Exam 2 Flashcards
Conceptual definition of variable
Variable that is a symbolic representation of an entity or quantity that can assume different values
Operational definition of variable
Description of how a construct will be observed in a particular study
Reliability (general)
The measure of consistency
Validity (general)
The measure of accuracy
Self-report measure
Record responses to a questionnaire or interview, parent/teacher reports if the child is young
Observational measure
Record observable behaviors or traces of behaviors; aggressive behaviors, examining teeth for signs of stress grind, consulting public records of marriages/divorces
Physiological measure
Record biological data; muscle movements when smiling/frowning, doing tasks while in an fMRI machine, skin conductance studies for stress
Categorical (nominal) scale
Data that does not automatically have a number; favorite ice cream flavor, breed of dog, etc.
Ordinal scale
Represents a ranked order or variable levels; top 100 _____
Interval scale
Represents a ranked order of variable levels in which there are equal intervals between levels and no “true zero” exists; IQ scores
Ratio scale
Represents a ranked order of variable levels in which they are equal; number of cookies a child eats in a time period
Test-retest
Measures how consistent people’s scores remain overtime when a test is repeated at least once across a particular time interval
Interrater
Measures how consistent ratings are between different observers or coders; most often used when evaluating observational data
Bystander effect studies
We can train people to code “helping” behavior that they observe, such as stopping to ask if someone is okay, helping to change a flat tire, etc.
When evaluating categorical data, which is better, kappa or Pearson correlation
Kappa because it measures the extent to white raters tended to place data points into the same category
Internal/internal consistency
Measures how consistently multiple items on a test “hang together” or effectively assess the same construct; self-esteem questionnaires
Two ways to measure internal/internal consistency
Average inter-item correlation (AIC) and Cronbach’s alpha (coefficient alpha)
Construct validity
How well has the conceptual variable been operationalized? How well did the researchers measure the variable?
External validity
How well the study’s results generalize to people or context beyond those used in the current study, how well does the participant sample represent the general population?
Statistical validity
The extent to which a study’s statistical conclusions are precise, reasonable, and replicable, how well do the numbers support the claim?
Internal validity
Can we eliminate alternative explanations for this perceived relationship?
Face validity
Assesses the extent to which a measure is subjectively considered to be a plausible operationalization of the construct, do they seem to be about what I think they’re about?
Content validity
Assesses the extent to which a measure adequately captures the major aspects of a construct
Criterion validity
Assesses whether performance on the measure is associated with a concrete behavioral outcome that theoretically ties to the construct
Convergent/Discriminant validity
Assesses the extent to which scores on the measure correlated with similar (convergent) or dissimilar (divergent) constructs
What is the relationship between reliability and validity?
You can have good reliability without good validity (being reliably wrong) but you CAN NOT have good validity without good reliability.
Purposive nonrandom sampling
Limiting your sample to only certain groups of people because this aligns with the purpose of your study.
Snowball nonrandom sampling
Study participants are asked to recommend other people who might wish to participate in the study
Quota nonrandom sampling
Researcher identifies subsets of the population of interest and sets a target number of participants from each group. The researcher samples from each group until the quota is reached.
