Experiments Flashcards
Lab Experiments
The IV is manipulated by the researcher and the experiment is carried out in a controlled lab setting
Lab Experiment advantages
-More controlled
-Higher internal reliability
-Less extraneous variables
-High construct validity
Lab experiment disadvantages
-May not be natural
-Lack ecological validity
-More effort to conduct
Field experiments
The IV is manipulated by the researcher, but the experiment is carried out in a participants normal surroundings
Field experiment advantages
-Higher ecological validity
-Less effort to conduct
-High construct validity
Field experiment disadvantages
-Harder to standardise
-Harder to get consent
-Low construct validity due to extraneous variables
-Lower internal reliability due to participants having different experiences
Quasi experiments
The IV is naturally occurring and not manipulated by the researcher
Quasi experiment advantages
-High ecological validity
-Helps us study variables we can’t manipulate
-Don’t have to ‘set up’ IV
Quasi experiment disadvantages
-Less opportunity to do experiments
-Difficult to conduct
-Low construct validity
Repeated measures design
Having the same people in each condition
Repeated measures advantages
-Less impacted by participant variables
-Easier to obtain sample as smaller
Repeated measures disadvantages
-Influenced by order effects-need to counter balance
-Reduced construct validity due to demand characteristics
Independent measures design
Using different people in each condition
Independent measures advantages
-No order effects
-Reduced demand characteristics
Independent measures disadvantages
-More impacted by participant variables
-Need a larger sample-more effort
Matched pairs design
Using different people in each condition, but attempting to make the participants as similar as possible by testing the individuals on key characteristics
Matched pairs advantages
-Less impacted by participant variables
-Not influenced by order effects
-Lower chance of demand characteristics
Matched pairs disadvantage
-More effort/time consuming
-Need a larger sample
-Can’t control all extraneous variable
Why must extraneous variables be controlled?
To establish a cause-and-effect relationship
Participant variables
Characteristics of the individual participant which may influence results (eg: age, intelligence, skill)
How can participant variables be controlled?
-Use a repeated measures or matched participants design
-Randomly assign participants to groups if using independent measures so variables are spread evenly across groups
Situational variables
Features of a research situation which may influence a participants behaviour and therefore the result (eg: order effects)
How to control situational variables?
-Having different people in each group-independent measures or matched participants design
-Counter-balance if repeated measures design is used
How to control environmental factors? (temperature, noise, etc)
-Impose controls to ensure as little difference as possible between conditions
Demand characteristics
Cues in an experiment which communicate to participants what is expected of them and may influence their behaviour
How to limit demand characteristics?
Do not tell participants the aim of the study
Single-blind research
Participants do not know the aim of the study
Double-blind research
Neither the researcher or participant know the aim of the study-to remove researcher effects/bias
Alternative hypothesis
A hypothesis that predicts the IV will affect DV
Null hypothesis
Predicts the IV will not effect the DV, and any difference seen will be to chance factors.
Two-tailed alternative hypothesis
Does not indicate the direction the difference the IV affects the DV will be
One-tailed alternative hypothesis
Predicts the IV will have a significant effect on the DV, and indicates in what direction
Target population
The group of people the researcher is interested in studying
Sample
The actual group of participants used in the research
Sampling methods
The ways in which researchers can obtain a sample of people from within the target population to take part in their study.
Self-selecting
When people volunteer to take part in the study after seeing it advertised
Self-selecting advantages
-Gain consent easily
-Removes researcher bias
-Easy to obtain
Self-selecting disadvantages
-Volunteers may not be generalisable (more co-operative)
-Expensive/time consuming
Opportunity
Selecting those most readily available at a given time and place selected by the researcher
Opportunity advantages
-Easy to obtain
-Likely to be more generalisable
-Good way to obtain target population
Opportunity disadvantages
-Potential researcher bias
-May get an unrepresentative sample
Random
Where each member of the target population is selected randomly and has an equal chance of being selected
Random advantages
-No researcher bias
-Equal chance of being selected
-Representative of samples
Random disadvantages
-Those selected may not consent
-May not be generalisable enough
-Outliers may be selected
Snowball
When participants are asked to contact friends and family, who contact their friends and family and so on
Snowball advantages
-Covenient
-Already have consent
-Easy to obtain sample
-Less researcher bias
Snowball disadvantages
-Friends/family may be too similar to make it generalisable
-Message may get distorted
Ideal sample size
20 per condition
Primary data
Data collected directly by the researcher
Secondary data
Analysing data that has already been produced (eg: crime statistics already possessed by police)
Mean advantages
-Involves all the data
Mean disadvantages
-Can only be used for numerical data
-Includes all outliers which may skew result
-Can give decimal figures
Median advantages
-Discounts outliers so not skewed
Median disadvantages
-Can only be used for numerical data
-May not be a specific middle value-decimal
-Doesn’t include all data collected
Mode advantages
-Can be used for quantitative and qualitative data
-Easy to calculate
-Always a whole value
Mode disadvantages
-May be no mode or more than one mode
-Doesn’t include all data points
Quantitative data advantages
-Easier to interpret and display
-Easily make comparisons
-No researcher bias
-Easily establish reliability of results
Quantitative data disadvantages
-Doesn’t show reasoning behind participants actions
-Harder to show anomalies
-Can lack ecological validity
Qualitative data advantages
-Shows reasoning behind participant actions
-More detail
Qualitative data disadvantages
-Harder to interpret and display
-Harder to make comparisons
-Harder to interpret and display
-Can have researcher bias-interpreted in multiple ways
Range advantages
-Quick and easy to calculate
-Shows how consistent data is
Range disadvantages
-Doesn’t show if the spread is even
-Can be skewed by outliers
-Doesn’t include all the data
Variance advantages
-Takes into account all values in the data set
-Less likely to be affected by outliers
Variance disadvantages
-More time consuming to calculate
-Not in the same units as original measure
Standard deviation advantages
-Same units as original measure
-Easy to calculate if variance already done
Standard deviation disadvantages
-Time consuming/difficult to calculate without variance
-Takes into account extreme outliers