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;
51 Cards in this Set
- Front
- Back
Reliability
|
refers to how consistent your measurement is
|
|
3 kinds of consistency and reliability in psychological measurement
|
1. internal consistency
2. temporal consistency 3. procedural consistency |
|
internal consistency
|
do the items of my measure consistently measure the same construct?
|
|
temporal consistency
|
how consistent is my measure over time? Test-retest reliability
|
|
procedural consistency
|
how consistent is my measure across different administrators? Interobserver agreement, interrator reliability
|
|
validity of a pscyhological measurement tool
|
refers to the extent to which we are measuring what we intend to measure
|
|
How does reliability relate to validity?
|
a measurement must be reliable in order to be valid
|
|
Types of measurement validity
|
1. face validity
2. content validity 3. predictive validity 4. concurrent validity |
|
face validity
|
it looks as if the test measures what it's supposed to measure
|
|
content validity
|
the content of teh test is representative of the content domain that I intend to measure
|
|
Predictive validity
|
My test specifically predicts some specific criteria
|
|
concurrent validity
|
my test relates in the appropriate way to some concurrent criterion
|
|
brilliant studies simply become bad studies without
|
good measurement instruments
|
|
What does validity refer to in psychology?
|
validity refers to how accurate or correct a psychological proposition (or measure) is.
|
|
in teh context of psychological studies validity usually refers to
|
how accurate (valid) the conclusions of the researchers are, given their research design and their results
|
|
internal validity
|
the extent to which we can be confident that the change of the IV resulted in teh changes of the DV.
|
|
Thus, if a study is high in internal validity, we can confidently conclude that in our study
|
the changes of our IV were causing the changes to the DV
|
|
3 requirements for confirming causation
|
1. covariation
2. temporal sequence 3. eliminating confounds |
|
external validity
|
refers to the extent to which a research finding is actually representative of what typically happens in the real world.
|
|
if a study is high in external validity, then we can confidently conclude
|
that the findings apply to situations outside of the research environment
|
|
external validity is closely associated with the issue of boundary conditions
|
Generalizability
|
|
2 basic issues of external validity
|
1. to what extent will my results generalize to other people
2. how much will my results generalize to other situations |
|
confounds threaten
|
internal validity
|
|
artifacts threaten
|
external validity
|
|
construct validity
|
the extent to which the independent and dependent variable actually represent the abstract theoretical variables that the researcher is interested in
|
|
basic issues
|
am i really manipulating the variables that i hoped to manipulate?
am i really measuring the variables i hoped to measure? |
|
conceptual validity
|
how well your specific research hypothesis follows from teh broader theory it was designed to test.
|
|
threats to the internal validity of psychological studies
|
1. history
2. maturation 3. selection |
|
history
|
some event takes place between the testing of the different levels of your IV
|
|
maturation
|
participants become older (more mature), wiser, stronger, or more experienced between the testing of teh different levels of your IV.
|
|
selection
|
participants are not randomly assigned to the different conditions of your IV
|
|
statistical regression
|
experimenters choose participants on the basis of their having scored very high or low on a test
|
|
differential mortality
|
people dropping out of study
|
|
demand characteristics
|
characteristics of the experimental procedure that subtly (or not so subtly) suggest how the participants are expected to act.
|
|
participant reactance
|
the research participant behaves so as to disconfirm the experimenters expectations
|
|
evaluation apprehension
|
a participants concerns about being judged either favorably or unfavorably by another person
|
|
some ways to counter participant reaction bias:
|
1.anonymous participation
2. cover story 3. secret or unobtrusive observations 4. the "bogus pipeline" 5. indirect or implicit assessment |
|
how can you control for confounds
|
remember there is no such thing as a perfect study
be careful and be prepared! use apporopriate control groups use random assignment if possible. measure and control for confounds. |
|
how can you determine if you have significant artifacts
|
-remember that it's virtually impossible to eliminate all artifacts
-administer your study with different populations |
|
artifacts are very difficult for us to recognize
|
becase we aren't aware of the water that we swim in
|
|
operational definitions
|
to specify the operations anyone must go through to set up the independent variable in the same way they did.
|
|
pilot experiment
|
a small scale version of the experiement you are planning
|
|
3 ways to determine reliability
|
test retest
alternative form split half |
|
alternative form
|
a second test having items similar to the first is constructed and given to the same peeps
|
|
split half
|
a single test is statistically split into halves and scores of the two halves are correlated
|
|
concurrent validity
|
tow measures taken at th same time to compare test score with criterion
|
|
choice reaction time
|
time it itakes to give one of several responses when one of several stimuli occur
|
|
composite dependent variable
|
gives some indication of overall performance
|
|
percent savings
|
percent saved=number of trials to learn - number of trials to relearn/ number of trials to learn x 100
|
|
physiological measures
|
behavior is a private event but the physiology of the body can be measured
|
|
dual task methodology
|
the primary task and a secondary task is performed
|