Terms Flashcards
Secondary Data
Info collected for purpose other than the current one.
Extraneous Variables
Variable that may effect the DV, other than the IV.
Deception
Participants not told the true aim of the study.
Participant Observation
Observer joins in the activity.
Field Experiments
Take place in a natural environment.
+ High ecological validity.
- Difficult to control extraneous variables.
Primary Data
Data collected directly from first-hand experience.
Snowball Sampling
Participants recruit others from people they know.
+ Locate people that are difficult to access.
- Not likely to be good cross-section from population.
Null Hypothesis
Statement of no effect; the variables have no relationship.
Confidentiality
Concerns communication of personal information and that it will be well protected.
Independent Groups Design
Different participants in each condition - performance compared.
+ External variables can be controlled.
- Low ecological validity.
- Participant variables can’t be controlled.
- Need more participants to get the same amount of data.
Deal with limitations:
Randomly allocate participants to conditions.
Dependent Variable
Measured (usually behaviour)
Covert Observation
Participants don’t know they are being observed.
Quantitative
Quantity, numbers, easily measured, easily compared.
Mean
Add up scores and divide by number of data.
Sensitive - uses all data.
Can be distorted by anomalous data.
Median
Middle score when data is in order.
Not easily distorted.
Not as sensitive.
Mode
Most frequent.
Easy to find.
Very insensitive.
Useless.
Nominal Data
Data in separate categories.
Ordinal Data
Data ordered (e.g. ascending order).
Interval Data
Data measured using equal intervals.
Ratio
True zero point as in most measures of physical quantities.
Qualitative
Quality, description, observed not measured.
Directional Hypothesis
States direction of predicted differences.
Self-Selected Sampling
Advert - volunteers.
+ Variety - representative, less bias.
- Volunteer bias - highly motivated to be helpful.
Opportunity Sampling
People that are easily accessible.
+ Quick and easy.
- Researcher bias - similar people - difficult to generalizable.
Time Sampling
Counting behaviour in a given time frame.
Matched Pairs Design
Different people perform in each condition but are matched in ways that matter to the study.
+ External variables can be controlled.
- Time consuming.
- Difficult to match participants characteristics.
Privacy
Person’s right to control the flow of information about themselves.
Operationalise
Ensure variables are in a testable form and can be measured.
Non-Participant Observation
Observer separate from those being observed.
Overt Observation
Participants know they are being observed.
Observer Bias
Observers expectations effect what they see/hear.
Reduces validity.
Stratified (Quota) Sampling
Randomly selected from sub-groups in population.
+ Representative.
- Difficult to identify sub-groups - time consuming.
Event Sampling
Counting number of time a behaviour appears in a target population.
Aim
Statement of what the researcher intends to find out.
Content Analysis
Observational study in which behaviour is observed indirectly through written material.
+ Appropriate for any form of media.
+ Usually no ethical issues.
+ High ecological validity.
- Behaviour studied out of context.
- Categories chosen beforehand, based on research expectations.
- Possibly subjective - different observers may interpret categories differently.
Demand Characteristics
Cue that makes the participants aware of the aim of the study.
Independent Variable
Manipulated to test the effect on the DV.
Valid Consent
Participants given comprehensive info on the study so they can make an informed decision on their participation.
Non-Directional Hypothesis
States there will be a relation between variables, but not the direction.
Lab Experiments
Takes place in controlled environment.
+ External variables can be controlled - participants have the same experience.
- Low ecological validity.
Random Sampling
Equal chance of being chosen.
+ No researcher bias.
- May not be representative.
- Time consuming.
Online Experiments
Assess participants via internet/social media. \+ Can assess large groups. \+ Diverse sample. \+ Cost effective. \+ Quicker analysis. - Methods used are limited. - Ethical issues hard to deal with.
Risk of Harm
Participants shouldn’t experience negative physical or psychological effects, beyond what is normal.
Repeated Measures Design
Same participants in conditions - performance compared.
+ External variables can be controlled.
- Order effect - repetition may affect result (boredom effect, practice effect and guessing).
Deal with limitations
Use different tests to reduce practice effect - counterbalancing.
Systematic Sampling
Every Nth number of the target population.
+ No researcher biases.
- May not be representative.
