Exam notes: Applied Research
Methods
Development and Mental Health
General information
Study questions resemble exam questions. Be able to answer the study questions
Exam about lectures, research papers and assignments. Multiple choice exam: 40-50 questions
Lecture 1: general topics
Scientific Research and Theory
Types of Scientific Research
• Observations: is about finding phenomena
o For example Skinner
• Correlations and Quasi-Experiments: is about finding relationships
o One goes up and other goes down/also up, but we don’t know why
o Maybe causal case XY but also XY so we don’t know the relation
• Experiments: is about finding causal explanations
o No other way to test causal relationships
• All of them: Developing and testing theories of experience and behavior
How do you tell a good theory from a bad one?
• Precision = The more precise a theory explains a phenomenon / the higher the accuracy
(nauwkeurigheid), the better. Its more precise when same results are obtained with repeated
measures.
• Parsimony = Try to explain a phenomenon with less things. The fewer assumptions to explain
something, the better. All things being equal, you should prefer the simplest possible explanation
for a phenomenon or the simplest possible solution to a problem.
• Testability = you can test the theory, and Falsifiability = the capacity for a theory to be proven
wrong: a theory that can explain everything is useless
Types of Validity
Validity = the test measures precisely what it aims to measure, meaning the data collected is
accurate and represents some truth compared to others outside of the study.
• Internal validity = to extent to which the study measures what it intended to study.
o did the intervention cause the results, rather than a confounded variable? (then you
measured something different and thus low internal validity) So is it a clean experiment, no
other variables influencing? is focused on the structure of a study and the accuracy of the
conclusions drawn based on a cause and effect relationship
• External validity = the extent to which you can generalize the findings of a study to other things.
o How far can the results be generalized?
o E.g: When studying depression in females it cannot be generalized to males.
• Construct validity = the extent to which your study accurately tests what it's supposed to.
, o Which aspect of the intervention caused the results?
o The extent to which the test or measurement fulfills its purpose. Important to operationalize
constructs into concrete and measurable characteristics based on your idea of the construct
and its dimensions. Construct validity answers questions about the measurement of a
concept or construct. E.g: Do you measure intelligence with an intelligence test?
• Statistical validity = the extent to which drawn conclusions of a research study can be considered
accurate and reliable from a statistical test. So: are the statistical conclusions correct?
Correlational research
Definitions (not from the lectures)
Correlation = the extent to which two variables are linearly related
(meaning they change together at a constant rate)
Regression = a measure of the relation between the mean value of one variable (e.g. output) and
corresponding values of other variables.
Correlational Research Questions
• Correlation: How closely are two
variables related?
• Regression: How can I predict one
variable if I know the other?
How can correlations be used and interpreted?
• Correlation: Direction and size of the relation between the two variables
• Regression: Prediction, is dependent on the correlation
Both can say nothing about causal interpretations!
An Example of the Causality Problem
• Depressed patients think more negatively about themselves than others
gives a correlation of depression and thinking. Possible explanations:
o Negative thinking causes depression?
o Depression causes negative thinking?
o Depression and negative thinking cause each other ?
o A third variable (genetic, neurological) causes both depression and negative thinking?
• The number of crimes and the number of churches in a city are correlated.
o Does religion cause crime? No; amount of people in a city is third variable.
• Sales of ice cream and drowning rates are correlated
o Does ice cream cause drowning? No; season is third variable.
• Shoe size is positively correlated with alcoholism, and negatively correlated with anxiety
o Do big feet cause alcoholism, but protect from anxiety? No; Gender is third variable.
Correlation and Causality
• The relation is not symmetric (kan het niet omdraaien)
o If there is causality between two variables, then they must be correlated
o But NOT if correlation between two variables, then also causality.
• And temporal order does NOT prove causality but it disproves it:
o If A is the cause of B, A must happen before B.
, o Prove that B cannot be the cause of A
• Even if two variables are both correlated and temporally ordered, the earlier one does not have
to be the cause of the later one!
• Correlation is a necessary, but not a sufficient precondition for causation!
Variables in experiments
• Independent Variables (IV = OV) = manipulated by experimenter:
o What is a good independent variable? must be manipulated.
o How many levels of the variable?
• Dependent Variables (DV = AV) = measured by experimenter:
o What is a good dependent variable ? Should be closely related to the concept you want to
measure.
o Beware of
Floor effects = effect when a data-gathering instrument has a lower limit to the data
values it can reliably specify (ondergrens voor de data die het betrouwbaar kan
specifieren) (definition not from lecture)
Ceiling effects = a measurement limitation that occurs when the highest possible or
close to that, reached, thereby decreasing the likelihood that the testing instrument has
accurately measured the intended domain. (definition not from lecture)
• Control Variables (controlled by experimenter):
o Holding them constant
o Turning them into independent variables, when independent don’t have an effect
E.g. naming gender in the discussion section
Between-Subjects versus Within-Subjects Designs
• Between-Subjects Designs (Independent Groups):
o Every subject experiences only one level of the independent variable
o Random assignment!
• Within-Subjects Designs (Repeated Measures):
o Every subject experiences every level of the independent variable
o When 4 levels (no division of participants with 2 levels and other with 2 levels bc every level)
o Order effects?
Problems of experimental designs
Particularly Critical in Clinical Psychology:
• Quasi-experiments instead of random assignment.
o Cannot random assignment of people who have disorder and who do not.
• External validity:
o Laboratory versus everyday life
o Patients versus analogue populations
• Low sample size low statistical power
Effect Size and Statistical Power
Effect Size and Statistical Power: Why bother?
• How many participants will I probably need in my study?
• Why do so many experiments in psychology yield non-significant results?
• Why should I better not believe many of the significant results I read about?
, What's It All About? Two Types of Errors
Problems in generalizing from the small experimental sample to the population The errors are
False positive = type 1 error = alpha error = about (not)
o rejects a null hypothesis that is actually true in the population rejecting the Ho
o Alpha is the probability of this error
False negative = type 2 error = beta error =
o not rejecting the null hypothesis when it is actually false in
the population decreases when power increases
o beta is the probability of this error
Definitions of effect size and statistical power
• Effect Size = How large is a difference / correlation / relationship? (cohen’s d is a measure of this)
the degree to which the null hypothesis is believed to be false thus how big is the effect?
• Statistical Power = Probability that this effect will be statistically significant in an experiment.
• Situations:
o Experiment in preparation: Determine necessary sample size
o Experiment completed: Determine power of the experiment
o Evaluation of published studies: Are the effects for real?
Effect sizes: cohen’s d as a simple example How large is d typically in psychology?
• 0.2 = small
• 0.5 = medium
• 0.8 = large
What affects power?
• Effects size:
o Larger effects are easier to find, do
not study an anticipated small effect!
• Sample size:
o Effects are easier to find with many
participants
• Alpha error:
o Increasing the alpha error reduces 530 348 270
the beta error (increasing one error
decreases the other)
• Thus, how larger the effect, how smaller 87 57 44
your sample size of each group you need.
Bigger alfa values are useless. 35 23 18
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper PaulineWolfs. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €7,49. Je zit daarna nergens aan vast.
