Summary: Quantitative Research Methods – Kasper van der Hoofd
Contents
1a: Data & RQ .............................................................................................................................................. 3
What is quantitative research and why it is useful ..................................................................................... 3
Why learn about qrm? ........................................................................................................................... 3
When to use quantitative research ............................................................................................................ 3
Characteristics of quantitative research – Qualitative to Quantitative – phases ..................................... 3
Can we always trust research findings? ................................................................................................. 4
Coming up with a research question ......................................................................................................... 4
1st step of Quantitative study: Initial observation (RQ) ............................................................................... 5
1b: Conceptual domain: Theory & Generating Hypotheses .......................................................................... 6
Theory ...................................................................................................................................................... 6
Moderator – WHEN? ............................................................................................................................. 7
Mediator – WHY? .................................................................................................................................. 8
2a: Observable domain: Data collection and measurement .......................................................................... 8
Measurement in social science research................................................................................................... 9
Indicators ............................................................................................................................................... 9
Survey .................................................................................................................................................... 11
Example: developing a survey ................................................................................................................ 11
1. You gain interest in a certain phenomenon ......................................................................................... 11
2. You define and conceptualize everything ............................................................................................ 11
3. Operationalizing .................................................................................................................................. 11
Answering categories: ......................................................................................................................... 14
Control variables ................................................................................................................................. 15
4. Measuring – Data collection ................................................................................................................ 15
Power of statistical test ........................................................................................................................ 17
5. Analyzing data .................................................................................................................................... 18
3: How to test reliability & validity ................................................................................................................ 19
Examining measurement validity in SPSS............................................................................................... 19
Exploratory factor analysis (EFA) ........................................................................................................ 20
Steps: .................................................................................................................................................. 21
Assessing reliability in SPSS................................................................................................................... 23
Examining descriptive statistics in SPSS................................................................................................. 24
4a: Hypothesis testing ............................................................................................................................. 25
Basics:................................................................................................................................................. 26
Correlation and regression analysis ..................................................................................................... 26
Correlation ........................................................................................................................................... 27
1
, Summary: Quantitative Research Methods – Kasper van der Hoofd
Regression .......................................................................................................................................... 28
Mediation hypothesis ........................................................................................................................... 31
Moderation hypothesis ........................................................................................................................ 35
5b: Experimental research .......................................................................................................................... 39
T-test ................................................................................................................................................... 43
2
, Summary: Quantitative Research Methods – Kasper van der Hoofd
1a: Data & RQ
What is quantitative research and why it is useful
Explaining phenomena by collecting numerical data that are analyzed using mathematically based
methods (in particular statistics) → really tries to quantify relationships
Why learn about qrm?
QRM allow you to make better decisions that are based on sound interpretations of data and evidence. You
can achieve a higher decision making speed, that is based on isights from statistical analysis
When to use quantitative research
For what types of questions can we use QRM?
• Type 1 – WHAT: Descriptive questions:
o Interested in quantitative answer:
▪ ‘How many students are enrolled in the premaster?’ Which program has the most
students?
▪ What are the differences in firm innovation performance across the world?
(descriptive)
o Interested in numerical change: ‘Are the numbers of premaster students in our university
rising compared to last year?
o How did rates of small business ownership develop after the financial crisis in 2008?
(numerical change)
• Type 2 – WHY: explaining phenomena → with QRM you can get a better sense of predictors
o Test relationships:
▪ ‘What is the relationship between students’ self-esteem and their average grade in
the premaster?’
▪ When employees are more proactice and take more initiatives, do they get higher
performance evaluations? (test relationship)
o Explain something:
▪ ‘What factors predict changes in student performance over time?’
▪ Why are only some CEOs able to successfully navigate their companies during the
Covid-19 crisis? (explain something)
QRM is usefull for descriptive (WHAT), but also in WHY-studies that test inferential types of relationships
that explain phenomena
Characteristics of quantitative research – Qualitative to
Quantitative – phases
Qualitative… → → → … Quantitative
State of prior theory Nascent Intermediate Mature
Open-ended inquiry Proposed relationships Focused questions
Nature of the
about a phenomenon between new and related to existing
research questions
of interest established constructs constructs
3
, Summary: Quantitative Research Methods – Kasper van der Hoofd
Type of data Both qualitative and
Qualitative Quantitative
collected quantitative
Interviews,
Primary methods of Interviews, observations, Surveys, structured
observations,
data collection secondary data, surveys interviews
secondary data
Content analysis and
Iterative content
Primary methods of statistical tests aimed at Statistical tests of
analysis focused on
data analysis preliminary testing of formal hypotheses
pattern identification
propositions and constructs
Suggestive theory, Provisional theory, which Supported theory that
Theoretical
invitation to further often integrates previously extends existing
contribution
work separate literatures theories
Can we always trust research findings?
• Study is flawed
• Conclusions derived from study are mis-interpreted or mis-presented
→ we need to know how to judge quality of research findings
Characteristics of ‘good’ quantitative research:
• Clearly defined purpose (developing, extending theory
→ when you are not clear, all things that follow will not be clear
• Transparent research process and design
• in the way it was carried out
• measures that were used
• what were findings (significant and not significant?)
→ people need insight to judge quality and to replicate findings
• High ethical standards (follow high standards, → follow code of conduct)
• Sound analyses using appropriate methods (that fit the RQ and the data at hand)
You have to follow state-of-the art methods → to get best and most reliable findings
Next: report them in a transparent way
• Unambiguous presentation of (null) findings → detail patterns and data (and also report not
significant data) → we dont want BIASED reporting
• Conclusions match the findings
If findings find very weak relationships, then the conclusion should not overstate the importance of
the findings
• Awareness of the study’s limitations (the more aware you are of limitations, and the better you
report them → allows others to judge the quality of your study because it shows that you have
thought about different things)
Coming up with a research question
RQ – derived from own interest, can also be driven by empirical observations driven by data (something
interesting that has been observed in the world)
4
, Summary: Quantitative Research Methods – Kasper van der Hoofd
You search for theory to make sense of observation(s)
→ you can draw hypotheses to make sense of observations (predictions)
→ you collect data (you measure variables that are included in the hypotheses)
→ use statistics to measure whether relationships are statistically significant
→ if hypotheses are rejected then new hypotheses can be formulated in new studies
1st step of Quantitative study: Initial observation (RQ)
Keep in mind: whether the question that you ask is suited for quantitative research? – or is qualitative more
suited? Additionally, the question has to be interesting
Don’t focus on “uninteresting” questions:
• Obvious, irrelevant, absurd, definitional, affirmation questions
Search for “interesting” questions:
• Identify an empirical anomaly that existing theories cannot explain (e.g., providing more rewards
lowers task performance)
• Identify a phenomenon that challenges our understanding of the world (e.g., how does the corona
crisis impact firm innovation?)
• An audience finds an idea interesting not because it tells them some truth they did not already know,
but instead because it tells them some truth they thought they already knew is wrong
→ INTERESTING QUESTIONS CHALLENGE PEOPLE’s ASSUMPTIONS
5
, Summary: Quantitative Research Methods – Kasper van der Hoofd
1b: Conceptual domain: Theory &
Generating Hypotheses
Theory
We need theory to help us:
• identifying relevant variables
• come up with a logic, that helps us relate the variables
→ One of the best ways to review theory, is by doing a literature review.
A literature review, essentially reviews what is already known about the topic that is investigated:
• What are the key theories (that can help us making sense of the field
• What are the main variables (that have been studied in prior research studies?
• What are the key findings
• Is there consensus about the evidence?
o Helps establishing what is lacking
→ Theory:
• identifies main variables that are needed to understand the phenomenon of interest
• explains how these variables are interrelated
• Explains why these variables are related
• The boundary conditions under which these relations are expected to hold
Variables
• Independent (IV): the key predictor of the outcome we want to explain
• Dependent (DV): the variable that is affected by change in independent variable
• Control: the variable that you keep constant to rule out alternative explanations that can also
affect the dependent variable
• Moderating: the variable that influences the relationship between the IV and the DV
o help to understand WHEN the relationship between the IV and DV is stronger/weaker
• Mediating: the variable that explains the relationship between the IV and the DV
o Mediating variables help to understand WHY the IV influences the DV
6
, Summary: Quantitative Research Methods – Kasper van der Hoofd
Relationships between variables
→ What do we expect? (hypotheses)
• Positive relationship
• Negative relationship
• Non-linear relationship (e.g. up to a certain level, there is a positive relationship there is a positive
relation between income and happiness. However, after this threshold there will be a negative
relationship)
→ Why do we expect these relationships? Clarify the logic for this relationship
Points of attention while defining a hypothesis: → make it quantifiable, concrete, and therefore,
testable
Simple direct effects hypothesis
The most basic model:
→ Variable X has a positive/negative relationship with variable Y
Independent Dependent variable
variable (X) (Y)
Moderator – WHEN?
Moderating: the variable that influences the relationship between the IV and the DV
→ help to understand WHEN the relationship between the IV and DV is stronger/weaker
moderator influences the relationship between the IV and DV by strengthening or weakening it
Moderator
Independent Dependent variable
variable (X) (Y)
Moderation = combined effect of two variables
→ in statistical terms it is known as an interaction effect
When testing an moderation effect you need: 2 hypotheses
1. A hypothesis that makes prediction regarding the direct relationship (between dependent and
independent variable)
2. A hypothesis that specifies the nature of the moderation effect (strengthen/weaken the
relationship)
7
, Summary: Quantitative Research Methods – Kasper van der Hoofd
Mediator – WHY?
Mediating: the variable that explains the relationship between the IV and the DV
→ Mediating variables help to understand WHY the IV influences the DV
Focuses on mechanisms that make causal chain possible
Mediator
Independent Dependent variable
variable (X) (Y)
When testing an moderation effect you need: 4 hypotheses
1. The direct relationship between the independent and the dependent variable
2. The direct relationship between the independent variable and the mediator
3. The direct relationship between the mediator and the dependent variable
4. The fact that the mediator mediates the relationship between the independent and the dependent
variable
2a: Observable domain: Data collection
and measurement
After deriving a RQ, developing a theory and hypothesis that you want to test with your study, you can
focus on collecting data and testing predictions. How to move from conceptual model to measures that
test our hypotheses?
8
, Summary: Quantitative Research Methods – Kasper van der Hoofd
Measurement in social science research
& key difficulties
• Classic definition of measurement
o “the assignment of numbers to objects or events according to rules” (Stevens, 1951)
o Objects: “things that can be seen or touched”
o Events: “result, consequence, or outcome”
• Measurement problem in social sciences
o Phenomena to be measured are typically not objects or events, but abstract concepts
▪ Work motivation, cognitive dissonance, innovation, transformational leadership, etc.
→ This makes it difficult to find good indicators or measures of said concepts
→ How to actually measure concepts?
Conceptual
domain
Construct X Construct Y
Logic Reliability & Validity Logic
Observable of Measures
domain
Observable Observable
Response X’ Response Y’
Try to find indicators of the construct
But a new problem arises → to what extent do the answers of the survey questions do the answers
really capture said construct accurately?
In other words: When we link the responses together, to what extent can we really say that we tested a
relationship when when only tested the outcomes?
→ We may have used inferior/inappropriate indicators/response → that may inaccurately represent
the relationship between both constructs
→ therefore we need to access the validity and reliability of the measures
Indicators
Use ‘unobtrusive’ measures
9
, Summary: Quantitative Research Methods – Kasper van der Hoofd
• Physical traces (evidence people leave behind in their physical environment)
• Non-participant observation
• Documentary sources
• Written and spoken words of subjects
→ you can use these things to learn about preferences, perceptions, and personalities of subjects
Provide clear insight into your logic about why certain things are indicators for said construct
Also provide evidence
With a table like this →
Next, provide evidence for these constructs → with a correlation matrix
10
