Lecture 1: Introduction to scientific reasoning
Psychology as a way of thinking
Producing vs. consuming research
Research producer = writing a study, doing research, …
→ Research consumer
Critical mindset is essential
Not all published research is correct or robust
Replication crisis in psychology
How scientists work
Science is based on empiricism
Scientists test theories
Scientists work on fundamental and applied problems
Science is continuously evolving
Scientists publish their findings in scientific journals
Scientists communicate with the general public via journalists
Empiricism
Empirical method is based on data obtained through:
Our senses (sight, hearing, touch)
Instruments that assist our senses (thermometer, questionnaires, timer) → indirectly
Both quantitative and qualitative methods can be used to collect empirical evidence
Empiricists aim to do research in a systematic, rigorous, and replicable manner
Empiricism is not based on own experiences, intuition, or authority figures
Scientists test theories
Theory-Data cycle (confirmatory research)
Note: There is a difference between exploratory (inductive) and confirmatory
(deductive) research!
Harlow (1958): Cupboard theory vs. Contact comfort theory
⇒ to understand why humans/animals attach to their parents. Cupboard theory →
mother provides food as reason to be attached. Contact comfort theory →
affection, cuddled by the motherm bonding,.. creates attachment
He tested this by taking baby monkeys away from their mothers. They were put in a
cage with a fake mother figure, and a colder mother figure with food. Here we see
the theory data cycle in action!
Characteristics of good theories
Supported by data (however, results from a single study cannot prove a theory)
Falsifiable
Parsimonious (“Occam’s razor”)
If you have multiple theories on a certain fenomenon, then the most simple
one will be the more correct one. → Simplicity wins!
Examples of non-falsifiable theories:
Facilitated communication treatment believers
See book “De ongelovige Thomas heeft een punt”
Theories are evaluated based on all available evidence
A theory can never be “proven”, but it can be falsified - one study wont be enough
to prove a theory
Replication is crucial
Scientists form a community
Lecture 1: Introduction to scientific reasoning 1
, Fundamental vs. applied research
Basic research - trying to advance our understanding of a certain fenomenon, interest in theoretical understanding
Applied research - does an application have an effect.
Continiously evolving
Theories are continuously tested, modified, and falsified
→ Research triggers follow-up research (example: impact of color on approach- and avoidance- motivation in context (Meier et al., 2012))
Publishing
Manuscripts are submitted to scientific journals (e.g., Nature, Science, Psychological Bulletin)
Peer-review process
Parties involved include editor, reviewers, authors
Reject, revise and resubmit, accept
Number of submitted and published articles at Nature between 1997 and 2017 - Percentage of articles accepted at Nature between 1997 and
2017
predatory journal; articles that paid the journal to be published!
Communication
The science news cycle
People can communicate through the media → BUT journalists dont always share what the researchers are doing entirely, so not always
trustworthy. VB “listening to Mozart makes u smarter - but the research was about intelligince test after listening to it vs not listening to it. The
media reported this in a different way.
Sources of information
Researchers: research allows us to
Use a comparison group
Control for third variables
Try to evaluate information without bias
Research is better than:
Experience
Intuition
Authority arguments
Research vs. experience
Experience has no comparison group
Example: treatment of breast cancer → without a comparison we cant know if this treatment is effective or not
Experience is confounded by third variables
We do not control our own experience for multiple, simultaneously acting factors that influence our feelings and behaviors
Using exercise to deal with anger
Bottom line: research is better than own experiences
Researchers:
use a control group
control for confounders
Lecture 1: Introduction to scientific reasoning 2
, attempt to evaluate information without bias
Example: Bushman (2002) - Catharsis hypothesis
participants were angered and then either sat quietly, punched a bag as exercise, or punched a bag while thinking about the person who
upset them.
Result: the “think about the person” group became the most aggressive afterward. Conclusion: catharsis doesn’t reduce anger—acting it out
can increase aggression.
Research is probabilistic → results give us the likeliness, not everyone will react the same way to certain things
We can also make decisions based on our intuition: Some cognitive biases:
A good story (its easy to be convinced by something that sounds like a logical and good story)
Availability heuristic/availability bias (its easier to believe things that are available in our memory, for example: causes of death)
Confirmation bias (we believe things that allign with our own existing ideas)
Confirmatory hypothesis testing
Bias blind spot (its difficult to recognise our own bias)
Authority figures
We need to be careful with our beliefs based on what authority figures tell us.
Scientific sources
Empirical articles published in scientific journals
Review articles published in scientific journals (e.g., meta-analyses)
Chapters in books (edited books)
Scientific books
Books for broad audience
Blogs, internet sites
Popular media (e.g., news magazines)
Pyramid of scientific evidence
OVERVIEW
Lecture 1: Introduction to scientific reasoning 3
, Lecture 2: Foundations of research
Three claims, four validities (chapter 3)
Variables
A variable is something that varies, meaning that it has at least two levels or values. We need to operationalise our constructs so we can study
them. There are different ways we can operationalise our constructs. (self-report questionnaire, checking records, teacher’s observations…) →
biases can show f.e. a halo bias. We need to reflect critically.
Variables versus constants
Independent versus dependent variables
Measured versus manipulated variables
Conceptual (theoretical) versus operational (meetbaar, manipuleerbaar) variables
Way of describing a
Definition Example 1 Example 2
variable
Construct, conceptual The name of the concept
Satisfaction with life Perseverence (in young children)
variable being studied
A careful, theoretical A person's cognitive evaluation of his or her life The ability to push through when confronted with ...
Conceptual definition
definition of the construct (Diener et al., 1985) obstacles (White et al., 2017)
Five questionnaire items on the Satisfaction with How long a child will choose to engage in a slow-paced,
How the construct is
Operational definition, Life scale, answered on a scale of 1 (strongly boring activity that involves pressing a button when
measured or manipulated
operationalization disagree) to 5 (strongly agree). An example item is: they see a picture of cheese on a screen and not
in an actual study
All in all, I am satisfied with my life pressing it when they see a cat on the screen
Imagine that you are interested in studying the following research question:
Do people who send text messages while driving a car have a higher likelihood of car accidents?
What is the independent variable? texting
What is the dependent variable? accidents
How could you measure the independent variable?
How could you manipulate the independent variable?
Three claims made by researchers
Frequency claims (a certain percentage experiences x)
►“89% of teenagers are active on social media.”
►“Most people who return to work after burning out relapse.”
Association claims
►“Speech delays could be linked to mobile devices.”
►“Employees that have to deal with aggressive customers are more likely to experience burnout.”
Causal claims (more specific)
►“Family meals curb/decrease eating disorders.”
►“To appear intimidating, just tilt your head down.”
1. Frequency claims
Examples:
“60% of drivers text while driving.”
“12% of the voters prefers political party X.”
Also called descriptive research
Claims are about 1 variable that is measured
However, 1 study can make frequency claims about multiple variables
2. Association claims
Lecture 2: Foundations of research 1
Psychology as a way of thinking
Producing vs. consuming research
Research producer = writing a study, doing research, …
→ Research consumer
Critical mindset is essential
Not all published research is correct or robust
Replication crisis in psychology
How scientists work
Science is based on empiricism
Scientists test theories
Scientists work on fundamental and applied problems
Science is continuously evolving
Scientists publish their findings in scientific journals
Scientists communicate with the general public via journalists
Empiricism
Empirical method is based on data obtained through:
Our senses (sight, hearing, touch)
Instruments that assist our senses (thermometer, questionnaires, timer) → indirectly
Both quantitative and qualitative methods can be used to collect empirical evidence
Empiricists aim to do research in a systematic, rigorous, and replicable manner
Empiricism is not based on own experiences, intuition, or authority figures
Scientists test theories
Theory-Data cycle (confirmatory research)
Note: There is a difference between exploratory (inductive) and confirmatory
(deductive) research!
Harlow (1958): Cupboard theory vs. Contact comfort theory
⇒ to understand why humans/animals attach to their parents. Cupboard theory →
mother provides food as reason to be attached. Contact comfort theory →
affection, cuddled by the motherm bonding,.. creates attachment
He tested this by taking baby monkeys away from their mothers. They were put in a
cage with a fake mother figure, and a colder mother figure with food. Here we see
the theory data cycle in action!
Characteristics of good theories
Supported by data (however, results from a single study cannot prove a theory)
Falsifiable
Parsimonious (“Occam’s razor”)
If you have multiple theories on a certain fenomenon, then the most simple
one will be the more correct one. → Simplicity wins!
Examples of non-falsifiable theories:
Facilitated communication treatment believers
See book “De ongelovige Thomas heeft een punt”
Theories are evaluated based on all available evidence
A theory can never be “proven”, but it can be falsified - one study wont be enough
to prove a theory
Replication is crucial
Scientists form a community
Lecture 1: Introduction to scientific reasoning 1
, Fundamental vs. applied research
Basic research - trying to advance our understanding of a certain fenomenon, interest in theoretical understanding
Applied research - does an application have an effect.
Continiously evolving
Theories are continuously tested, modified, and falsified
→ Research triggers follow-up research (example: impact of color on approach- and avoidance- motivation in context (Meier et al., 2012))
Publishing
Manuscripts are submitted to scientific journals (e.g., Nature, Science, Psychological Bulletin)
Peer-review process
Parties involved include editor, reviewers, authors
Reject, revise and resubmit, accept
Number of submitted and published articles at Nature between 1997 and 2017 - Percentage of articles accepted at Nature between 1997 and
2017
predatory journal; articles that paid the journal to be published!
Communication
The science news cycle
People can communicate through the media → BUT journalists dont always share what the researchers are doing entirely, so not always
trustworthy. VB “listening to Mozart makes u smarter - but the research was about intelligince test after listening to it vs not listening to it. The
media reported this in a different way.
Sources of information
Researchers: research allows us to
Use a comparison group
Control for third variables
Try to evaluate information without bias
Research is better than:
Experience
Intuition
Authority arguments
Research vs. experience
Experience has no comparison group
Example: treatment of breast cancer → without a comparison we cant know if this treatment is effective or not
Experience is confounded by third variables
We do not control our own experience for multiple, simultaneously acting factors that influence our feelings and behaviors
Using exercise to deal with anger
Bottom line: research is better than own experiences
Researchers:
use a control group
control for confounders
Lecture 1: Introduction to scientific reasoning 2
, attempt to evaluate information without bias
Example: Bushman (2002) - Catharsis hypothesis
participants were angered and then either sat quietly, punched a bag as exercise, or punched a bag while thinking about the person who
upset them.
Result: the “think about the person” group became the most aggressive afterward. Conclusion: catharsis doesn’t reduce anger—acting it out
can increase aggression.
Research is probabilistic → results give us the likeliness, not everyone will react the same way to certain things
We can also make decisions based on our intuition: Some cognitive biases:
A good story (its easy to be convinced by something that sounds like a logical and good story)
Availability heuristic/availability bias (its easier to believe things that are available in our memory, for example: causes of death)
Confirmation bias (we believe things that allign with our own existing ideas)
Confirmatory hypothesis testing
Bias blind spot (its difficult to recognise our own bias)
Authority figures
We need to be careful with our beliefs based on what authority figures tell us.
Scientific sources
Empirical articles published in scientific journals
Review articles published in scientific journals (e.g., meta-analyses)
Chapters in books (edited books)
Scientific books
Books for broad audience
Blogs, internet sites
Popular media (e.g., news magazines)
Pyramid of scientific evidence
OVERVIEW
Lecture 1: Introduction to scientific reasoning 3
, Lecture 2: Foundations of research
Three claims, four validities (chapter 3)
Variables
A variable is something that varies, meaning that it has at least two levels or values. We need to operationalise our constructs so we can study
them. There are different ways we can operationalise our constructs. (self-report questionnaire, checking records, teacher’s observations…) →
biases can show f.e. a halo bias. We need to reflect critically.
Variables versus constants
Independent versus dependent variables
Measured versus manipulated variables
Conceptual (theoretical) versus operational (meetbaar, manipuleerbaar) variables
Way of describing a
Definition Example 1 Example 2
variable
Construct, conceptual The name of the concept
Satisfaction with life Perseverence (in young children)
variable being studied
A careful, theoretical A person's cognitive evaluation of his or her life The ability to push through when confronted with ...
Conceptual definition
definition of the construct (Diener et al., 1985) obstacles (White et al., 2017)
Five questionnaire items on the Satisfaction with How long a child will choose to engage in a slow-paced,
How the construct is
Operational definition, Life scale, answered on a scale of 1 (strongly boring activity that involves pressing a button when
measured or manipulated
operationalization disagree) to 5 (strongly agree). An example item is: they see a picture of cheese on a screen and not
in an actual study
All in all, I am satisfied with my life pressing it when they see a cat on the screen
Imagine that you are interested in studying the following research question:
Do people who send text messages while driving a car have a higher likelihood of car accidents?
What is the independent variable? texting
What is the dependent variable? accidents
How could you measure the independent variable?
How could you manipulate the independent variable?
Three claims made by researchers
Frequency claims (a certain percentage experiences x)
►“89% of teenagers are active on social media.”
►“Most people who return to work after burning out relapse.”
Association claims
►“Speech delays could be linked to mobile devices.”
►“Employees that have to deal with aggressive customers are more likely to experience burnout.”
Causal claims (more specific)
►“Family meals curb/decrease eating disorders.”
►“To appear intimidating, just tilt your head down.”
1. Frequency claims
Examples:
“60% of drivers text while driving.”
“12% of the voters prefers political party X.”
Also called descriptive research
Claims are about 1 variable that is measured
However, 1 study can make frequency claims about multiple variables
2. Association claims
Lecture 2: Foundations of research 1
