1/60
Vocabulary practice flashcards covering correlational designs, longitudinal and multivariate research, experimental methodology, validity threats, null results, and empirical applications from the Pennycook papers for PSY 1004 Exam 3.
Correlational Research
Research in which variables are measured as they naturally occur rather than manipulated or assigned by the researcher.
Quantitative Variables
Variables measured on a numerical scale, which are best displayed using scatterplots when examining the relationship between two such variables.
Categorical Variables
Variables that sort cases into discrete groups, best displayed using bar graphs or grouped plots.
Correlation Coefficient (r)
A statistic quantifying the strength and direction of a linear relationship between two quantitative variables, where the sign (+ or ā) indicates direction and the absolute value (0.00 to 1.00) indicates strength.
Effect Size
A statistic that measures the strength or practical magnitude of an association (such as the magnitude of r).
Statistical Significance (p-value)
An indicator of whether an observed association is unlikely to be due to chance alone, which is heavily influenced by sample size.
95% Confidence Interval (CI)
A range of plausible values for the true population association; if it includes 0, a zero association remains plausible and the result is not statistically significant at α=.05.
Outliers
Extreme data points that fall far from the rest of the distribution and can artificially inflate, deflate, or reverse a correlation coefficient, especially in small samples.
Restricted Range
A condition occurring when a sample only includes a narrow slice of the full possible range of values on one or both variables, artificially weakening (attenuating) the observed correlation.
Curvilinear Relationship
A non-linear association (such as a U-shaped or inverted-U relationship) where the direction of the relationship changes across levels of a variable, which can cause linear correlation (r) to be near zero.
Directionality Problem
Ambiguity in correlational research regarding whether Variable A causes Variable B or Variable B causes Variable A due to a lack of temporal precedence.
Third-Variable Problem (Confound)
The possibility that an unmeasured external variable causes both variables under study, creating a spurious (false) association between them.
Moderation (Interaction)
A phenomenon that occurs when the strength or direction of the association between two variables changes depending on the level of a third variable (the moderator).
Cross-Sectional Correlation
The correlation between two different variables measured at the exact same point in time to assess concurrent association.
Autocorrelation
The correlation of the same variable with itself measured at two different points in time, used to assess the stability of a construct over time.
Cross-Lag Correlation
The correlation between one variable at Time 1 and a different variable at Time 2, helping establish temporal precedence (A1āāB2ā vs. B1āāA2ā).
Predictor Variable
An explanatory or measured input variable in regression analysis used to predict changes in the outcome.
Criterion Variable
The primary outcome variable being predicted in regression analysis, representing the target variable of interest.
Temporal Precedence
Establishing that a proposed causal variable occurs earlier in time than its proposed effect.
Multiple Regression
A statistical technique that uses multiple predictor variables simultaneously to predict a single criterion variable.
Adjusted Regression Coefficient (Partial Slope)
The association between a specific predictor and the criterion variable while holding all other measured predictors constant (controlling for them).
Unstandardized Coefficient (b)
A regression slope expressed in the original measurement units of the predictor and criterion variables.
Standardized Coefficient (β)
A regression slope expressed in standard deviation units, allowing direct comparison of relative predictive strength across predictors measured on different scales.
Mediator
An intermediate internal variable that explains the mechanism or pathway through which Variable A leads to Variable B (AāMediatorāB).
Independent Variable (IV)
The variable actively manipulated by the researcher, with levels representing the specific experimental conditions created.
Dependent Variable (DV)
The variable measured by the researcher to assess the effect of the manipulated independent variable.
Control Variables
Extraneous factors held constant across all experimental conditions to prevent systematic variation.
Covariance
Demonstrating that changes in the manipulated independent variable are systematically associated with changes in the dependent variable.
Random Assignment
Assigning participants to experimental conditions purely by chance to equalize groups at baseline and address internal validity.
Random Sampling
Selecting participants from a population purely by chance to ensure the sample represents the population and address external validity.
Selection Effects
Systematic pre-existing differences between groups prior to manipulation, occurring when participants choose their group or non-random assignment is used.
Matched Groups
An experimental procedure where participants are grouped into blocks based on a key baseline characteristic, and members of each block are then randomly assigned to conditions.
Internal Validity
The extent to which alternative explanations and confounds are ruled out, proving that the independent variable caused the dependent variable.
External Validity
The extent to which experimental results generalize to other populations, settings, or times.
Construct Validity
The degree to which operationalizations accurately capture the theoretical independent variable and dependent variable constructs.
Posttest-Only Design
An independent-groups design where different participants are assigned to different independent variable conditions and measured once on the dependent variable after manipulation.
Pretest/Posttest Design
An independent-groups design where different participants are assigned to different independent variable conditions and measured on the dependent variable both before and after manipulation.
Repeated-Measures Design
A within-groups design where the same participants experience all levels of the independent variable sequentially.
Concurrent-Measures Design
A within-groups design where the same participants are exposed to all independent variable levels simultaneously and make a single choice or response.
Order Effects
A threat in within-groups designs where the sequence of condition presentation alters participant performance, encompassing practice, fatigue, and carryover effects.
Practice Effect
An order effect where participant performance improves over the course of a study due to repetition or familiarity.
Fatigue Effect
An order effect where participant performance declines over time due to tiredness or boredom.
Carryover Effect
An order effect where spillover effects from an earlier condition contaminate participant responses in subsequent conditions.
Counterbalancing
Systematically varying the order of condition presentation across participants to distribute order effects evenly across all levels of the independent variable.
Design Confound
An unintentional secondary variable that varies systematically along with the levels of the independent variable, compromising internal validity.
Maturation Threat
A threat to internal validity involving natural biological, psychological, or developmental changes in participants over time.
History Threat
A threat to internal validity involving an external event occurring between pretest and posttest that influences participants broadly.
Regression to the Mean
The tendency of extreme scores at initial measurement to move closer to the average on re-testing due to random measurement error.
Attrition (Mortality)
The systematic loss of participants over time, particularly differential dropout between conditions, threatening internal validity.
Testing Effect
A threat to internal validity where changes in participant performance are caused specifically by taking the pretest prior to the posttest.
Instrumentation Threat
A threat to internal validity caused by changes in measuring instruments, observer coding criteria, or calibration over time.
Observer Bias
A threat occurring when experimenter expectations unconsciously influence data recording or participant treatment.
Demand Characteristics
Cues in an experiment that reveal the study hypothesis, causing participants to change their natural behavior.
Placebo Effect
Participant improvement driven purely by their belief that they are receiving an active treatment.
Ceiling Effect
A measurement problem where scores cluster at the maximum possible limit of the dependent variable across all conditions.
Floor Effect
A measurement problem where scores cluster at the minimum possible limit of the dependent variable across all conditions.
Statistical Power
The probability of correctly detecting a true effect when one exists, maximized by increasing sample size, strengthening manipulations, and reducing experimental noise.
Cognitive Reflection Test (CRT)
A measure of analytic thinking evaluated in Pennycook & Rand (2019) that assesses an individual's inclination to override intuitive gut responses with deliberate reasoning.
Headline Discernment
The capacity to differentiate true from false news headlines in accuracy ratings.
Accuracy Prompts (Nudges)
An experimental manipulation (such as rating the accuracy of a single neutral headline at baseline) tested by Pennycook et al. (2021) to prime the concept of accuracy.
Sharing Discernment
The relative difference in social media sharing intentions between true versus false news headlines, which served as the primary dependent variable in Pennycook et al. (2021).