Term
|
Definition
| "A design in which there are two levels of each of two independent variables, yielding four conditions" |
|
|
Term
|
Definition
| A counterbalancing technique in which the treatment orders are randomized in successive blocks of presenting those conditions |
|
|
Term
|
Definition
| When the effect of one independent variable on a dependent variable reverses at different levels of a second independent variable |
|
|
Term
|
Definition
| "A message, signal, or pertinent fact that makes you doubt the accuracy or credibility of a persuasive message" |
|
|
Term
|
Definition
| An experimental design in which each level of every independent variable occurs with all levels of the other independent variable |
|
|
Term
|
Definition
| Experimental results that occur when the effects of one independent variable depend on the levels of other independent variables |
|
|
Term
|
Definition
| When the effect of one independent variable is the same at all levels of another independent variable |
|
|
Term
|
Definition
| Experimental design in which subjects are matched on some variable assumed to be correlated with the dependent variable and then randomly assigned to conditions |
|
|
Term
|
Definition
| An experimental design that contains both between-subjects and within-subjects manipulations of the independent variables |
|
|
Term
|
Definition
| A device that simulates only a portion of a system; see simulator |
|
|
Term
|
Definition
| When subjects are randomly assigned to conditions in a between-subjects design |
|
|
Term
|
Definition
| The process of unbiased assignment of subjects to conditions or unbiased variation of condition order |
|
|
Term
|
Definition
| "Several measures are taken on the same participant, such as several learning trials or numerous psychophysical judgments; a type of within-subjects experiment" |
|
|
Term
|
Definition
| A device that duplicates the functions of a real system to allow controlled experiments to be performed |
|
|
Term
|
Definition
| An improvement in the effect of a persuasive message with the passage of time |
|
|
Term
| Treatment x treatment x subjects |
|
Definition
| A within-subjects factorial design with two independent variables |
|
|
Term
|
Definition
| A frequently used design in which a therapy (B) is instituted after measuring a particular behavior (A); a poor research design |
|
|
Term
|
Definition
| See reversal (ABA) design |
|
|
Term
| Alternating-treatments design |
|
Definition
| A small-n design in which two or more independent variables alternate |
|
|
Term
| Changing-criterion design |
|
Definition
| A small-n design in which the criterion behavior needed to produce an outcome changes |
|
|
Term
|
Definition
| An analysis of the antecedents and consequences of a particular behavior; usually undertaken before the implementation of a behavioral treatment |
|
|
Term
|
Definition
| Designs in which inferential statistics are used to determine whether two or more groups differ reliably |
|
|
Term
|
Definition
| A small-n design in which different behaviors (or different participants) receive baseline periods of varying lengths prior to the introduction of the independent variable |
|
|
Term
|
Definition
| "Small-n design in which a participant’s behavior is measured under a baseline (A) condition, then an experimental treatment is applied during the B phase and any changes in behavior are observed; finally, the original baseline (A) conditions are reinstituted to ensure that the experimental treatment was responsible for any observed change during the B phase" |
|
|
Term
|
Definition
| Research design using a small number of subjects |
|
|
Term
| why can't observation treatment observation be reversed? |
|
Definition
the teatment is not under the experimeters control
most natural treatments are likely to have carryover effects |
|
|
Term
| what are two threats to interal validty that occur naturally in treatments |
|
Definition
history of the participants
any changes in the participant that occur over time |
|
|
Term
| The comparison case in DV case study is a ____because |
|
Definition
| "nonequivalen control because it was selected, not randomly assigned" |
|
|
Term
| "in matching, be alert for" |
|
Definition
synergism
regression artifacts |
|
|
Term
|
Definition
| confounds time of treatment or testing (can't know what caused the change in participant) |
|
|
Term
|
Definition
| confound cohort of the participants with the target variable |
|
|
Term
| "if you text effects of age both longitudinally and sectionally, and they are the same, what conclusions can be drawn?" |
|
Definition
| Ruling out cohorts an time of testing as important confoundings. |
|
|
Term
|
Definition
| the effects of one variable are dependent on the level of the other variable |
|
|
Term
| when using age and a independent variable will expect a____ |
|
Definition
|
|
Term
| enhance internal validity with age as a variable |
|
Definition
| "cross sequential design and the search for interactions, convergining operations." |
|
|
Term
| quasi experiments are similar to ex post facto for 2 reasons |
|
Definition
they can have: naturally occuring variables
subject variables |
|
|
Term
| "In natural treatments, what can confound the results" |
|
Definition
history
maturation
selection bias
nonequivalent control groups
|
|
|
Term
| what are two quasi-experimental procedures with low internal validity |
|
Definition
one-shot case studies
intterupted time series design |
|
|
Term
| threats to internal validity on the two worst forms of quasi? |
|
Definition
| mortality and delayed effects |
|
|
Term
| characterisitcs in matching can… |
|
Definition
regresses towards mean
interacts with the matched variables |
|
|
Term
| whats a problem that comes from nonequivalent control group design? |
|
Definition
|
|
Term
| One shot cases don't have? |
|
Definition
baseling observations
control condition |
|
|
Term
| why is quasi better than ex post facto? |
|
Definition
| statements about causation are prefered to correlation |
|
|
Term
| A benefit from interrupted time series? |
|
Definition
| use several depedent variables and can be examined like a case study |
|
|
Term
| difficulties of inerrupted time series 2 |
|
Definition
effects of treatment are: delayed
masked by other variables |
|
|
Term
| improve internal validity in case studies |
|
Definition
deviant case analysis
nonequivalent control
mulitple depedent variables |
|
|
Term
| improve internal validity in interrupted time series? |
|
Definition
non equivalent control
multiple dependent variables |
|
|
Term
|
Definition
multiple independent variables
every subjects receives all combinations of levels of all IV's
need counterbalancing |
|
|
Term
|
Definition
| "Variables that covary with a target in statistical analyses, such as regression, that are accounted for in the analysis of the target" |
|
|
Term
|
Definition
| Taking a large sample of the population of various ages at one time and testing them; contrast with longitudinal method |
|
|
Term
|
Definition
| A quasi-experimental design used when age is a subject variable to try to control for cohort and time-of-testing effects; involves testing several different age groups at several different time periods; see longitudinal method and cross-sectional method |
|
|
Term
|
Definition
| A possible confound in research that inadvertently takes place between measurements because of historical changes in the participant |
|
|
Term
|
Definition
| Experimental results that occur when the effects of one independent variable depend on the levels of other independent variables |
|
|
Term
| Interrupted-time-series design |
|
Definition
| A quasi-experiment that involves examination of a naturally occurring treatment on the behavior of a large number of participants |
|
|
Term
|
Definition
| Testing one group of people repeatedly as they age; contrast with cross-sectional method |
|
|
Term
|
Definition
| See subject variable and matched-groups design |
|
|
Term
|
Definition
| Changes in people over time because of growth and other historical factors; may be a source of confounding in quasi-experiments |
|
|
Term
|
Definition
| A possible source of confounding in research resulting from participants dropping out either because they will not participate or because they cannot participate |
|
|
Term
|
Definition
| A type of regression analysis that uses more than one predictor variable |
|
|
Term
| Nonequivalent control group |
|
Definition
| "In quasi-experiments, a control group that is not determined by random assignment but is usually selected after the fact and is supposed to be equivalent to the naturally treated group" |
|
|
Term
| Observation-treatment-observation |
|
Definition
| A quasi-experimental design; usually includes a nonequivalent control group |
|
|
Term
|
Definition
| A quasi-experiment in which the behavior of a single individual is studied and “explained” in terms of life events; see case study and deviant-case analysis |
|
|
Term
|
Definition
| An experiment in which the independent variable occurs naturally and is not under direct control of the experimenter; see ex post facto |
|
|
Term
|
Definition
| An artifact in the measurement of change on a variable when groups of subjects who scored at the extremes on the variable are tested again; see regression to the mean |
|
|
Term
|
Definition
| "Tendency for extreme measures on some variable to be closer to the group mean when remeasured, due to unreliability of measurement" |
|
|
Term
|
Definition
| Occurs when subjects are not selected randomly; see subject attrition and matching |
|
|
Term
|
Definition
| "Some characteristics of people that can be measured or described but cannot be varied experimentally (e.g., height, weight, sex, IQ)" |
|
|
Term
|
Definition
| A term used in medical research to describe an interaction; see interactions |
|
|
Term
|
Definition
| "A quasi-experimental design used when age is a subject variable in order to control time of testing effects; subjects of a particular age (e.g., 19-year-olds) are tested at different time periods" |
|
|
Term
|
Definition
| unbiased assignment of subjects to conditions |
|
|
Term
| multi factor experiments have ___ and they have____ |
|
Definition
- one independent variable
- may have more than two levels |
|
|
Term
| what happens when more than two indep variables are varied? |
|
Definition
| it's a factorial design in which all possible combination of the levels of the independent variables are examined |
|
|
Term
| what is the primary reason for using multifactor designs? |
|
Definition
|
|
Term
| when there are two indep variables there are.... |
|
Definition
| two main effects, each associated with each variable, and there can be an interaction |
|
|
Term
|
Definition
one or more variables are varied within subjects;
one or more variables are varied between subjects |
|
|
Term
| why is AB an invalid way to evaluate therapy? |
|
Definition
| because changes occurring during treatment in the b phase may be cause by other factors that confounded with the factor of interest |
|
|
Term
|
Definition
| the functional relation between what leads to the target behavior and the consequences that it produces |
|
|
Term
| multiple baseline design with several behaviors is? |
|
Definition
|
|
Term
| multiple baseline design with several people is? |
|
Definition
|
|
Term
| the case study is a ____design |
|
Definition
|
|
Term
| the small-n design assumes that valid conclusions can be made by... |
|
Definition
| making large numbers of observations on few subjects in a controlled situation |
|
|
Term
|
Definition
|
|
Term
| What is the indepedent variable in ABA? |
|
Definition
|
|
Term
|
Definition
| interspersed between baseline periods are independent variable phases |
|
|
Term
| What is an alternating treatments design? |
|
Definition
a reversal design, involving numerous baseline periods and more than two levels of the independent variable.
ACABCBCB |
|
|
Term
| when is the multiple baseline design useful? |
|
Definition
when the independent variable has carryover
behaviors to be altered are not independent of each other |
|
|
