RMS final exam study guide

Design

1. Confounding Variables: 

• These are variables that influence both the IV and DV, making it unclear whether changes in the DV are due to the IV or the confound.

• Example: Studying the effect of exercise on weight loss while ignoring diet, which could also impact weight loss.

2. Internal Validity Threats: 

• Selection Bias: Non-random assignment of participants to groups.

• History: Events outside the experiment influencing results.

• Maturation: Natural changes in participants over time.

• Regression to the Mean: Extreme scores tend to normalize over time.

• Attrition: Participants dropping out affects group composition.

• Testing Effects: Practice or fatigue from repeated testing.

• Instrumentation: Changes in measurement tools or procedures.

3. Independent Variables (IV): 

• The variable that is manipulated to observe its effect on the DV.

• Example: Amount of sleep (4, 6, or 8 hours).

4. Dependent Variables (DV): 

• The variable that is measured; its value depends on the IV.

• Performance on a memory test.

5. Independent vs. Dependent Groups:

   • Independent (Between-subjects): Different participants in each group. Each experiences a single condition.

   • Dependent (Within-subjects): Same participants in all conditions.

6. Random Assignment vs. Random Selection:

   • Random Assignment: Ensures equal chance of participants being in any group, reducing bias within the study.

   • Random Selection: Increases external validity by ensuring the sample represents the population.

7. Manipulation Checks: Measures used to confirm that the IV was effectively manipulated.

• Example: If testing stress effects, measure participants’ stress levels to confirm the manipulation was effective.

8. Demand Characteristics: Cues that reveal the experiment’s purpose, causing participants to alter their behavior.

• Participants alter their behavior based on perceived expectations.

• Minimize by using double-blind designs or deception.

General Statistics

1. Frequency Tables: Organize data into categories and show counts or percentages.

   • When to use: Categorical data.

2. Bar Graphs: Compare categories using bars.

   • When to use: Categorical data.

3. Histograms: Display frequency distributions of continuous data.

   • When to use: Continuous data.

4. Line Graphs: Show trends over time or continuous variables.

   • When to use: Time-series or interval data.

5. Measures of Central Tendency: 

• Mean: Average.

• Median: Middle score.

• Mode: Most frequent score.

• Variability: Spread of scores (range, variance, standard deviation).

6. Types of Distributions: Normal, skewed (positive/negative), bimodal, etc

7. Characteristics of a Normal Distribution: Symmetrical, bell-shaped, mean = median = mode.

Z-scores: Standardized scores indicating how far a score is from the mean in standard deviations. FORMULA

• Why useful: Compare scores across different scales.

8. Sampling Distributions: Distribution of a statistic (e.g., mean) across samples.

   • Why use: Infer population parameters.

9. Descriptive vs. Inferential Statistics:

   • Descriptive: Summarize data (e.g., mean).

   • Inferential: Conclude a population.

10. Null vs. Alternative Hypothesis:

    • Null: No effect. (H0)

    • Alternative: Some effect exists. (H1​)

11. Criterion Values/Alpha Levels: Threshold (e.g., 0.05) for rejecting the null hypothesis.

12. Representativeness of a Sample: The degree to which the sample reflects the population.

13. Reject vs. Fail to Reject:

    • Reject: Evidence supports the alternative hypothesis.

    • Fail to Reject: Insufficient evidence to support the alternative.

14. One-Tailed vs. Two-Tailed:

    • One-tailed: Predicts direction.

    • Two-tailed: Tests for any difference.

z-Test

1. When to Run a z-Test: Compare the sample mean to the population mean with known population SD.

2. z Obtained vs. z Critical:

   • z Obtained: Calculated z-score.

   • z Critical: Cutoff value based on alpha.

3. Standard Error of the Mean (SEM): SEM=σnSEM=n​σ​, measures variability in the sample mean.

4. Type I and Type II Errors:

   • Type I: Rejecting true null hypothesis (false positive).

   • Type II: Failing to reject false null hypothesis (false negative).

5. Power: Probability of correctly rejecting the null.

   • Influenced by: Sample size, effect size, and error variability.

One-Sample t-Test

1. When to Run: Compare the sample mean to the population mean when the population SD is unknown.

2. t-Test vs. z-Test:

   • z-Test: Known population SD.

   • t-Test: Estimate SD from the sample.

3. t-Distribution vs. Sampling Distributions: t-distribution is wider and accounts for small sample sizes.

4. Assumptions: Independence, normality, interval/ratio data.

5. Effect Size: Quantifies magnitude of difference (e.g., Cohen’s d).

6. Confidence Interval: Range likely to include population mean.

7. APA Style Reporting: DO THIS

Two-Sample t-Test

1. When to Run: Compare the means of two groups.

2. Assumptions: Independence, normality, equal variances (for independent samples).

3. Between vs. Within Groups Variability:

   • Between: Differences between group means.

   • Within: Variability within each group.

4. Effect Size: Quantifies group differences.

One-Way ANOVA

1. Difference from t-Test: Tests 3+ group means simultaneously.

2. Why Not Multiple t-Tests: Increases Type I error rate.

3. Assumptions: Normality, independence, equal variances.

4. Null vs. Alternative:

   • Null: All group means equal.

   • Alternative: At least one differs.

5. Between vs. Within Groups Variance: Variance due to IV vs. random error.

6. F-Distribution: Positively skewed; used in ANOVA.

7. Posthoc Tests: Run when the F-test is significant to identify specific differences.

Two-Way ANOVA/Factorial Designs

1. Assumptions: Independence, normality, equal variances.

2. Notation: e.g., 2x3 factorial design (2 IVs, one with 2 levels, one with 3 levels).

3. Why Use Factorial Designs: Examine interactions between IVs.

4. Main Effects vs. Interactions:

   • Main Effects: Independent effect of each IV.

   • Interactions: Combined effect of IVs.

5. APA Reporting: Include F(df between, df within) = value, p < .05, partial eta squared. DO THIS