Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
35 Cards in this Set
- Front
- Back
Construct Validity
|
the degree to which the IV and DV accurately reflect or measure what they are intended to.
True underlying thing you are trying to measure |
|
Internal Validity
|
whether causal statements about relationships between variables is warranted
If causal variables cannot be determined, this is a confound in internal validity can make casual statement about relationship between variables |
|
External Validity
|
the extent to which one can generalize from the research setting and participant population to other settings and populations.
|
|
Control Variable
|
variable held constant across all levels of the IV
|
|
Control Condition
|
the baseline used that the IV that is being manipulated is compared to
|
|
Correlation
|
relation between X and Y
|
|
Causation
|
three possible relationships
X-Y Y-X Z-Y & Z-X |
|
Validity
|
truth of observations
|
|
Reliability
|
consistency of behavioral measures
If measure is not reliable, then it is not valid |
|
Confounding Varible
|
Something that varies together with the independent variable
|
|
External Validity *
|
Being able to replicate research
|
|
Protocol
|
One way to minimize inadvertent researcher bias
|
|
Construct Validity
|
When measures fit with results
Operationally define all variables will maximize this. |
|
Demand Characteristics (Reactivity)
|
When people do things in an experiment that they wouldn't normally do
|
|
Person Confound
|
Avoided by random assignment
|
|
Between-Subject Design is Better than Within-Subject Design
|
When the independent variable is likely to have a permanent effect
|
|
Internal Validity
|
Validity specifically concerned with being able to make causal statements about relationships between variables
|
|
Completely/Fully Counterbalanced Experimental Design
|
All possible treatment orders are used
|
|
Deception
|
The best way and most common way to reduce the impact of reactivity on the Independent Variable
|
|
Random Assignment (most helpful)
|
Preferred to matching because we may not know all of the relevant variables to match
|
|
Within-Subjects Design
|
When the IV is time
|
|
Double-Blind Experiment
|
Participant reactivity and researcher bias are minimized
|
|
Science
|
Is always changing and evolving
|
|
Faithful Subject
|
Participant who responds honestly on each trial even if he/she know the major hypothesis
|
|
Pilot Research
|
Preliminary research prior to a complete project
|
|
Factorial (In Experimental Design)
|
Testing all combinations of the IVs
|
|
Mundane Realism (Ecological Validity)
|
Is less relevant/important than generalizability
|
|
Latin Square Design
|
All possible treatment orders are used
|
|
Control Group
|
is often untreated
|
|
Within-Subject Design is better than Between-Subject Design
|
Fewer participants are needed
There is less chance of difference between groups Each participant is his/her own control |
|
Parallel Form Reliability
|
Two test coving the same material with different questions but having the same scores obtained on both tests
|
|
Deception (must)
|
Be justified by significant value
Not cause significant physical harm or emotional distress Be explained to the participants as early as possible |
|
Basic Research
|
It may be decades before it produces useful application
|
|
Ceiling Effect
|
If participants do too well on a task to observe meaningful differences in behavior
|
|
Difference Between Control and Experimental Conditions
|
This supports our hypothesis
|