Chapter 8 Flashcards
How are all other variables controlled in experimental method
Experiment control
Randomization
Confounding variable
Variable that varies along with the independent variable
It occurs when efffects of independent variable and an uncontrolled variable are intertwined so you cannot determine which of the variables is responsible for the observed effect
When the results of an experiment can confidently be attributed to the effect of the independent variable, the experiment is said to have…
Internal validity
How does one achieve good internal validity?
Researcher myst design and conduct the experiment so that only the independent variable can be the cause of the results
A researcher using a Posttest only design must..:
Obtain two equivalent groups of participants
Introduce the independent variable
Measure the effect of the independent variable on the dependent variable
Why must pisttest only design use two EQUIVALENT groups?
To eliminate any selectional differences
Selection differences
Differences in the type of subjects who make up each group in an experimental design
This situation occurs when participants elect which group they are to be assigned to
Difference between pretext only design and pretext post test
In protest post test, a pretext is given before the experimental manipulation is introduced
Why in pretext posttest, is pretext given before manipulation
Makes it possible to ascertain that groups were infact equivalent from the beginning
It is not always needed in a large sample of participants that has random assignment
drop out factor is caused
Attrition or mortality
Fallbacks of a pretest
It can sensitize participants to what you are studying, enabling them to figure out what you are studying and why
Solomon design
Half of participants receive only the post test and half receive both the pre test and the post test
Independent group design (also known as between subjects design)
When participants are randomly assigned to the various conditions so that the each participated in only ONE group
Comparisons are made between different groups of participants
Repeated measure design
Participants receive all conditions
When each participant is measured after receiving each level of the independent variable.
(Also called within subject design)
Random assignment
Decision to assign an individual to a particular condition is completely random and beyond the control of the researcher
Advantages of repeated measure design
Fewer research participants are needed because each individual participated in all conditions
They are extremely sensitive to finding statistical significant differences between group.
Two ways to assign participants to a group
Indenpendent group design (only participated in one )
repeated measure design (participated in all)
Advantages of pretest post test
Ensure consistency (aka experimental control)
Can help with subject selection criteria
Can determine change between pretest and post test
Can determine if attrition (aka mortality) affected results
Pre test post test disadvantages
Demand characteristics (like filler ideas)
Time, money, effort, etc
How to ensure equivalency between two groups
Random assignment
Repeated measures design (ak within subjects) advantages
Fewer Ss needegreater statistical “sensitivity” by deducting random error
Disadvantages of experimental design
Order effects
Counterbalancing
Threats to internal validity highlighted by quasi experiment designs
History Maturation Testing Instrument decay Progression toward the mean Attrition or drop out
Order effect
The order of presenting the treatments effect the dependent variable
Order effect and repeated measure design
A problem with repeated measure design stems from the fact that different conditions must be presented in a particular sequence
Greater recall in the high meaningful condition can be attributed to the fact that the high meaningful task came second
Practice effect
Performance in the second task might improve merely because of practice gained on the first task
Fatigue effect
When research participant becomes tired, bored or distracted
Disadvantages of repeated measure design
Order effect
Practice effect
Fatigue effect
Carryover effect
Carryover effect
When effect of first treatment carries over to influence the response of the second treatment
Ex-after reading about less sever crime, the more severe crime might seem worse to participants than it normally would
Counterbalancing
When all possible orders of presentation is included in the experiment
This helps with order effect
How to deal with order effect
Counterbalancing or Latin square
Time interval between treatments
Latin actuate
A technique to control for order effects without having all possible orders
Each condition appears at each ordinal position
Each condition prefers and follows each condition one time
How to counteract fatigue effect
A rest period/time in between treatments
Repeated measure design over independent group advantages
A reduction in the number of participants is required to complete the experiment
Greater control over participant differences and thus greater ability to detect an effect of tha independent variable
When can you not use repeated measure design
An experimental procedure that produces a realistically permanent change in an individual
Ex-psychotherapy treatment or surgical procedure
Matched pair designs
Instead of randomly assigning participants to groups, people get matched on participant variable such as age or personality trait
The variable will be strongly related to the dependent variable
What does a matched pair design ensure
That groups are equivalent prior to introduction of independent variable
When is matching most likely used
When there’s only a few participants available or there is a strong relationship between dependent measure and the matching variable
