Week 8 (variables, experimental research & non-experimental) Flashcards
What are the 3 classes of variables?
Behavioural
Stimulus
Subject
What is a behavioural variable?
It is any observable response produced by a subject
Eg. Often dependent variables the organism produces such as heart rate
What is a stimulus variable?
Environmental factors that have actual or potential effects on behavioural variables
Normally independent variables that we try to control
What are subject variables?
Characteristics that can be used to classify a subject for research purposes
Aren’t possible to control normally
Eg. Gender, age, mental health, medical condition
What is the dependent variable?
A variable that is measured or observed
How is the DV measured
?
Can be measured directly
Eg physical arousal, virus antibodies
Sometimes cannot be measured directly and is represented by a contrast
Eg pleasure, love, pain
What is the independent variable?
A variable that is manipulated by the researcher in true experiments
Requires at least 2 levels or conditions
When can the IV not be manipulated?
Sometimes it is impossible - eg sex
Sometimes it is unethical - eg smoking, drugs
What is an experiment called when they have more than one independent variable?
Factorial experiments
What is an operational definition?
Procedures or operations that specify how to manipulate or measure a construct
What are the benefits of multiple operational definitions?
Different aspects of the construct can be measured
Convergent validity
Allows for multiple ways to design studies
What are extraneous variables?
Variables other than the IV that affects the DV
Contribute to random (unexplained) error (variability) to the data
Systematically occur
What is a control variable?
Extraneous variables that are held constant to avoid confounding
What are the goals of experimental designs (3)?
Maximise (hypothesised) causal effect
Minimise influence of extraneous variables
Control confounding variables
What do experiments attempt to do in relation to between and in group designs?
Maximise between groups variance
Minimise within groups variance
Explain between groups variance?
Caused by an IV and is systematic
But can be affected by confounding variables that may act to increase difference between groups so that the IV appears more important than it is
Decrease differences between groups so that the IV appears less important than it is, masking IV effects
Explain within groups variance
Variance typically caused by unknown factor - random (error) variance
Extraneous variables, individual differences and measurement error (unreliable instruments)
Explain quasi-experimental designs
Inability to manipulate the IV
Unable to control all confounding variables
Can have control variables
Same statistical methods as experimental design
Explain the process of a single group, post test only (one shot case study)?
Identify a sample
Introduce IV
Measure DV after intervention
Compare sample with others that haven’t had IV
No control group - poor quality study
Why is a single group, post test only (one shot case study) under threat to its internal validity?
Bias (sampling method) Maturation History Regression to the mean Placebo effects
Explain the process of a single-group, protest-posttest experiment
Identify a sample
Measure DV before intervention (pre test)
Apply IV
Measure DV after intervention (post test)
Compare pretest and post test
Slightly stronger than single group post test only due to measurements prior to
What are the threats to the internal validity in single-group, pretest-post test?
Bias (sampling) Maturation History Regression to the mean Pre-test sensitisation Placebo effects
Why should we use control groups and random assignment?
Control group reduces threats of history, maturation and regression to the mean
Random assignment reduces selection bias
What is a two-group posttest-only, control group design?
Simplest form of between subjects design
Is a true experiment because of control for confounding variables and random assignment
Good to use when pre-testing isn’t possible
Process:
Sample split into control and experimental group
No treatment or apply IV
Measure DV after intervention and compare
Explain 2group, pretest-post test
Simplest form of mixed design
True experimental design: random assignment
Controls maturation and history
Process:
Random allocation into control or experimental groups
Measure DV
Apply IV or not
Measure DV after
Calculate the effect by comparing post and pre tests
What are the benefits of within subjects design?
Every participant is involved in every condition
Maximises group equivalence - perfect March when compared with self
Allows estimation of variance due to individual differences
Removes noise
What are the issues with within-subject designs?
Order effects
Carry over effects
Sequence effects - condition order matters
How do we minimise order effects in within subjects design?
Randomise condition order for each participant
For small number of conditions - complete counterbalancing eg. ABC ACB BAC BCA CAB CBA
partial counterbalancing - Latin square: each condition occurs one in each possible position
For large sample size!!
What is a main effect?
Term only used in factorial designs
Effect of one factor, ignoring the other.
One main effect for each factor in the design
What is an interaction effect?
Effect in which the effect of one factor depends on the levels of another factor
Factorial designs and chi square analysis?
Chi square test of independence tests an interaction - will be significant if there is a large enough interaction
Does not test main effect - if there is a main effect but no interaction, test will not be significant
How many operational definitions can a variable have?
A variable can have multiple operational definitions