Lecture 2
Two worldviews and their Nomothetic approach → human behaviour is generalizable, predictable and motivated by events
definitions personality and other people. Understanding behaviour is best done by isolating factors.
Idiographic approach → each person is unique, unpredictable and self-motivated. Understanding
behaviour is best done by listening to the participants perspective and considering the whole
situation.
Characteristics of Testing a hypothesis
nomothetic approach: Systematic manipulation (isolating factors)
Clear causal relation
Provides a generalizable result
Characteristics of No hypothesis, but a data driven approach
idiographic approach: Real communicative setting
Diverse set of observations
Qualitative research Methods that are build around the idea that human traits and behaviours can be translated
into numbers and analyse them.
Quantitative research Methods that are build around the idea that human traits and behaviours are best
understood by observations and interviewing people.
Typically qualitative Interviews, focus groups, case studies, observational studies, diary studies.
methods
Typically quantitative Experiments, focus groups, content analysis, meta analysis, computational modelling
methods
Triangulation The act of combining methods to neutralize weaknesses of single research methods. “It is not
all binary choices”
WEIRD-bias Acronym for Western Educated Industrialized Richa and Democrat: most participants studies.
Induction Observation -> Pattern -> Hypothesis -> Theory
If you have a certain observation, you might generalize that by saying that people who play
violent video games are more likely to become violent. (mostly for new studies)
Deduction Theory -> Hypothesis -> Observations
It starts with a theory and based on that people formulate a hypothesis. After writing the
hypothesis, people will start testing or conducting experiments to test their hypothesis. “If violent
videogames cause aggression, gamers will, on average, have a higher incidence of committing
violent acts than non-gamers.”
Research purposes Exploration, description, explanation, prediction, control, interpretation and criticism.
Dimensions of research
Lecture 3
Theory Simplified model of the world that can be used to explain or predict behavior
Requirements theory Cover a broad area (not too narrow)
Should be falsified (must be possible to prove partially wrong, in order for better
theories)
Generate hypotheses (be productive)
Substantiated by converging evidence
, Four types of variables + 1. Nominal – assigns categories or labels
characteristics 2. Ordinal – ranks categories in order
3. Interval – known, equal intervals (like Celsius and Fahrenheit)
4. Ratio – has a true zero point.
Hypotheses Hypotheses are statements that specify the relation we expect to find and. Hypotheses should be
testable and are not said to be true or false.
Two types of correlation 1. Two-tailed: relationship between two variables.
hypothesis 2. One tailed: specifycing the direction of the relationship (positive or negative related)
These are simply predictions that if one variable changes by a certain amount, another variable
will also change by a certain amount. Correlation is not causation.
Causal hypothesesmee All hypotheses containing: causes/leads to/contribute to. This means the writer of the hypothesis
wants to specify the relationship they expect to find between variables. Tested with
experiments.
Constructs Variables that are not able to be seen or touched like abstract concepts as attitude or motivation.
Dependent variable The outcome variable.
Independent variable A re those that we study in relation to the outcome variable. They might cause or correlate with
variation in the outcome variable. IV’s are often manipulated by the researcher.
Interaction effect Sometimes, the relation between the IV and a DV is affected by the level of another IV
Moderators When the strength and or the relationship between the IV and de DV depends on the level of
another variable, this other variable acts as a moderator.
Recognized by: for people/readers/woman who are
Difference moderator and Strong idea about what moderates what (causality)
interaction effect Not primarily interested in the effect of moderating variable on the DV, only in its
moderating role (otherwise it would just be an independent variable)
Mediation When there is mediation, the causal effect of IV on DV if indirect. The IV influences the mediating
variable which in turn influences the DV. In other words, the mediator is the transmitter between
the IV and the DV.
Lecture 4
How to prove a causal Experiment
hypothesis?
We can say that VARA 1. A precedes B in time
causes B if these conditions 2. A and B are Covary (they need to be positively/negatively correlated)
are met: 3. It is certain that A causes B - To do that, we need to observe B under different conditions
of A while, making sure that all else is equal.
Solomon four group design Group 1: R O1 X O2
Group 2: R 01 O2 include this group to isolate the effect of the experimental
manipulation
Group 3: R X O2 include this group to isolate the effect of pretesting
Group 4: R O2 include this group to isolate the effect of the experimental
manipulation AND pretesting
Risks of pretest: having participated in the pre-test can itself also lead to better scores on the post-test, o to make
sure that students have actually learned from the teaching and not from taking the pre-test,
When are pretests not when participants (provided the sample is large enough) are randomly assigned to conditions
necessary?
What is a factorial design? A 2x2 design with 2 or more IVs to study interaction effects
Between subject design Between-subject design: a given participant is exposed to only one level of each independent
variable.
Participant 1 Sees 2 clear profile pictures
Participant 2 Sees 2 blurred profile pictures
Two worldviews and their Nomothetic approach → human behaviour is generalizable, predictable and motivated by events
definitions personality and other people. Understanding behaviour is best done by isolating factors.
Idiographic approach → each person is unique, unpredictable and self-motivated. Understanding
behaviour is best done by listening to the participants perspective and considering the whole
situation.
Characteristics of Testing a hypothesis
nomothetic approach: Systematic manipulation (isolating factors)
Clear causal relation
Provides a generalizable result
Characteristics of No hypothesis, but a data driven approach
idiographic approach: Real communicative setting
Diverse set of observations
Qualitative research Methods that are build around the idea that human traits and behaviours can be translated
into numbers and analyse them.
Quantitative research Methods that are build around the idea that human traits and behaviours are best
understood by observations and interviewing people.
Typically qualitative Interviews, focus groups, case studies, observational studies, diary studies.
methods
Typically quantitative Experiments, focus groups, content analysis, meta analysis, computational modelling
methods
Triangulation The act of combining methods to neutralize weaknesses of single research methods. “It is not
all binary choices”
WEIRD-bias Acronym for Western Educated Industrialized Richa and Democrat: most participants studies.
Induction Observation -> Pattern -> Hypothesis -> Theory
If you have a certain observation, you might generalize that by saying that people who play
violent video games are more likely to become violent. (mostly for new studies)
Deduction Theory -> Hypothesis -> Observations
It starts with a theory and based on that people formulate a hypothesis. After writing the
hypothesis, people will start testing or conducting experiments to test their hypothesis. “If violent
videogames cause aggression, gamers will, on average, have a higher incidence of committing
violent acts than non-gamers.”
Research purposes Exploration, description, explanation, prediction, control, interpretation and criticism.
Dimensions of research
Lecture 3
Theory Simplified model of the world that can be used to explain or predict behavior
Requirements theory Cover a broad area (not too narrow)
Should be falsified (must be possible to prove partially wrong, in order for better
theories)
Generate hypotheses (be productive)
Substantiated by converging evidence
, Four types of variables + 1. Nominal – assigns categories or labels
characteristics 2. Ordinal – ranks categories in order
3. Interval – known, equal intervals (like Celsius and Fahrenheit)
4. Ratio – has a true zero point.
Hypotheses Hypotheses are statements that specify the relation we expect to find and. Hypotheses should be
testable and are not said to be true or false.
Two types of correlation 1. Two-tailed: relationship between two variables.
hypothesis 2. One tailed: specifycing the direction of the relationship (positive or negative related)
These are simply predictions that if one variable changes by a certain amount, another variable
will also change by a certain amount. Correlation is not causation.
Causal hypothesesmee All hypotheses containing: causes/leads to/contribute to. This means the writer of the hypothesis
wants to specify the relationship they expect to find between variables. Tested with
experiments.
Constructs Variables that are not able to be seen or touched like abstract concepts as attitude or motivation.
Dependent variable The outcome variable.
Independent variable A re those that we study in relation to the outcome variable. They might cause or correlate with
variation in the outcome variable. IV’s are often manipulated by the researcher.
Interaction effect Sometimes, the relation between the IV and a DV is affected by the level of another IV
Moderators When the strength and or the relationship between the IV and de DV depends on the level of
another variable, this other variable acts as a moderator.
Recognized by: for people/readers/woman who are
Difference moderator and Strong idea about what moderates what (causality)
interaction effect Not primarily interested in the effect of moderating variable on the DV, only in its
moderating role (otherwise it would just be an independent variable)
Mediation When there is mediation, the causal effect of IV on DV if indirect. The IV influences the mediating
variable which in turn influences the DV. In other words, the mediator is the transmitter between
the IV and the DV.
Lecture 4
How to prove a causal Experiment
hypothesis?
We can say that VARA 1. A precedes B in time
causes B if these conditions 2. A and B are Covary (they need to be positively/negatively correlated)
are met: 3. It is certain that A causes B - To do that, we need to observe B under different conditions
of A while, making sure that all else is equal.
Solomon four group design Group 1: R O1 X O2
Group 2: R 01 O2 include this group to isolate the effect of the experimental
manipulation
Group 3: R X O2 include this group to isolate the effect of pretesting
Group 4: R O2 include this group to isolate the effect of the experimental
manipulation AND pretesting
Risks of pretest: having participated in the pre-test can itself also lead to better scores on the post-test, o to make
sure that students have actually learned from the teaching and not from taking the pre-test,
When are pretests not when participants (provided the sample is large enough) are randomly assigned to conditions
necessary?
What is a factorial design? A 2x2 design with 2 or more IVs to study interaction effects
Between subject design Between-subject design: a given participant is exposed to only one level of each independent
variable.
Participant 1 Sees 2 clear profile pictures
Participant 2 Sees 2 blurred profile pictures
