Single Subject Research
The no-treatment phase, labeled A, that works as a kind of control condition.
What is baseline?
Assigning scores to individuals so those scores represent some specific characteristic of those individuals.
What is measurement?
Same scale, same people, taken one week and one year apart. The scores of all tests correlate at +.80 or higher. This shows this type of reliability
What is test-retest reliability?
Survey research has two key features: self-report measures and careful attention to this, which decides who answers the survey.
What is sampling?
Every member of the population has an equal chance of being selected, like drawing names from a hat. This is this type of sampling.
What is simple random sampling?
A survey item reads: "How satisfied are you with your pay and your supervisor?" What is wrong with this item?
What is a double-barreled item? Split it into two questions.
The term given to waiting until behavior is fairly consistent within a phase before switching conditions.
What is the steady state strategy?
Intelligence, self-esteem, and anxiety and traits you cannot observe directly are variables called
What are psychological constructs?
Two observers independently watch the same children on the playground and rate each child's aggression. Their ratings closely agree. This shows this type of reliability.
What is interrater reliability?
A respondent must first decide whether "alcoholic drinks" includes beer and wine. Of the five steps (interpret, retrieve, judge, format, edit), name the step they are on.
What is interpreting the question?
A researcher asks current participants to recruit their friends to join the study and so on, and so on. They are using this type of sampling.
What is snowball sampling?
A researcher says: "One person rated their pain an 8 and another rated it a 4, so the first person hurts twice as much." The researchers flaw is
What is treating ordinal data as ratio data? A pain rating has no true zero.
Baseline, treatment, then baseline again. If behavior changes back when treatment is removed, the treatment is the likely cause.
What is a reversal (ABA) design?
Used with observation an example is counting how many times a child helps a classmate during free play or another action.
What is a behavioral measure?
People who score high on one item of a self-esteem scale tend to score high on the other items too, and Cronbach's alpha checks it. This shows this type of reliability.
What is internal consistency?
Researchers asked about dating frequency first, and it changed how people then rated their life satisfaction. An earlier question changing answers to a later one is called this.
What is an item-order effect (a context effect)?
A researcher randomly picks several towns, then randomly picks residents within each town. This is this type of sampling.
What is cluster sampling?
A researcher says: "Our scale had an alpha of .92, so it is a valid measure of anxiety." What is wrong with this claim?
What is confusing reliability with validity? Alpha shows internal consistency only.
Treatment starts at staggered times across participants, behaviors, or settings, so it never has to be taken away.
What is a multiple-baseline design?
Using multiple means of testing and measurement on a single construct in several different ways and getting similar results is termed.
What are converging operations?
A new motivation scale should NOT correlate with a vocabulary test, because the two constructs are distinct. This shows this type of validity.
What is discriminant validity?
The acronym BRUSO is used to describe a good survey question. BRUSO stands for these words
What is Brief, Relevant, Unambiguous, Specific, and Objective.
A random sample receives a survey, but the people who answer differ systematically from the people who do not. This type of bias results.
What is nonresponse bias?
A student doing a self-management single subject research design says: "I tracked my study hours for a week, added a reward, and my hours went up. The reward caused it." The problem is
What is an AB design with no reversal? Remove the reward (ABA) to rule out other explanations.
The calculation on the amount of treatment responses more extreme than the most extreme baseline response. It adds to visual inspection but does not replace it.
What is the percentage of non-overlapping data (PND)?
In this level of measurement type the order, equal differences, and a true zero are present, so "twice as much" numerically makes sense.
What is the ratio level?
Students take a test anxiety measure today, and their scores line up with their exam grades next month. This shows this type of validity.
What is predictive validity (a type of criterion validity)?
Several agreement statements about one attitude, with opposite-worded items reverse coded, then summed into one total score. Name this type of scale.
What is a Likert scale?
In 1936 a magazine poll collected millions of ballots and still predicted the wrong presidential winner. Name the poll.
What is the Literary Digest poll?
tudents write their names on an honesty survey and hand it straight to the professor running the study. Nearly everyone reports that they never cheat. This response problem most likely explains the results
What is socially desirable responding