Step 1: Planning the Research

Planning the Research

Statistical Hypotheses: Formalize predictions about population relationships.
- Null Hypothesis: Predicts no effect/relationship.
- Alternative Hypothesis: States a prediction of effect/relationship.
- Examples:
  - Effect Hypothesis:
    - Null: Meditation has no effect on math scores.
    - Alternative: Meditation improves math scores.
  - Correlation Hypothesis:
    - Null: Parental income and GPA have no relationship.
    - Alternative: Parental income and GPA are positively correlated.

Overall Strategy for Data Collection and Analysis: Determines the statistical tests applicable.
Types of Designs:
- Experimental Design: Assesses cause-effect relationships using statistical comparison or regression.
- Correlational Design: Explores relationships without causality assumptions with correlation coefficients and significance tests.
- Descriptive Design: Studies characteristics of a population or phenomenon using statistical inference.
Participant Comparison Levels:
- Between-Subjects Design: Group level comparisons between different treatments.
- Within-Subjects Design: Comparisons of repeated measures from the same participants.
- Mixed (Factorial) Design: Combination of between- and within-subject comparisons.

Operationalizing Variables: Define how variables will be measured.
Levels of Measurement:
- Categorical Data (groupings): Nominal (e.g., gender) or Ordinal (e.g. language ability).
- Quantitative Data (amounts): Interval scale (e.g., test score) or ratio scale (e.g., age).
Importance of Measurement Level: Affects statistical choice and hypothesis testing.
Relevant Participant Characteristics: Often collected alongside primary variables.

Experimental Example:
- Age: Quantitative (ratio)
- Gender: Categorical (nominal)
Correlational Example:
- Parental Income: Quantitative (ratio)
- GPA: Quantitative (interval)