically significant.
Common Techniques:
• Hypothesis testing (t-test, z-test)
• Chi-square test
• ANOVA (Analysis of Variance)
• Confidence intervals
Example: Testing whether male and female customers differ
significantly in purchase frequency.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
3. Correlation and Regression Analysis
Purpose: To examine relationships between variables and
predict outcomes.
Use: Identifies which factors influence customer behavior.
Common Techniques:
❖ Correlation Analysis: Measures the strength and direction
of relationships between variables.
❖Simple Regression: Predicts the value of one variable from
another.
❖Multiple Regression: Predicts outcomes using several
independent variables.
Example: Predicting sales based on advertising expenditure
and price.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
4. Multivariate Analysis
Purpose: To analyze more than two variables simultaneously.
Use: Helps understand complex market phenomena and
segment customers.
Common Techniques:
❖ Factor Analysis (to identify underlying factors)
❖ Cluster Analysis (for market segmentation)
❖ Discriminant Analysis (to classify groups)
❖ MANOVA (Multivariate Analysis of Variance)
Example: Grouping customers into segments based on their
buying behavior and demographics.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
5. Non-Parametric Tests
Purpose: Used when data does not meet normal distribution
assumptions.
Use: Suitable for ordinal or nominal data.
Common Techniques:
❖Mann-Whitney U Test
❖Kruskal-Wallis Test
❖Wilcoxon Signed-Rank Test
Example: Comparing customer satisfaction levels between
two service branches when data is ranked.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
6. Time Series and Forecasting Analysis
Purpose: To analyze data collected over time and predict
future trends.
Use: Useful for sales forecasting, demand planning, and
market trend prediction.
Common Techniques:
❖ Moving averages
❖ Exponential smoothing
❖ ARIMA models
Example: Forecasting monthly sales for the next year based
on past performance.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
7. Conjoint and Discrete Choice Analysis
Purpose: To understand how consumers value different
product attributes.
Use: Helps in product design, pricing, and positioning.
Example: Determining which combination of features (price,
quality, color) is most preferred by customers.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
8. Structural Equation Modeling (SEM)
Purpose: To test complex relationships between observed
and latent (unobserved) variables.
Use: Often applied in brand loyalty, customer satisfaction,
and behavioral intention studies.
Example: Analyzing how perceived quality and brand trust
influence customer loyalty.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Summary of Descriptive Statistical Tools in Marketing Research
Tool Purpose / What It Measures Type of Data Example Use in Marketing Research
Shows how often each value or All types (nominal, ordinal, Number of customers in each age
Frequency Distribution
category occurs in the dataset. interval, ratio) group.
Calculates the average (arithmetic
Mean Interval / Ratio Average customer satisfaction score.
center) of the data.
Identifies the middle value when data is
Median Ordinal / Interval / Ratio Median household income of buyers.
arranged in order.
Identifies the most frequently occurring
Mode Nominal / Ordinal Most preferred product brand.
value or category.
Expresses category frequencies as a
Percentage All types % of customers preferring each brand.
proportion of the total (×100).
Measures how much data points Variability in monthly sales between
Variance Interval / Ratio
deviate from the mean (spread). regions.
Square root of variance; indicates Consistency of customer satisfaction
Standard Deviation (SD) Interval / Ratio
average deviation from the mean. scores.
Measures asymmetry of the data
Detecting if income distribution of
distribution around the mean. • Positive
Skewness Interval / Ratio customers is skewed (many low
skew: long right tail • Negative skew:
incomes, few high).
long left tail
Measures the peakedness or flatness of
Understanding whether most
the data distribution. • High kurtosis =
Kurtosis Interval / Ratio customers rate satisfaction near the
sharp peak (many similar values) • Low
mean or vary widely.
kurtosis = flat (more spread out)
Compares two or more categorical
Gender vs. Brand preference or
Cross-tabulation variables to identify relationships or Nominal / Ordinal
Region vs. Purchase frequency.
patterns.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
How to do descriptive statistics in SPSS
Analyze + Descriptive statistics+ Descriptive
Descriptive Statistics
Minimu Maximu Std.
N m m Mean Deviation Skewness Kurtosis
Std. Std.
Statistic Statistic Statistic Statistic Statistic Statistic Error Statistic Error
Government 204 1.00 4.80 2.1431 .93551 .772 .170 -.172 .339
Financial Facilites
Cereal@crops 204 1.00 5.00 2.7826 1.00668 -.003 .170 -.961 .339
Horticulture 204 1.00 5.00 2.7127 .96950 .151 .170 -.665 .339
Livestock 204 1.00 4.80 2.7434 1.03702 .012 .170 -1.067 .339
Extension 204 1.00 5.00 2.6262 .98846 .349 .170 -.661 .339
Valid N (listwise) 204
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Reliability Test in SPSS: A Reliability Test in SPSS is used to check how
consistent or dependable your questionnaire or measurement scale is.
Reliability = Consistency of measurement
How to do reliability test in SPSS: Compute Cronbach’s alpha
Analyze + Scale + Reliability analysis +(selects some items) from statistics
option.
• Reliability Statistics
• Cronbach's Alpha Cronbach's Alpha Based on Standardized Items
N of Items
• .866 .869 5
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
How to do hypothesis test in SPSS:
Prior to conducting hypothesis testing, it is essential
to assess the normality of the data. The results of
this assessment determine the appropriate
analytical approach: parametric tests are applied if
the data follow a normal distribution, whereas non-
parametric tests are used when the data deviate
from normality
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
How to do Normality test in SPSS: A Normality Test in SPSS
checks whether your data follows a Normal Distribution (bell-
shaped curve).
In simple words:
It tests whether your data are normally distributed or skewed.
SPSS provides two main normality tests:
Test Name Best For Interpretation
Shapiro–Wilk Test Small samples (n < 50) More powerful for small datasets
Kolmogorov–Smirnov Test Large samples (n ≥ 50) Commonly used for big samples
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
How to Do It in SPSS:
❖ Go to Analyze → Descriptive Statistics → Explore
Move your variable(s) to “Dependent List”
Click Plots → Normality plots with tests → Continue →
OK
• SPSS will show:
• Histograms
• Q–Q plots
• Test results (Shapiro–Wilk and Kolmogorov–Smirnov)
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
GovernmentFinancialFacilites .129 204 .000 .922 204 .000
Cereal@crops .103 204 .000 .970 204 .000
Horticulture .079 204 .004 .979 204 .004
Livestock .092 204 .000 .962 204 .000
Extension .103 204 .000 .969 204 .000
a. Lilliefors Significance Correction
p-value (Sig.) Interpretation Conclusion
p > 0.05 Data are normally distributed ✅ Normal
p < 0.05 Data are NOT normally distributed ❌ Not normal
Because the p-value is below 0.05, it indicates that the data are
not normally distributed
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Sample chart with only the names of common parametric
and non-parametric tests
Scenario Parametric Test Non-Parametric Test
One group mean One-sample t-test One-sample Wilcoxon
Two independent groups Independent samples t-test Mann-Whitney U
Two related groups Paired-samples t-test Wilcoxon signed-rank
Three or more independent groups One-way ANOVA Kruskal-Wallis
Three or more related groups Repeated measures ANOVA Friedman
Correlation between variables Pearson correlation Spearman correlation
Predicting DV from IVs Linear regression Non-parametric regression*
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Scenario
DV: Cereal crops (continuous or ordinal, but non-normal)
IV: Government financial support (continuous or ordinal,
non-normal)
Normality test result: Not normally distributed →
parametric tests like linear regression or Pearson
correlation are not suitable.
Hypotheses:
• H₀: : There is no significant association between government financial
support and cereal crop production.
• H₁: There is a significant positive association between government
financial support and cereal crop production
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Choosing a Non-Parametric Test
You have one independent variable and one dependent
variable, both non-normal.
In this case, the non-parametric alternative to
correlation/regression is:
❖ Spearman Rank-Order Correlation
❖ If Spearman correlation is significant (p < 0.05), you can
conclude that government financial support impacts cereal
crops.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Why not Mann-Whitney or Kruskal-Wallis?
• Those are used when the IV is categorical with 2 or more groups.
• In your case, the IV seems continuous or ordinal, so correlation is
suitable.
Steps in SPSS
❖ Go to: Analyze → Correlate → Bivariate…
❖ Move Cereal crops (DV) and Government financial support (IV) into
the Variables box.
❖ Tick Spearman under Correlation Coefficients.
❖ Click OK
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Continue…
Correlations
GovernmentFina
ncialFacilites Cereal@crops
Spearman's rho GovernmentFinancialFacilite Correlation Coefficient 1.000 .724**
s
Sig. (2-tailed) . .000
N 204 204
Cereal@crops Correlation Coefficient .724** 1.000
Sig. (2-tailed) .000 .
N 204 204
**. Correlation is significant at the 0.01 level (2-tailed).
If Spearman correlation is significant (p < 0.05), you can conclude that government
financial support impacts cereal crops.
H₀: There is no significant association between government financial support and cereal crop production.
H₁: There is a significant positive association between government financial support and cereal crop production.
Now based on the above test the p < 0.005 → clearly less than 0.05.
Reject H₀: There is no significant association between government financial support and cereal crop
production
Accept H₁: There is a significant positive association between government financial support and cereal
crop production.
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved
Thanks
Copyright © 2020, 2017, 2014 Pearson Education, Inc. All Rights Reserved