Week 5 Flashcards
Reliability
The extent to which a measure is free from random error - How much our measure reproduces the same score repeatedly.
Test Retest Reliability
Administer same questionnaire with same people at multiple times.
Internal Consistency
Do the items on the scale produce similar results?
Inter-rater Reliability
When judging behaviour, do two people judge it the same?
Validity
The extent to which something is “true”; that what one says is accurate.
Construct Validity
The extent to which the measured variable actually measures the construct it intends to.
Face Validity
Measure appears to reflect the construct being measured.
Content Validity
Measure captures all aspects of the domain.
Criterion Validity
Ex: Test anxiety should be negatively correlated with performance on important exams.
A category of construct validity regarding people’s scores on a measure correlating with other variables typically related to the construct.
Concurrent Validity
Extent to which the measure relates to a behaviour at the same time, or even before the measure. Ex: politicians or librarians and extroversion.
Predictive Validity
Extent to which measure predicts relevent behaviours.
Convergent Validity
How well measures of the same thing produce the same results.
Ex: Using a homemade extroversion measure, and comparing results to a well-established extroversion measure.
Discriminant Validity
Extent to which our measure does not relate to other measures.
Free Format/Open Ended
Provide own answer
Fixed Format/Closed Ended
Choose from a list of alternatives.
Response Set
A pattern of responding to questions on a self-report measure that’s not related to question content.
Yay Saying
Agreeing with everything.
Nay Saying
Disagreeing with everything.
Reverse Score
To deal with response sets, questions that are worded differently and reverse the meaning of the order of response.
Social Desirability
Responding in a way that is socially acceptable rather than saying what you actually think.
BRUSO
Brief, Relevant, Unambiguous, Specific, Objective.
Double-Barreled Questions
A question that is asking two different things at once.
Loaded Question
Questions that show bias towards one opinion through strong language.
External Validity
How well the results can be generalized to the intended population.
Stratified Random Sampling
Population divided into different subgroups, or strata, then random sample taken from each stratum.
Convenience Sampling
Selecting people because they’re convenient to recruit.
Sampling Frame
List of all members of the population eligible to be in your study.
Sampling Bias
When a sample is not representative of the population.
Nonresponse Bias
When the people who do not respond differ in some ways than those who do.
Constructs
Traits that are covert or difficult to measure.
Self-Report Measures
Those in which participants report on their own thoughts, feelings, and actions.
Behavioural Measures
Those in which some other aspect of participants behaviour is observed and recorded.
Physiological Measures
Those that invoke recording any of a wide variety of physiological processes.
Converging Operations
When psychologists use multiple operational definitions of the same construct.
Nominal
Categorical variables that involve assigning scores that are category labels without ordering.
Ordinal
Assigning scores so that they represent the rank order of individuals without assuming gaps between intervals.
Ratio
Assigning scores in such a way there is a true zero point that represents the complete absence of the quantity. Also shares features of the other three levels of measurements.
Interval Scales
Assigning scores using numerical scales in which intervals have the same interpretation throughout without having a true zero.
Survey Research
Quantitative and qualitative method of research that involves measuring variables using self-reports and has a large attention on issue of sampling.
Respondents
Participants of survey research.
When is open ended preferred
When answer to question is unsure, and can be easily converted into categories later on.
When is closed ended preferred
Preferred when you have a good idea of potential responses and need a well defined variable or construct.
Probability Sampling
Utilizing some form of random sampling to ensure that everyone in the population has an equal chance of being chosen.
Ex: Random Sampling
Nonprobability Sampling
When the researcher cannot specify the probability that each member of the population will be selected for the sample, ex: convenience sampling.
Sampling Frame
List of all members of population from which to select respondents.
Stratified Random Sampling
Population divided into different subgroups, or strata, then random samples taken from each stratum.
Cluster Sampling
Larger clusters of individuals are sampled and then individuals within each closer randomly sampled.
Sampling Bias
When a sample is selected in such a way that is not representative of the entire population.
Nonresponse bias
If survey nonrespondents differ from survey respondents in systematic ways.
