Summary: Business Research Methods [HMA80a] – Axel Temmerman
Business Research Methods
Summary (2024 - 2025)
Brushing up - Hypothesis Testing & Linear Regression........................................2
1. Hypothesis testing and Linear regression.................................................................................... 2
Hypothesis testing......................................................................................................................2
Linear regression....................................................................................................................... 6
2. Introduction SPSS........................................................................................................ 12
Part 1: Logistic Regression.................................................................................... 13
Logistic regression: example................................................................................................... 13
Another example: prcancer.sav............................................................................................... 14
1. The Logistic Regression Model.................................................................................................. 15
Example cancer:...................................................................................................................... 15
2. Regression Coefficients..............................................................................................................17
Estimation method................................................................................................................... 17
Interpretation: in terms of probabilities.....................................................................................18
Interpretation: in terms of odds VERY IMPORTANT FOR EXAM............................................18
3. Testing hypotheses about the model.......................................................................................... 22
Likelihood ratio test.................................................................................................................. 22
4. Quality........................................................................................................................................ 25
5. Assumptions............................................................................................................................... 29
Linearity?................................................................................................................................. 29
Outliers.....................................................................................................................................29
Quasi-multicollinearity (QMC).................................................................................................. 29
Quasi-complete separation (QCS)...........................................................................................30
Part 2: Factor Analysis........................................................................................................ 32
Correlation and Factors....................................................................................................32
1. Correlation Matrix................................................................................................... 34
2. Factors................................................................................................................... 37
3. Interpretation.......................................................................................................... 45
4. Factor scores..........................................................................................................48
Part 3: Reliability analysis...................................................................................................51
Part 4: Cluster Analysis.......................................................................................................56
Cluster analysis methods................................................................................................. 56
Hierarchical clustering................................................................................................ 56
K-means clustering.....................................................................................................64
1
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
Brushing up - Hypothesis Testing & Linear Regression
1. Hypothesis testing and Linear regression
Hypothesis testing
Different types of data
● Qualitative data
○ Nominal data
E.g. Type of car, gender (dummy variable), city you live in
○ Ordinal data
E.g. Attitude against something (totally agree, …, totally disagree),
level of education
● Quantitative data
E.g. Price, exam score, number of customers, GDP, wage
Obtaining the data
Where does the data come from?
● Experiments (ex: medical experiments)
● Observation (ex: facebook, shops)
● Survey (e.g.: find 500 customers and ask about satisfaction with a product)
When looking for data on a specific question:
● Sometimes we can obtain data on all individuals concerned (e.g.: data on all
clients having used the product): population data
● Sometimes we can only obtain data on some individuals concerned (e.g.:
data on 500 clients out of 10000 clients having used the service: sample data
Hypothesis testing – In practice
When we want to test a statement about a population, using sample data, we
perform hypothesis testing
(When we have data on the whole population, hypothesis testing is not necessary!)
A hypothesis = A statement (about a population parameter)
Examples:
● TV Adertisements increase sales
● Anti-immigration statements by a political party increase voting shares for that
party
→ How do we test whether a specific statement is true?
2
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
Hypothesis testing = formal procedure used by statisticians to accept or reject a
hypothesis
Formal procedure (assuming you already have the data:
1. State the hypotheses (H1 and H1)
2. Formulate analysis plan (specify formula or test statistic and the significance
level)
3. Analyze data (compute value of test statistic from data)
4. Interpret results (reject H0 in favor of H1, or not)
Null hypothesis versus Alternative hypothesis
➔ Alternative hypothesis = research hypothesis
Something new, something controversial
E.g. statistics exam:
E.g. wages:
➔ Can we reject H0?
One-sided tests versus two-sided tests
● One-sided test
E.g. statistics exam:
● Two-sided test
E.g. wages:
Specify the test statistic and significance level
We need a formula based on which to decide whether to reject H1 in favor of H1
⇒test statistic
There are different formulas for test statistics, depending on the type of data and on
hypothesis H0 and H1
Specify the test statistic and significance level
The formal testing procedure can lead to errors
● Type 1 error: the null hypothesis is true but we reject it
● Type 2 error: the null hypothesis is false but we fail to reject it
→ How likely are we to commit an error?
3
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
We choose the significance level:
The significance level is the probability to make a type 1 error.
Usually:
Analyze data (compute value of test statistic from data)
● This is done by statistical software (we use SPSS)
○ Open database
○ Ask SPSS to test a hypothesis
○ ⇒ SPSS output
→ Interpret results (reject H0 in favor of H1, or not)
In SPSS output, check the significance level of the test statistic (p-value)
If p-value < ⇒ reject H0 in favor of H1.
Hypothesis test: example 1
One-sided test
Statement: “More than half of the students pass their statistics exam”
Hypotheses:
(is the proportion of students who pass significantly higher than 50%?)
→ We have data on a sample of 50 students
T-distribution:
SPSS output:
4
Business Research Methods
Summary (2024 - 2025)
Brushing up - Hypothesis Testing & Linear Regression........................................2
1. Hypothesis testing and Linear regression.................................................................................... 2
Hypothesis testing......................................................................................................................2
Linear regression....................................................................................................................... 6
2. Introduction SPSS........................................................................................................ 12
Part 1: Logistic Regression.................................................................................... 13
Logistic regression: example................................................................................................... 13
Another example: prcancer.sav............................................................................................... 14
1. The Logistic Regression Model.................................................................................................. 15
Example cancer:...................................................................................................................... 15
2. Regression Coefficients..............................................................................................................17
Estimation method................................................................................................................... 17
Interpretation: in terms of probabilities.....................................................................................18
Interpretation: in terms of odds VERY IMPORTANT FOR EXAM............................................18
3. Testing hypotheses about the model.......................................................................................... 22
Likelihood ratio test.................................................................................................................. 22
4. Quality........................................................................................................................................ 25
5. Assumptions............................................................................................................................... 29
Linearity?................................................................................................................................. 29
Outliers.....................................................................................................................................29
Quasi-multicollinearity (QMC).................................................................................................. 29
Quasi-complete separation (QCS)...........................................................................................30
Part 2: Factor Analysis........................................................................................................ 32
Correlation and Factors....................................................................................................32
1. Correlation Matrix................................................................................................... 34
2. Factors................................................................................................................... 37
3. Interpretation.......................................................................................................... 45
4. Factor scores..........................................................................................................48
Part 3: Reliability analysis...................................................................................................51
Part 4: Cluster Analysis.......................................................................................................56
Cluster analysis methods................................................................................................. 56
Hierarchical clustering................................................................................................ 56
K-means clustering.....................................................................................................64
1
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
Brushing up - Hypothesis Testing & Linear Regression
1. Hypothesis testing and Linear regression
Hypothesis testing
Different types of data
● Qualitative data
○ Nominal data
E.g. Type of car, gender (dummy variable), city you live in
○ Ordinal data
E.g. Attitude against something (totally agree, …, totally disagree),
level of education
● Quantitative data
E.g. Price, exam score, number of customers, GDP, wage
Obtaining the data
Where does the data come from?
● Experiments (ex: medical experiments)
● Observation (ex: facebook, shops)
● Survey (e.g.: find 500 customers and ask about satisfaction with a product)
When looking for data on a specific question:
● Sometimes we can obtain data on all individuals concerned (e.g.: data on all
clients having used the product): population data
● Sometimes we can only obtain data on some individuals concerned (e.g.:
data on 500 clients out of 10000 clients having used the service: sample data
Hypothesis testing – In practice
When we want to test a statement about a population, using sample data, we
perform hypothesis testing
(When we have data on the whole population, hypothesis testing is not necessary!)
A hypothesis = A statement (about a population parameter)
Examples:
● TV Adertisements increase sales
● Anti-immigration statements by a political party increase voting shares for that
party
→ How do we test whether a specific statement is true?
2
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
Hypothesis testing = formal procedure used by statisticians to accept or reject a
hypothesis
Formal procedure (assuming you already have the data:
1. State the hypotheses (H1 and H1)
2. Formulate analysis plan (specify formula or test statistic and the significance
level)
3. Analyze data (compute value of test statistic from data)
4. Interpret results (reject H0 in favor of H1, or not)
Null hypothesis versus Alternative hypothesis
➔ Alternative hypothesis = research hypothesis
Something new, something controversial
E.g. statistics exam:
E.g. wages:
➔ Can we reject H0?
One-sided tests versus two-sided tests
● One-sided test
E.g. statistics exam:
● Two-sided test
E.g. wages:
Specify the test statistic and significance level
We need a formula based on which to decide whether to reject H1 in favor of H1
⇒test statistic
There are different formulas for test statistics, depending on the type of data and on
hypothesis H0 and H1
Specify the test statistic and significance level
The formal testing procedure can lead to errors
● Type 1 error: the null hypothesis is true but we reject it
● Type 2 error: the null hypothesis is false but we fail to reject it
→ How likely are we to commit an error?
3
, Summary: Business Research Methods [HMA80a] – Axel Temmerman
We choose the significance level:
The significance level is the probability to make a type 1 error.
Usually:
Analyze data (compute value of test statistic from data)
● This is done by statistical software (we use SPSS)
○ Open database
○ Ask SPSS to test a hypothesis
○ ⇒ SPSS output
→ Interpret results (reject H0 in favor of H1, or not)
In SPSS output, check the significance level of the test statistic (p-value)
If p-value < ⇒ reject H0 in favor of H1.
Hypothesis test: example 1
One-sided test
Statement: “More than half of the students pass their statistics exam”
Hypotheses:
(is the proportion of students who pass significantly higher than 50%?)
→ We have data on a sample of 50 students
T-distribution:
SPSS output:
4
