what is correlational design ?
what does it explore?
show?
Correlational Design
* Exploring the strength and
direction of a relationship
between two variables
* Show how one thing changes
as another changes
* Observe ‘natural’ events;
snapshot of ‘how things are’
* No cause and effect
what does experimental design compare?
Experimental Design
* Comparing 2 conditions (within-
subjects design) or 2 groups
(between-subjects design)
* Controlled comparison of
situations; can identify cause and
effect
* Can establish cause and effect
which design can establish cause and effect and which cannot ?
correlational cannot
experimental can
examples of correlational design
+
which is for parametric and non-parametric data
Pearson correlation for
parametric data
* Spearman correlation for
non-parametric data
regression
examples of experimental design
+
which is for parametric and non-parametric data
Independent samples t-test
for between-subjects
(independent) design
* Paired samples t-test for
within-group (repeated) design
- Independent measures designs
- Repeated measures designs
- Mixed & Factorial designs (2+ IVs)
what are types of regression
simple linear regression
multiple regression
what the difference between independent measures design, repeated measures design, mixd and factorial designs
Independent measures designs
(2+ groups, single IV)
* Repeated measures designs
(2+condition, single IVs)
* Mixed & Factorial designs (2+ IVs)
more than one factorial interest
what are correlational designs used for
measure strength and direction of relationship between two variables
wha are regression designs used for
use one (or more) predictor variables to explain variability on an outcome varaiable
is the predictor good at explaining the varibaility of an outcome
what is meant by simple linear regression
does one thing predict another?
what is meant by multiple regression?
can i use multiple predictors to predict one thing
more than one predictor
what an experimental method
compare two (or more) controlled situations
in one the cause is absent in the other the cause is present
designs with single iv
repeated measures
independent measures
designs with one IV with two or more conditions/groups
repeated measures
independent measures
designs with more than one iv
factorial designs
mixed designs
what is meant by factorial design
the effect of one variable is interacting with the other
manipulating more than one iv in a experiment
when a factorial design typically used
when we think our two ivs may interact wit one another
what is meant by interaction
we meant the effect of one iv is different depending on the levels of the other iv
whats a measurement error
when we measure things there is always a degree of error
psychological variables are subject to different amounts or error because they are typically influenced by a rnage of factors
instead of iv and dv what do we use in correlarional research
predictor and outcome
systematic variation
variability in scores between groups/conditions brough abaout as a result of the experimental manipulation
desirable
unsystematic variation
variabilityin scores between groups/conditions that is due to random and/or uncontrolled factors
undesirable
randomisation
keeps unsytematic variaion as low as possible in order to maximise systematic variation
ways of randomisation
experimental control
randomisation
counter balancing
