PSYCO104 Lecture Notes - Lecture 3: Repeated Measures Design, Dependent And Independent Variables
PSYCO 104 - Lecture #3
●Cont. of previous lecture
●Research Method
○Correlational Studies
■Examines relationship between/among variables
■Seeks to examine general relationships between variables of interest for
particular research question
■Involves measuring at least two variables (e.g. X and Y), then determining
whether they are statistically related
■Remember that variables in correlational research are only ever
measured, never manipulated
■Advantages
●Gauges the strength of relationship present
●Can be used to make predictions about variables
●Identifies ‘real-world’ associations
●Can be used to study things we couldn’t ethically manipulate in
experiments
■Disadvantages
●While the fact that these kinds of studies take place outside of the
laboratory is a sort of advantage (good ecological validity), it also
means they are uncontrolled
●“Correlation is not causation”: these kinds of studies don’t allow
inferences about cause and effect to be drawn, they merely
demonstrate associations
■Bidirectionality problem:
which variable is affecting which (if either)?
■Third variable problem:
is there another unaccounted for variable
involved in the relationship?
■Correlation Coefficient:
Correlations are mathematically described by a
correlation coefficient – values range from -1.0 to +1.0
●(–) sign indicates direction (positive or negative)
●absolute value indicates strength
○~ 0.1 = small effect
○~ 0.3 = medium effect
○~ 0.5 = large effect effect
●• For positive correlations:
○As X increases, Y also increases
○As X decreases, Y also decreases
●For negative correlations:
○As X increases, Y decreases
○As X decreases, Y increases
■Scatter Plots: these are used to visualize correlation data
○Experimental Methods
■Examines cause and effect relationship
■Three main characteristics of experiments:
●A variable is manipulated (the independent variable, e.g. whether
a participant receives a drug or placebo)
●Another variable is measured to determine what effect, if any,
results from the manipulation (the dependent variable, e.g.
whether a participant’s blood pressure changes after receiving a
treatment)
●Other factors that might influence results are controlled for
■Unlike the other study designs that were discussed, experiments do allow
inferences about cause and effect to be made
■This is due to the combination of directly manipulating the independent
variable(s), while controlling for other factors that may influence the
dependent variable(s)
■The Power of Prediction
yifanyang and 39701 others unlocked
42
PSYCO104 Full Course Notes
Verified Note
42 documents
Document Summary
Seeks to examine general relationships between variables of interest for particular research question. Involves measuring at least two variables (e. g. x and y), then determining whether they are statistically related. Remember that variables in correlational research are only ever measured, never manipulated. Can be used to make predictions about variables. Can be used to study things we couldn"t ethically manipulate in experiments. While the fact that these kinds of studies take place outside of the laboratory is a sort of advantage (good ecological validity), it also means they are uncontrolled. Correlation is not causation : these kinds of studies don"t allow inferences about cause and effect to be drawn, they merely demonstrate associations. Correlations are mathematically described by a correlation coefficient values range from -1. 0 to +1. 0. ( ) sign indicates direction (positive or negative) Scatter plots: these are used to visualize correlation data. A variable is manipulated (the independent variable, e. g. whether a participant receives a drug or placebo)