Pre-Master
Business Administration
2019-2020
Management Research Methods II
,MRM 2 – Week 1 – Summary – One-Way ANOVA
Summary of all techniques
,Course details
This is not a statistics course. The focus is not on understanding the
mathematical process behind the techniques, but to understand the concept
behind the technique and how they can be applied using SPSS.
Different types of analytics
PV and OV
* OV = Outcome variable OR DV = Dependent variable à Test variable (to be
explained) (e.g. sales)
* PV = Predictor variable OR IV = Independent variable à Variable that explains
(e.g. advertising)
à In the example you want to understand how advertising affects sales.
P-value
* The p-value stands for the probability of obtaining a result (or test-statistic
value) equal to (or more extreme than) what was actually observed, assuming
that H0 is true. à A low p-value indicates that the null hypothesis is unlikely.
How to approach a data set?
1. Check the data for outliers with for example a histogram.
2. What are the predictor and outcome variable?
à Usually the outcome variables are expressed in numbers.
3. Find relationships between variables.
à Techniques like regression that help you understand them.
Conceptual models
Conceptual models are visual representations of relations between theoretical
constructs (and variables). By model we mean a simplified description of reality.
, Measurement scales of variables
* Categorical (nominal & ordinal) – subgroups are indicated by numbers
* Quantitative (discrete, interval & ratio) – We use numerical scales with equal
distances between values.
* In social sciences we sometimes treat ordinal scales as (pseudo)-interval
scales (e.g. Likert scales)
An example of a conceptual model
RQ: What factors determine student satisfaction?
With H1: Teachers that are more committed
increase student satisfaction level.
With H2: Teachers that are more committed will
increase the satisfaction level of students, when
they have good communication skills. The effect is
stronger in certain settings. “Communication
skills” is a moderating variable; it moderates the
relationship between two other variables. The
relationship of commitment and student
satisfaction is positive; the degree of positivity is
moderated by teacher’s communication skills.
With H3: The positive effect of teacher’s
commitment on student satisfaction is mediated
by quality of the course material. The
relationship goes via another variable.
“Lecture slides quality” here, is a mediating
variable; one that mediates the relationship
between two other variables. This entails that A
(commitment) leads to B (quality of slides), which ultimately leads to C (student
satisfaction).
ANOVA (ANALYSIS OF VARIANCE
There are two measurements of variability:
1. Variance à The average of the squared differences from the Mean.
2. Sum of Squares à The sum of the squared differences from the Mean.
Anova helps us to investigate the differences between groups with a certain
level of statistical confidence:
* By comparing the variability between the groups against the variability
within the groups.
* We want to see how much of the variability in our outcome variable can
be explained by our predictor variable. However, we probably won’t be
able to explain all the differences in exam scores, solely by creating our
group who receive different programs/predictor variables.
Business Administration
2019-2020
Management Research Methods II
,MRM 2 – Week 1 – Summary – One-Way ANOVA
Summary of all techniques
,Course details
This is not a statistics course. The focus is not on understanding the
mathematical process behind the techniques, but to understand the concept
behind the technique and how they can be applied using SPSS.
Different types of analytics
PV and OV
* OV = Outcome variable OR DV = Dependent variable à Test variable (to be
explained) (e.g. sales)
* PV = Predictor variable OR IV = Independent variable à Variable that explains
(e.g. advertising)
à In the example you want to understand how advertising affects sales.
P-value
* The p-value stands for the probability of obtaining a result (or test-statistic
value) equal to (or more extreme than) what was actually observed, assuming
that H0 is true. à A low p-value indicates that the null hypothesis is unlikely.
How to approach a data set?
1. Check the data for outliers with for example a histogram.
2. What are the predictor and outcome variable?
à Usually the outcome variables are expressed in numbers.
3. Find relationships between variables.
à Techniques like regression that help you understand them.
Conceptual models
Conceptual models are visual representations of relations between theoretical
constructs (and variables). By model we mean a simplified description of reality.
, Measurement scales of variables
* Categorical (nominal & ordinal) – subgroups are indicated by numbers
* Quantitative (discrete, interval & ratio) – We use numerical scales with equal
distances between values.
* In social sciences we sometimes treat ordinal scales as (pseudo)-interval
scales (e.g. Likert scales)
An example of a conceptual model
RQ: What factors determine student satisfaction?
With H1: Teachers that are more committed
increase student satisfaction level.
With H2: Teachers that are more committed will
increase the satisfaction level of students, when
they have good communication skills. The effect is
stronger in certain settings. “Communication
skills” is a moderating variable; it moderates the
relationship between two other variables. The
relationship of commitment and student
satisfaction is positive; the degree of positivity is
moderated by teacher’s communication skills.
With H3: The positive effect of teacher’s
commitment on student satisfaction is mediated
by quality of the course material. The
relationship goes via another variable.
“Lecture slides quality” here, is a mediating
variable; one that mediates the relationship
between two other variables. This entails that A
(commitment) leads to B (quality of slides), which ultimately leads to C (student
satisfaction).
ANOVA (ANALYSIS OF VARIANCE
There are two measurements of variability:
1. Variance à The average of the squared differences from the Mean.
2. Sum of Squares à The sum of the squared differences from the Mean.
Anova helps us to investigate the differences between groups with a certain
level of statistical confidence:
* By comparing the variability between the groups against the variability
within the groups.
* We want to see how much of the variability in our outcome variable can
be explained by our predictor variable. However, we probably won’t be
able to explain all the differences in exam scores, solely by creating our
group who receive different programs/predictor variables.
