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;
34 Cards in this Set
- Front
- Back
____________ Validity: Items look as though they measure what they are supposed to measure.
|
Face
|
|
A measure for body image is high in ________ validity because it appears to ask questions pertaining to participants thoughts and opinions of their bodies.
|
Face
|
|
____________ Validity: How much the content of a test represents the domain of interest
|
Content
|
|
A standardized test would be considered high in __________ validity if it covers what was actually taught in the students' classrooms.
|
Content
|
|
____________ Validity: How much the test scores correlate with an external criterion
What are the 2 subtypes? |
Criterion Related
Concurrent; Predictive |
|
____________ Validity: The correlation between test scores and another valid measure given about the same time (also a type of ________________ validity)
|
Concurrent; Criterion Related
|
|
____________ Validity: The correlation between scores and later predicted behavior (also a type of _______________ validity)
|
Predictive; Criterion Related
|
|
A new depression measure is high in ____________ validity when it correlates with the Beck Depression Inventory.
|
Concurrent
|
|
The GRE has high _________ validity if it correlates with graduate school GPA.
|
Predictive
|
|
____________ Validity: How much scores correlate with other scores or attributes that would be predicted by a well-established theory.
What are the 2 subtypes? |
Construct
Convergent, Divergent |
|
____________ Validity: If scores correlate with other measures that they are theoretically expected to (a type of ___________ validity)
|
Convergent; Construct
|
|
____________ Validity: Scores have no relationship with measures of concepts that are theoretically expected to not be related (a type of ___________ validity)
|
Divergent; Construct
|
|
A creativity measure has ________ validity because it correlates with participant art class grades as theoretically expected.
|
Convergent
|
|
A creativity measure correlates with art grades but has no relationship with student math and science grades. This measure has ___________ validity.
|
Divergent
|
|
Describe the main threats to internal validity aka possible rival hypotheses for your results.
SH*T RAM!! |
• Selection bias: groups still differ on some unaccounted for variable that influences results
• History: people may respond differently to an intervention based on personal history • Instrumentation: Low reliability or validity of measures • Testing effects: Repeatedly taking tests can influence reliability of your measurement • Regression to the mean • Attrition: Differential attrition between groups may result in selection bias. Attrition always will lower statistical power. • Maturation: people generally grow and may get better simply with time. |
|
Describe the the main threats to external validity.
|
• Reactivity to the outcome measure
• Pre-test and post-test sensitization • Generalization across outcome measures • Demand characteristics • Carryover effects: the order a treatment is administered may influence your results. For example, a person receives Medication, then Therapy, the No Treatment. Who is to say that observed differences are not due to carryover effects from the earlier treatment? • Treatment interaction effects: selection, history, and testing effects may result in some people disproportionally benefiting from a treatment. For example, some people may be more likely to participate in a study but are they really representative of the population as a whole? • Timing of measurement/novelty effects • Combination of treatment and subject selection, experimental setting, history |
|
Describe what experimental design is, including the strengths and weaknesses of this design.
|
• Random selection from the population and random assignment into groups
• Strengths: high internal validity so causal inferences can be made • Weaknesses: tends to have lower external validity; more difficult/expensive to carry out |
|
Describe what quasi-experimental design is, including the strengths and weaknesses of this design.
|
• Conditions of a true experiment are approximated
• Random assignment of participants into groups is not possible (e.g., you are administering two levels of an intervention in pre-existing classrooms…you could not assign students to classes). • Strengths: approximates experimental research but easier to carry out • Weaknesses: Selection bias may influence results; must be more cautious about causal inferences |
|
Describe what case control design is, including the strengths and weaknesses of this design.
|
• Observational research of the magnitude and type of relations among variables.
• Characteristics of participants are not manipulated directly • Individuals in pre-existing groups are examined with respect to a variable of interest • Typically only have one time point, so you just get a snapshot of preexisting differences • Strengths: efficient use of resources; attrition is not an issue • Weaknesses: causal inferences cannot be made • E.g. Are there differences between Men and Women with respect to X? |
|
___________ ____________ is the kind of reliability that measures the degree to which compared halves of the test measure the same construct
|
Internal Consistency
|
|
Name two statistical procedures you could use to measure the internal consistency of a test.
|
Split-half Reliability & Cronbach's Alpha
|
|
__________ ___________ is the kind of reliability that measures the degree to which scores from the same test correlate with each over over time.
|
Test-retest Reliability
|
|
__________ _________ is the kind of reliability that measures the degree to which scores from alternate forms of a test correlate with one another.
|
Parallel Forms
|
|
__________ _________ is the kind of reliability that measures the degree to which different examiners consistently and accurately score the same test.
|
Inter-rater Reliability
|
|
What are the four main factors that influence the reliability of a measure?
|
Length of test, Accuracy of scoring, Group variability, and Item difficulty
|
|
A test that is ___________ is typically more reliable. Longer or shorter?
|
Longer
|
|
True or False: Heterogeneous groups yield higher reliability on a measure than homogeneous ones.
|
TRUE
|
|
Threat to Internal Validity: An unanticipated event occurs while the experiment is in progress and affects the dependent variable.
What are ways to control for this threat? |
History
Use two groups, measure outcome variable at the same time, and keep the treatment short, or isolate the participants during the treatment. |
|
Threat to Internal Validity: Normal developmental changes in participants between the pretest and posttest that might affect the results.
What are ways to control for this threat? |
Maturation
Random assignment helps ensure that maturation will occur at the same rate across groups. |
|
Threat to Internal Validity: Changes in scores due to the fact that participants who score low on a pretest will score higher on the posttest, and participants who score high on the pretest will score lower on the posttest.
What are ways to control for this threat? |
Regression Toward the Mean
Random Assignment (because this threat often occurs when individuals are chosen BECAUSE they have extreme standing on a variable) |
|
Threat to Internal Validity: The effect of participants dropping out of a study.
What are ways to control for this threat? |
Attrition
Administering a pretest. |
|
Threat to Internal Validity: Changes in scores due to taking the test more than once.
What are ways to control for this threat? |
Testing Effects
Be sure that the pros of administering a pretest outway the possibility of testing effects. |
|
Threat to Internal Validity: Changes in the measuring device (including low reliability or validity) or procedure over the course of a study.
What are ways to control for this threat? |
Instrumentation
Be sure to use measures that have been deemed valid and reliable, keep procedures consistent throughout the study. |
|
Threat to Internal Validity: Differences between groups that exist before implementation of the treatment intere with the validity of the study.
What are ways to control for this threat? |
Selection Bias
Random selection and assignment |