elementary psych 9-3

Definition: measures two or more variables to assess whether they are associated; no manipulation of variables.
Core idea: correlation does not imply causation; an association between variables does not prove one causes the other.
Key metric: correlation coefficient $r$ ranges from $-1$ to $1$ ; $r=0$ indicates no correlation; positive r>0 means both variables increase together; negative r<0 means one increases while the other decreases.
Graph directions:
- Positive correlation: both increase together.
- Negative correlation: one increases as the other decreases.
- No correlation: no predictable pattern.
Third-variable problem: apparent correlation can be due to a third variable (confound) not accounted for.
Example concepts from lecture:
- Height vs reading level shows a positive correlation but is likely driven by age/grade level as a third variable.
- Ice cream sales vs homicide rates (common example for illustrating correlation without causation).
Important takeaway: correlations show relationships, not causality; beware spurious correlations.

Definition: scientific procedure where one or more variables are manipulated and then measured to assess cause-effect.
Key terms:
- Independent Variable (IV): the variable that is deliberately manipulated.
- Dependent Variable (DV): the outcome that is measured.
- Control group: baseline condition used for comparison.
- Experimental group: receives the manipulation of the IV.
- Placebo: inert treatment used to control for expectations.
Example scenario (lecture): does wearing name-brand shoes affect basketball performance? IV = shoe brand; DV = basketball scores; Control group uses generic shoes; Experimental group uses name-brand shoes; Placebo control possible for expectations.
Data collection and inference: after manipulation, conduct statistical analysis to determine if differences between groups are likely not due to chance.
For exams: you don’t need to know specific statistical tests; focus on the idea that you test for significance of group differences.

Random assignment: each participant has an equal chance of being in the experimental or control group, helping ensure groups are comparable.
Why it’s important: helps ensure observed effects are due to the IV, not preexisting differences.
Fair assignment caveats: avoid systematic bias (e.g., grouping by gender, seating location) that could confound results.

Biases: factors that systematically affect performance (e.g., time of day, temperature, fatigue) and can skew results.
Demand characteristics: cues that reveal the researcher’s expectations, causing participants to alter their behavior.
Placebo effect: improvements due to participants’ expectations rather than the active treatment.
Remedies:
- Use placebo controls where appropriate.
- Implement double-blind designs to reduce both participant and experimenter expectancy effects.
- Ensure procedures minimize cues about expected outcomes.

Definition: both participants and data-collectors are unaware of treatment assignments.
Structure: one researcher knows group assignments (unblinded) but does not collect data; all others collect data blind to conditions.
Goal: reduce bias in data collection and analysis; blind is removed only after data collection is complete.

Scenario: Previous research indicates students learn more when engaging in group activities rather than individual work.
- Independent Variable (IV): the group condition (group activities vs individual work).
- Dependent Variable (DV): learning outcomes or scores.
Answer: IV = group condition; DV = learning outcomes/scores.

Correlational vs Experimental:
- Correlational: measure associations, no manipulation; cannot infer causation.
- Experimental: manipulate IV, measure DV; supports causal conclusions under proper controls.
Key controls and biases:
- Random assignment, control group, placebo, double-blind design.
- Be mindful of time-of-day, environmental factors, and demand characteristics.