Please enable JavaScript.
Coggle requires JavaScript to display documents.
Research (Quantitative) Methods, longitudinal research - Coggle Diagram
Research (Quantitative) Methods
Qualitative (descriptive/interpretive, naturalistic, external validity)
Naturalistic Observations
Clinical Case Studies
Interview
Quantitative (statistical, experimental, objective, internal validity)
Hypothesis
a stement that is testable and falsifiable
null hypothesis assumes that there be no difference in observations
alternative hypothesis assumes that there will be a difference in observations
Variables (1)
anything that can take a variying value
not very relevant to psychlogy, as it is concerned with "internal" characteristics
Variables (2)
Independent Variable
: the one the researches manipulates
Dependent Variable
: the one that changes as the I.V. changes
X (I.V.) causes Y (D.V) --> only in experimental design
Controls
: when the researches manpulates the I.V., all other possible variables stay the same
Confounding Variables (3)
variables that can potentially distort the relationship between the I.V. and D.V.
CVs are better controlled in a "
true labratory
" experiment -->
high Internal Validity
(how well confounding variables are controlled) but
low Ecological Validity
(relation to reality)
Population Validity
: generalizability / ecological validity
Operationalization
e.g. Depression is a theoretical construct (cannot be observed directly)
in order to be directly observed, it needs to be quantifiable
Correlational Studies
No variables manipulated by researcher
No indepedent and dependent variables --> there are two co-variables measured
We cannot say that x causes y (no causation) --> we can only say that there is a link between x and y
https://docs.google.com/document/d/1IAFkCVlHrio_l2Xb2zzjdxAWkU00hDhMlJT9_srp4GM/edit
Coefficient of Correlation (r)
Numerical indication of
magnitude
and
direction
of the relationship between two variables:
Negative Correlation
Positive Correlation
If the correlation is 0 --> no link between co-variables
0.10 - 0.29 --> small correlation
0.30 - 0.49 --> medium correlation
0.50 - 0.100 --> large correlation
Experimental Design
Sampling Techniques
Representativeness
: when the sample reflects all essential characteristics of a target population
Random Sampling
: every member of the target population has an equal chance of becoming part of the sample
Statiffied Sampling
: 1. decide the essential characteristics that the sample must reflect / 2. study the distribution of these in the target population / 3. recruit those who reflect the characteristics
Convenience Sampling
: recruiting easily available participants
Self - Selection
: recruiting volunteers after advertising
Quasi - Experiment
participants are randomly assigned to a condition on the independent variable
participants are grouped based on a characteristic of interest, such as gender, ethnicity, age or scores on a scale
Natural Experiment
naturally ocrruring variables
Bias
Demand Characteristics
: participants acting a certain way because they are aware of them participating
Researcher Bias
: unintentional error in design, sampling, analysis, interpretation or dissemination of findings
Maturation
: changes in the participants caused by time passing by
History
: outside events that happen to participants during experimentation
Experimental mortality:
participants dropping out
Testing Effect
: participants becoming acquianted with the task
Social Desirability
: giving responses that are social accepted instead of what one truly believes
Surveys
Repeated Measures Design
(same participants take part in each condition of the independent variable) =/=
Independent Measures Design
(different participants are used in each condition of the independent variable(
Terms
Sample: participants in the study
Sampling: the process to recruit participants
Credibility: how much results affect reality
Bias: making unintentional errors in research design, process or interpretation of data
Generalizability: applying results beyond a specific sample
Statistical Significance (P value)
the likelihood that the correlation was obtained by chance
the smaller the p value, the more statistically significant my findings are
more than 5% - p not significant - result is non-significant
less than 5% - p<0.05 - result statistically significant
less than 1% - p<0.01 - result very significant
less than 0.1% - p<0.001 - result highly significant
Ethics
right to withdraw: the ability to quit an experiment
no harm (physical or mental)
debriefing: being notified on deception and study results
informed consent - voluntary
anonymity & confidentiality
dual relationships: can't do research on people one knows - participation won't be voluntary
IRB
: Insitutional Review Board --> approval for research ethics
Dissemination
: publication of findings or research
Ethics in Research Dissemination
Confidentiality in handling personal/sensitive data
Data fabrication
Plaigarism / Publication credit
Sharing resereach data for verification (transparecy)
Social implications of the findings (+where results are published)
longitudinal research
mortality