1/151
Looks like no tags are added yet.
Multivariate Correlations
Multivariate correlations examine relationships between three or more variables at once, allowing for a more nuanced understanding of complex relationships in data.
Temporal Precedence
Establishing that one variable occurs before another in time.
Covariation
Demonstrating that changes in one variable are associated with changes in another.
Control of Third Variables
Using statistical techniques or design elements (like randomization) to account for potential confounding variables that could explain the observed relationship.
Multivariate Designs
Involve multiple variables and can analyze complex relationships among them, often using techniques like multiple regression.
Correlational Designs
Primarily examine the relationship between two variables, without necessarily controlling for other potential influences.
Developmental Approach
Involves studying how variables change over time, often in the context of age-related changes, to better understand development across the lifespan.
Cross-Sectional Designs
Compare different age groups at one point in time to infer developmental changes.
Cohort Effects
Differences between age groups that result from historical or social factors unique to each cohort rather than from age itself.
Longitudinal Designs
Follow the same group of participants over time to observe how variables change.
Selective Attrition
Loss of participants over time, which can bias results if the attrition is related to the study variables.
Cross-Sectional Correlations
Measure associations between variables at a single time point.
Autocorrelations
Measure the correlation of a variable with itself over time.
Cross-Lag Correlations
Measure the relationship between one variable at one time and another variable at a later time to help infer directional relationships.
Multiple Regression Designs
A statistical method that allows for examining the relationship between one dependent variable and multiple independent variables, helping control for confounding variables.
Predictor Variable (IV)
The variable that is believed to predict or cause changes in the other variable.
Criterion Variable (DV)
The outcome or variable being predicted in the analysis.
Beta (β)
The standardized coefficient in a regression analysis that indicates the strength and direction of the relationship between a predictor and the criterion variable, holding other variables constant.
Mediation
Occurs when a third variable explains the relationship between the independent and dependent variables.
Moderation
Occurs when a third variable changes the strength or direction of the relationship between the independent and dependent variables.
Third Variable
A confounding factor that may falsely appear to cause a relationship between two variables but actually influences both independently.
Design Confounds
Occur when there is an additional variable that unintentionally varies along with the independent variable (IV), potentially impacting the dependent variable (DV) and skewing results.
Participant Confounds
Characteristics of participants that vary systematically with the independent variable, potentially influencing the DV in unintended ways.
Systematic Variability
Variability that is related to the manipulation of the IV, creating potential confounds.
Unsystematic Variability
Random or unrelated variability that doesn't correlate with the IV and is generally not problematic, though it may add noise to the data, making it harder to detect effects.
Placebo Effect
Occurs when participants experience changes simply because they believe they are receiving an effective treatment, rather than due to the actual treatment itself.
Selection Effects
When participants in different experimental groups differ systematically due to non-random assignment, potentially influencing the outcome.
Between-Subjects Design
Participants are randomly assigned to different groups, with each group experiencing only one level of the IV.
Post-Test Only Design
Participants are assigned to different groups, and the DV is measured only after the experimental manipulation.
Pre-Test Post-Test Design
Participants are measured on the DV before and after the treatment to assess change over time.
Within-Subjects Design
Each participant experiences all levels of the IV, often in different sessions.
Concurrent Measures Design
All participants are exposed to all levels of the IV at nearly the same time, often in a single session, with the DV measured immediately.
Repeated Measures Design
Participants are exposed to each level of the IV in different trials, with the DV measured after each exposure.
Internal Validity
Refers to the extent to which a study can confidently demonstrate a causal relationship between the IV and DV, free from alternative explanations (confounds).
Time Intervals
The time between measurements or conditions can introduce potential confounds, as other events may impact the DV over time.
Maturation Threats
Changes in participants over time, such as development, fatigue, or natural growth, that may affect the DV independently of the IV.
History Threats
Events occurring between pre- and post-tests that impact participants' behavior or responses.
Regression Threats
The tendency for extreme scores to regress toward the mean over time, which can skew results, especially in repeated measures.
Repetition Issues
Repeated exposure to similar tasks or questions can lead to practice effects, boredom, or changes in response strategy over time.
Attrition Threats
The loss of participants over time, particularly if those who drop out differ systematically from those who stay.
Instrumentation Threats
Changes in the measurement instrument or observer criteria over time that can affect the DV measurement.
Testing Threats
When participants' scores change due to repeated testing or familiarity with the test format rather than due to the IV.
Null Effects
A null effect occurs when there is no statistically significant difference between groups, or no effect of the IV on the DV.
Between-Group Variance
Differences between groups might be too small to detect a significant effect.
Weak Manipulations
If the IV is not manipulated strongly enough, it may not affect the DV.
Insensitive Measures
Measurement tools that cannot detect subtle changes or differences in the DV.
Floor and Ceiling Effects
When most scores are clustered at the low (floor) or high (ceiling) end of the measurement scale, limiting the ability to detect differences.
Manipulation Check
A secondary measure to confirm that the IV manipulation had the intended effect on participants.
Pilot Tests
A preliminary test to evaluate the strength of the manipulation and measurement sensitivity before conducting the main experiment.
Within-Group Variance
Variability within groups, making it harder to detect an effect.
Measurement Error
Inaccuracies in how the DV is measured, leading to variability in results.
Individual Differences
Variability between participants, such as personality traits or abilities, that can obscure effects of the IV.
Situational Noise
External factors, such as environmental conditions or distractions, that add variability within groups.
Maternal Responsiveness
How quickly a mother responds to her infant's cries.
Infant Fussiness
The level of fussiness or crying frequency/intensity in the infant.
Cross-Sectional Comparisons
Correlate maternal responsiveness and infant fussiness within the same time period.
Expected Pattern of Results
If maternal responsiveness causes a reduction in infant fussiness over time, the expected pattern would be significant cross-lag correlations.
Depressive Affect
Levels of depressive symptoms tracked over seven days.
Binge Eating
Frequency or intensity of binge eating episodes tracked over the same period.
Cross-Sectional Comparisons (Depression Study)
Correlations between depressive affect and binge eating within the same day.
Autocorrelations (Depression Study)
Correlations of each variable with itself across consecutive days.
Cross-Lag Correlations (Depression Study)
Correlations between one variable at an earlier time and the other variable at a later time.
Cross-Lag Correlations (Directional Relationship)
Indicate a potential directional relationship where depressive affect on one day is positively correlated with binge eating the next day.
Independent Variable (IV)
Serving bowl size.
Levels/Conditions of IV
Large bowl vs. Medium bowl.
Dependent Variable (DV)
Amount of pasta consumed (measured in calories or weight).
Control Condition
Medium bowl (serving size is presumably less influenced).
Treatment Condition
Large bowl (hypothesized to lead to larger portions due to visual or psychological influence).
Possible Confounds
Environmental factors, hunger levels, social influences, and type of pasta or food quality that could influence participants' eating behaviors.
Random Assignment
Ensures participants are randomly assigned to each condition to balance any pre-existing differences between groups.
Matched Groups
Matching participants on factors such as hunger level or baseline portion size to ensure similar groups across conditions.
Pre-Test
Administering a brief survey about eating habits, hunger levels, or preferences to match participants effectively.
Independent Variable (IV) in Note-taking Study
Note-taking method.
Levels/Conditions of IV in Note-taking Study
Laptop note-taking vs. Longhand (notebook and pen) note-taking.
Dependent Variable (DV) in Note-taking Study
Test performance, measured separately on factual and conceptual questions.
Control Condition in Note-taking Study
There is no explicit control group, as both groups used note-taking methods.
Treatment Condition in Note-taking Study
Each note-taking method can be considered a treatment condition, comparing the effects of each on test performance.
Typing vs. Writing Style
Typing may lead students to take verbatim notes, while writing by hand often requires summarizing and processing information.
Typing Speed
Differences in individual typing or writing speed may affect how much information students capture.
Prior Preference or Experience
If participants are more accustomed to one note-taking method, it could affect how effectively they take notes and perform on tests.
Blinding
Keeping participants unaware of study hypotheses.
Matching on Prior Experience
Balancing groups based on prior familiarity.
Baseline Test
Pre-test to control for initial differences.
Factual Questions
Scores similar between laptop and longhand groups.
Conceptual Questions
Longhand group scores higher on understanding.
Empathy Condition
Designers engage with wheelchair users directly.
Increase Sample Size
Larger participant pool enhances statistical power.
Within-Groups Design
Each participant experiences all conditions.
Effectiveness Scale
Judging designs based on effectiveness criteria.
Longhand Note-Taking
May enhance conceptual understanding and retention.
Laptop Note-Taking
Similar results for factual recall as longhand.
Polarized Views
Extreme opinions developed during political course.
Post-Test
Final assessment after intervention concludes.
Statistical Fluctuation
Natural variation affecting observed results.
Study Hypothesis
Proposed explanation tested through research.
Benefit of Within-Groups Design
Controls individual differences by using same participants.
Control for Individual Differences
Match participants on key characteristics statistically.
Empathy Intervention
Enhances designers' understanding of challenges faced.
Causal Inference
Suggests relationships but does not confirm causation.
Multiple-Regression Designs
Measure multiple predictors to analyze dependent variable.