Preview this deck
Cross-sectional study
Front
| 5 | 0 | ||
| 4 | 0 | ||
| 3 | 0 | ||
| 2 | 0 | ||
| 1 | 0 |
Active users
1
All-time users
1
Favorites
0
Last updated
2 years ago
Date created
Dec 2, 2021
Cards (113)
Terms
(113 cards)
Cross-sectional study
An observational study that involves measurement at one point in time.
Independent variable
The manipulated variable in a true or quasi-experiment.
Open trial
A study in which all participants receive the targeted treatment and are aware of the treatment
Allegiance effects
A confound involving differences due to sites that are particularly associated with one of the treatments, resulting in a heightened level of expectation for the efficacy of that tx and unusually large tx effects.
PRISMA statement
A set of recommendations for the conduct of meta-analyses
FDA Approval Process: Phase 0
A small-sample study involving administration of a single, subtherapeutic dose of the medication to humans to verify that the medication acts in humans as expected.
Number needed to treat
An effect size statistic in circumstances where both variables are dichotomous estimating the number of additional patients who would need to receive the active treatment to achieve one more positive outcome.
FDA Approval Process: Phase III (Therapeutic Confirmatory):
Large, multisite studies focusing on efficacy.
Beta Level
The acceptable (researcher-defined) probability of a Type I error.
Blind
Unaware of a condition or independent variable.
Fatigue effect
An order effect involving deterioration over time.
Exclusionary criteria
Criteria of excluding a participant from a study
Type I error
An error that involves rejecting the null hypothesis when the null hypothesis is true.
Retrospective study
A cross-sectional study in which information about past events is gathered.
Prospective study
A study in which events are used to predict events at later periods
Washout period
An interval between administration of repeated measures treatments to allow return to baseline.
Coefficient alpha
The most popular statistic for the estimation of internal reliability.
Double-blind study
Study where participants and providers are unaware of participants’ treatment conditions
Number needed to harm
An effect size statistic in circumstances where both variables are dichotomous estimating the number of additional patients who would need to receive the active treatment to achieve one more negative outcome (e.g., some side effect).
Conflict of interest
Economic or other factors that could compromise a researcher’s objectivity in a research study.
Analysis of covariance
An analysis similar to analysis of variance that allows for equalization of participants on some potential confound.
Phase IV (Therapeutic Use):
Research after approval, focusing on safety and new uses.
Alternative hypothesis
The proposition that some effect exists in a population
Alpha level
The acceptable (researcher-defined) probability of a Type I error.
Pilot study
A preliminary small-sample study that is often weak in terms of both internal and external validity.
Relative risk
The probability of improvement in the active treatment relative to the control condition
Efficacy studies
Studies in which internal validity is emphasized, to the detriment of external validity
Effect size statistics
Statistics that are designed to estimate the size of an effect in a sample or population. Effect: Some hypothesized relationship or pattern. Effectiveness studies: Studies in which external validity is emphasized, to the detriment of internal validity.
Practice effect
An order effect involving improvement over time
Power
The probability of rejecting the null hypothesis when the null hypothesis is false; the complement of beta.
Phi coefficient
The correlation coefficient in circumstances where both variables are dichotomous
Standardized mean difference
Several effect size statistics (e.g., d) that present the difference between two groups relative to the size of the standard deviation within groups
Last observation carried forward
Intent to treat comparison using the final observation for all participants in the study.
Comparator
A treatment alternative to the original treatment.
Covariate
A confound controlled through an analysis of covariance
Reliability
Degree to which variability within a set of scores is attributable to true variation rather than random error.
Dependent variable
The outcome variable in a true or quasi-experiment.
Internal reliability
A method of estimating reliability based on responses across multiple items of a test
Case Studies
Relatively unsystematic observation of single individuals; contrasted with single-subject designs
Multiple baseline design
A single-subject design that involves introducing intervention at different times across participants or settings
Crossover study
A repeated measures study in which the same participants receive all treatments.
Nuisance variables
Variables that could cause spurious relationships between variables in a study.
Pharmacokinetics
The body’s effect on a drug.
CONSORT statement
A set of recommendations for the conduct of randomized controlled trials.
Statistical conclusion validity
Activities that compromise the validity of the conclusions derived from quantitative data.
Order effects
Confounds in repeated measures studies having to do with the order in which treatments are received
Partial counterbalancing
Counterbalancing using some possible orders of treatments
Critical value
The value for some test statistic at which the null hypothesis is rejected given a predefined alpha level.
Assumptions
Statistical requirements underlying the use of a statistical method
Construct validity
Having to do with the relationship between the operations performed in the process of conducting the study and the latent constructs they are intended to represent.
Open-label trial
A study in which all participants receive the targeted treatment and are aware of the treatment
Single-blind study
A study in which participants are unaware of their treatment condition
Preexisting groups design
A between-group study with non-random assignment
Repeated measures design
A comparison of matched groups, because of a dependency between groups or a within-group design.
Institutional Review Board
A committee that must approve any research conducted by an organization hoping to receive federal funding for research.
Black Box Warning
A warning of serious side effects associated with a medication.
FDA Approval Process: Phase I (Human Pharmacology):
A small-sample study that focuses on safety, pharmacodynamics, pharmacokinetics, and dose ranging in humans
ABAB design
A single-subject design involving sequential administration and removal of the active treatment
Parameter
A population statistic.
Moderators
Third variables that influence the strength of the relationship between the treatment variable and outcomes.
Confounds
Variables that could cause spurious relationships between variables in a study
Observed cases
A comparison of treatments based only on participants who completed the study
Inclusionary criteria
Criteria for including a participant in a study
Pharmacodynamics
A drug’s effect on the body
Significance testing
An inferential strategy developed to evaluate whether null hypotheses are true in a population.
Null hypothesis
The proposition that a given effect is absent in a population
Carryover effect
A confound in repeated measures studies in which one treatment influences response to the next.
Informed consent
Agreement to participate in research under conditions of full disclosure.
Absolute risk
The probability of improvement in the active treatment
Pretest-posttest design
A study involving a single group that receives the active treatment
Waitlist control
An alternative to the active treatment involving no treatment
Longitudinal design
A study that involves observation over an extended period of time.
Analysis of variance
A statistical method developed by Sir Ronald Fisher that involves one or more categorical independent variables of two or more groups and one-dimensional dependent variable.
Dose-response curve
Variation in relationship between tx and outcome across medication dosage
Observational research
A study in which there is no assignment or manipulation by the researcher
Type II error
An error that involves retaining the null hypothesis when the null hypothesis is false.
Titrating
A method for determining the endpoint of a reaction, and therefore the precise quantity of a drug needed to achieve the reaction.
Fishing expedition
Study where numerous analyses are conducted but only those that are significant are reported.
In vitro research
Research conducted in test tubes
Test-retest reliability
Method of estimating reliability based on responses across multiple administrations of some test.
True experiment
A study in which the researcher randomly assigns participants to a condition and differences in condition on some outcome that is evaluated.
Market exclusivity
The period during which FDA approval permits a company to market a medication exclusively.
Intraclass correlation coefficient
A statistic used to estimate reliability
External validity
Degree to which inferences are likely to generalize across populations and contexts
Validity
The extent to which inferences derived from a research study must be considered conditional or compromised by limitations of the study itself.
Survival analysis
A set of statistics used to study the attrition process itself.
In vivo research
Research conducted with humans or in culture
Between Group Design
A comparison of two or more separate groups.
Spurious relationships
Relationships between variables that are attributable to a third variable.
Correlation coefficient
A standardized measure of the degree to which one variable can be predicted from the other
Full counterbalancing
Counterbalancing using all possible orders of treatments.
Internal validity
The degree to which the inferred causes of a relationship are justified by the results
Within-group design
A study in which the same participants are exposed to all conditions.
Interaction
A statistical product of a moderator relationship in analysis of variance
Nested design
An experimental design in which the variables are arranged in an explicit hierarchy.
Therapeutic window
The range of dosages in which the medication is efficacious.
Meta-analysis
A variety of methods used to estimate outcomes across a series of studies (usually effect size statistics) to provide a summative analysis of some body of literature
Placebo
An alternative overtly similar to the active treatment that should not demonstrate ameliorative physiological effects but should activate similar expectations of improvement.
Power analysis
A procedure for estimating the appropriate sample size to achieve a desired level of power
Counterbalancing
Systematic changes in the order of administration of treatment’s in a repeated measures study to minimize order effects
Triple-blind study
A study in which participants, care providers, and data gatherers are unaware of participants’ treatment conditions.
Intent to treat
Analysis involving all individuals for whom there was an intention of tx regardless of participation level.
FDA Approval Process: Phase II (Therapeutic Exploratory):
The first large-sample studies conducted to evaluate safety, dose-ranging, and efficacy in humans.
Odds ratio
An effect size statistic in circumstances where both variables are dichotomous based on the relative odds of improvement in two treatments
Classical test theory
A dominant approach to evaluating measurement devices in psychology that focuses on the estimation of reliability.
Dose ranging
Identifying the range of dosages likely to define the therapeutic window
Relative risk ratio
An effect size statistic in circumstances where both variables are dichotomous based on the relative probability of improvement in two treatments
Inter-rater reliability
A method of estimating reliability based on responses across multiple raters using the same test.
Confidence intervals
A range of values derived from a sample in which a parameter is likely to fall.
Quasi-experiment
A study that involves conditions to which participants are not randomly assigned
Randomized controlled trials
A true experiment in which individuals are assigned to treatment conditions
Nocebo
Deterioration or side effects in response to a placebo; or, any adverse event associated with a treatment
Mixed-factors ANOVA
An analysis of variance involving at least one between group and at least one within-group independent variable.