Experimental Design Flashcards
what is the experimental approach?
The researcher manipulates one or more variables and controls and measures any change in other variables.
what are the variables in experimental research?
independent variabel (IV)
dependent variable (DV)
extraneous variable (EV)
what is the independent variable?
The variable that is manipulated
what is the dependent variable?
The variable that is measured
what is the extraneous variable?
A variable that might have an influence on the DV but is not in the focus of interest
what is a confound?
A confound is any uncontrolled extraneous variable that co-varies with the independent variable and could provide an alternative explanation of the results
what are the types of independent variables?
situational variables
task variables
instructional variables
what are situational variables?
Features in the environment that participants may encounter (e.g. number of bystanders in an experiment on helping)
what are task variables?
Vary the type of tasks performed by participants (e.g. different problems in a problem solving experiment)
what are instructional variables?
Vary the type of instruction participants get (e.g. perform a memory task by trying to form a mental image of the item or not)
what are subject variables?
Independent variables that are not directly manipulated by the experimenter such as natural groups (e.g. age groups, gender) or participant variables (e.g. political attitude, religious beliefs).
what are the problems with subject variables?
- cannot be independently manipulated
- therefore causal inferences are problematic
- differences in the DV between groups can be caused by any other factor on which the groups differ.
which groups are control group experiments made of?
- Experimental group: Gets a specific treatment or manipulation.
- Control group: Does not get an effective treatment or manipulation.
what are problems with the dependent variable?
- ceiling effects
- floor effects
what are ceiling effects?
The average scores for the DV are so high that no difference can be determined between conditions.
what are floor effects?
The average scores for the DV are so low that no difference can be determined between conditions.
what is the regression to the mean?
There is a statistical tendency for people who score extremely low or high on a test to score closer to the middle the next time.
what can be done to control extraneous variables? (5)
- Keeping the variable constant across experimental groups (e.g. only include people with specific face recognition abilities)
- Build the EV into design to investigate it (e.g. including face recognition abilities as a factor in the design)
- Matching the groups regarding the EV
- Random assignment to experimental groups
- Statistically controlling the variable (e.g. measuring the extraneous variable and statistically considering the influence)
what are the forms of validity in experimental research?
- external validity
- internal validity
what is the external validity?
The extent to which one can generalize the findings of a study to other situations, people, settings and measures.
what are the main criticisms of external validity?
- participants
→ Most studies in cognitive and social psychology are carried out with participants from WEIRD (Western, Educated, Industrialized, Rich, and Democratic) societies. - Experimental setting
→ Experimental settings are artificial and completely detached from real life. They lack ecological validity.
what is internal validity?
Internal validity is the degree of confidence that the causal relationship you are testing is not influenced by other factors or variables
threats to internal validity (studies extending over time)
- History: An unrelated event influences the outcomes
- Maturation: The outcomes of the study vary as a natural result of time.
- Instrumentation: Different measures are used in pre-test and post-test phases
- Testing: The pre-test influences the outcome of the post-test.
threats to internal validity (participant problems)
- Selection bias: Groups are not comparable at the beginning of the study
- Attrition: Dropouts from participants
what is the trade-off between internal and external validity?
A causal relationship can be tested in an artificial lab setting or in the ‘real world’. A lab setting ensures higher internal validity because external influences can be
minimized. However, the external validity diminishes because a lab environment is different than the ‘outside world’ (that does have external influencing factors).
what are between subject designs?
Groups of participants are assigned to the experimental conditions
what are the advantages of between-subject designs?
- Avoid carry-over effects
- Require less testing time because each participant is only exposed to one
experimental condition/treatment
what are the disadvantages of between-subject designs?
- Require more participants
- Are sensitive to sampling biases
Which analysis can be used?
Randomized Treatment and Control (2 groups) with Posttest-Only Design (between-participant design)
independent sample t-test
which analysis can be used to compare means between experimental
and control group conditions for significant differences?
independent sample t-test
which analysis can be used?
Randomized Multiple Treatments and Control with Post-test-Only Design
one-way anova
which analysis can be used to compare means across different experimental conditions for significant differences?
one-way anova
which analysis can be used? Randomized Multiple Treatments and Control with Pre-test and Post-test Design.
3x2 repeated-measures ANOVA
which analysis can be used to compare means across different experimental conditions for significant differences?
3x2 repeated-measures ANOVA
what are quasi-experimental designs?
Non-random assignment to conditions.
what is the ethical aspect of quasi-experimental designs?
For ethical reasons it is often not possible to experimentally
manipulate specific factors.
which analysis can be used? Simple ex-post-facto design
independent sample t-test
which analysis can be used to compare means between experimental and control group
conditions for significant differences?
independent sample t-test
what are within-subject designs?
Each participant is exposed to each level of the independent variable.
what are advantages of within-subject designs?
Within-subject designs reduce error related to individual differences.
what are disadvantages of within-subject designs?
- Major problem are carry-over effects from one condition to the other.
- Require more testing time for each participant
what are carry-over effects?
Performance in one condition might influence performance in the other condition
(e.g. if participants have to do an implicit and explicit task on stimuli)
what is counterbalancing?
Manipulating the sequence of conditions.
what is across subject counterbalancing?
The order of conditions is manipulated across participants.
Participant 1: A then B
Participant 2: B then A
what is reverse counterbalancing?
Each participant gets both orders of conditions in a reversed way.
Participant 1: ABBA
what is experimenter bias?
Any systematic errors in the research process or the interpretation of its results that are attributable to a
researcher’s behavior, preconceived beliefs, expectancies, or desires about results.
what are measures against experimenter bias?
- Trying to standardize the experiment as much as possible
- Double blind experiments: the experimenter doesn‘t know the assignment of participants to conditions
what is participant bias?
The influence that research participants’ knowledge about aspects of the research has on their responses to
experimental conditions and manipulations
what is the Hawthorne effect?
The effect on the behavior of individuals of knowing that they are being observed or are taking part in research.
- Evaluation apprehension (participants want to be positively evaluated)
what is the demand effect?
Effects induced by demand characteristics (cues that might indicate the research objective) to participants. These cues can lead participants to change their behaviors or responses based on what they think the research is about.
what are ways to deal with demand effects?
- Avoiding any demand characteristics: One should avoid cues that indicate the aim of the study.
- Using placebo groups: Participants think they are in the treatment group but in fact they are not.
- Deception: Giving participants false information about the purpose of the experiment. -> Deception is ethically problematic
- Manipulation check: Asking participants what they think the experiment was about.