Quantitative and design methods in business research
Inhoudsopgav
Quantitative and design methods in business research...................................................................................1
Lecture 1: Introduction.................................................................................................................................. 1
Lecture 2: Design vs. behavioral research....................................................................................................... 8
Lecture 3: Factor analysis............................................................................................................................. 12
Lecture 4: (Multivariate) Analysis of (Co-)Variance.......................................................................................25
Lecture 5 – Regression Analysis.................................................................................................................... 40
Lecture 6 – Guest lecture business analyst @PwC (forecasting using linear regression models).....................52
Lecture 7 – Structual Equiation Modeling..................................................................................................... 53
.................................................................................................................................................................... 65
.................................................................................................................................................................... 65
.................................................................................................................................................................... 65
Lecture 1: Introduction.................................................................................................................................. 1
Lecture 2: Design vs. behavioral research....................................................................................................... 7
Lecture 3: Factor analysis............................................................................................................................. 11
Lecture 4: (Multivariate) Analysis of (Co-)Variance.......................................................................................23
Lecture 5 – Regression Analysis.................................................................................................................... 35
Lecture 6 – Guest lecture business analyst @PwC (forecasting using linear regression models).....................46
Lecture 7 – Structual Equiation Modeling..................................................................................................... 47
Lecture 8 - Cluster Analysis……………………………………………………………………………………………………………….…………62
Lecture 1: Introduction
Multivariate techniques in this course:
- Factor analysis
- (M)AN(C)OVA
- Regression analysis
- Structural Equation Modeling
- Cluster analysis
,Multivariate analysis
Multivariate Data analysis comprises all statistical methods that
simultaneously analyze multiple measurements on each individual or
object under investigation. (For example three characters of one firm like
size, geographical location and strategy)
Motivation:
- Measurement
- Explanation & prediction (e.g. if you use this strategy, this can be
the consequence)
- Hypothesis testing
Basic concepts
- Measurement scales
o Nonmetric
o Metric
- Measurement and measurement error
- Statistical inference
- Types of techniques
Data can be:
- Nonmetric or qualitative
o Nominal scale (yes/no)
o Ordinal scale (e.g. education level, so there is order. But the
differences between the levels do not really have a meaning)
- Metric or quantitative
o Interval scale (numbers make sense also in subtracting. The
difference makes sense/is interpretable, but do not make
sense in multiplying, e.g. temperature)
o Ratio scale (numbers that make sense also in multiplying and
dividing. e.g. length)
The level of measurement is critical in determining the appropriate
multivariate technique to use!
Measurement error
- All variables have some error
- Measurement error = distorts observed relationships and makes
multivariate techniques less powerful
o Random measurement error
o Systematic measurement error
, - Researchers use summated scales, for which several variables are
summed or averaged together to form a composite representation of
a concept
- A: Shooting at the same point over and over again. It is also the right
point (e.g. the professional instructor)
- B: Shooting at the same point over and over again but it is not the
right point. (e.g. the instructor in the morning)
- C: Shooting everywhere in the circle.
- D: Shooting everywhere, and not even always in the circle.
Validity: you measure what you want to measure. Reliability: the observed
variable’s degree of precision and thus the lack of random measurement
error (measuring the right thing).
It relates to each other by: One who is not able to shoot creates for a high
random error. When being free of error, you have high validity.
Statistical significance and power
The null hypothesis is mostly about the fact that there is no effect.
- Type I error: the probability of rejecting the null hypothesis when it
is true
- Type II error: the probability of failing to reject the null hypothesis
when it is false
, - Power: the probability of rejecting the null hypothesis when it is false
- Effect size: the actual magnitude of the effect of interest (e.g. the
difference between means or the correlation between variables)
- Alpha: As alpha is set at smaller levels, statistical power decreases.
Typically, alpha = 5%
- Sample size: as sample size increases, power increases. With very
large sample sizes, even very small effects can be statistically
significant, raising the issue of practical significance vs. statistical
significance (are small effects really relevant, is it that large that we
should care about it? = practical significance. You may better refer
to practical significance as substantial).
If you want your alpha to be smaller, you need more observations.
Smaller amount of errors more data. Power can be increased by
choosing a less stringent alpha level (e.g. 10 instead of 5). Smaller
effect sizes always require larger sample sizes to achieve the
desired power. Any increase in power is most likely achieved by
increased sample size.
Types of multivariate techniques
Inhoudsopgav
Quantitative and design methods in business research...................................................................................1
Lecture 1: Introduction.................................................................................................................................. 1
Lecture 2: Design vs. behavioral research....................................................................................................... 8
Lecture 3: Factor analysis............................................................................................................................. 12
Lecture 4: (Multivariate) Analysis of (Co-)Variance.......................................................................................25
Lecture 5 – Regression Analysis.................................................................................................................... 40
Lecture 6 – Guest lecture business analyst @PwC (forecasting using linear regression models).....................52
Lecture 7 – Structual Equiation Modeling..................................................................................................... 53
.................................................................................................................................................................... 65
.................................................................................................................................................................... 65
.................................................................................................................................................................... 65
Lecture 1: Introduction.................................................................................................................................. 1
Lecture 2: Design vs. behavioral research....................................................................................................... 7
Lecture 3: Factor analysis............................................................................................................................. 11
Lecture 4: (Multivariate) Analysis of (Co-)Variance.......................................................................................23
Lecture 5 – Regression Analysis.................................................................................................................... 35
Lecture 6 – Guest lecture business analyst @PwC (forecasting using linear regression models).....................46
Lecture 7 – Structual Equiation Modeling..................................................................................................... 47
Lecture 8 - Cluster Analysis……………………………………………………………………………………………………………….…………62
Lecture 1: Introduction
Multivariate techniques in this course:
- Factor analysis
- (M)AN(C)OVA
- Regression analysis
- Structural Equation Modeling
- Cluster analysis
,Multivariate analysis
Multivariate Data analysis comprises all statistical methods that
simultaneously analyze multiple measurements on each individual or
object under investigation. (For example three characters of one firm like
size, geographical location and strategy)
Motivation:
- Measurement
- Explanation & prediction (e.g. if you use this strategy, this can be
the consequence)
- Hypothesis testing
Basic concepts
- Measurement scales
o Nonmetric
o Metric
- Measurement and measurement error
- Statistical inference
- Types of techniques
Data can be:
- Nonmetric or qualitative
o Nominal scale (yes/no)
o Ordinal scale (e.g. education level, so there is order. But the
differences between the levels do not really have a meaning)
- Metric or quantitative
o Interval scale (numbers make sense also in subtracting. The
difference makes sense/is interpretable, but do not make
sense in multiplying, e.g. temperature)
o Ratio scale (numbers that make sense also in multiplying and
dividing. e.g. length)
The level of measurement is critical in determining the appropriate
multivariate technique to use!
Measurement error
- All variables have some error
- Measurement error = distorts observed relationships and makes
multivariate techniques less powerful
o Random measurement error
o Systematic measurement error
, - Researchers use summated scales, for which several variables are
summed or averaged together to form a composite representation of
a concept
- A: Shooting at the same point over and over again. It is also the right
point (e.g. the professional instructor)
- B: Shooting at the same point over and over again but it is not the
right point. (e.g. the instructor in the morning)
- C: Shooting everywhere in the circle.
- D: Shooting everywhere, and not even always in the circle.
Validity: you measure what you want to measure. Reliability: the observed
variable’s degree of precision and thus the lack of random measurement
error (measuring the right thing).
It relates to each other by: One who is not able to shoot creates for a high
random error. When being free of error, you have high validity.
Statistical significance and power
The null hypothesis is mostly about the fact that there is no effect.
- Type I error: the probability of rejecting the null hypothesis when it
is true
- Type II error: the probability of failing to reject the null hypothesis
when it is false
, - Power: the probability of rejecting the null hypothesis when it is false
- Effect size: the actual magnitude of the effect of interest (e.g. the
difference between means or the correlation between variables)
- Alpha: As alpha is set at smaller levels, statistical power decreases.
Typically, alpha = 5%
- Sample size: as sample size increases, power increases. With very
large sample sizes, even very small effects can be statistically
significant, raising the issue of practical significance vs. statistical
significance (are small effects really relevant, is it that large that we
should care about it? = practical significance. You may better refer
to practical significance as substantial).
If you want your alpha to be smaller, you need more observations.
Smaller amount of errors more data. Power can be increased by
choosing a less stringent alpha level (e.g. 10 instead of 5). Smaller
effect sizes always require larger sample sizes to achieve the
desired power. Any increase in power is most likely achieved by
increased sample size.
Types of multivariate techniques
