Research Methods Chapter 2 PDF

A correlation measures the strength and direction of the relationship between two variables. It is represented by the correlation coefficient (r), which ranges from -1.00 to +1.00: ​ Positive Correlation (+): As one variable increases, the other tends to increase. ○​ Example: Hours...

A correlation measures the strength and direction of the relationship between two variables. It is represented by the correlation coefficient (r), which ranges from -1.00 to +1.00: ​ Positive Correlation (+): As one variable increases, the other tends to increase. ○​ Example: Hours studied and test scores (r = +0.8). ​ Negative Correlation (-): As one variable increases, the other tends to decrease. ○​ Example: Stress levels and quality of sleep (r = -0.6). ​ Zero Correlation (0): No relationship exists between the variables. ○​ Example: Shoe size and intelligence (r = 0). Strength of Correlations - ​ Weak: 0 to ±0.3 ​ Moderate: ±0.3 to ±0.6 ​ Strong: ±0.6 to ±1.0 What is a p-Value? - The p-value indicates whether a correlation is statistically significant or likely due to random chance. ​ p < 0.05: Statistically significant; less than 5% chance results are due to chance. ​ p < 0.01: Highly significant; less than 1% chance results are due to chance. ​ p < 0.001: Very highly significant; less than 0.1% chance results are due to chance. ​ p > 0.05: Not significant; results may be due to random variability. Combining Correlation and p-Value - ​ A weak correlation (e.g., r = 0.12) can still be significant if p < 0.05, especially with a large sample size. ​ A strong correlation (e.g., r = 0.8) might not be significant if p > 0.05, often due to small sample size or high variability. ​ Key Takeaway: Assess both r (strength) and p-value (reliability) to interpret data accurately. Asterisk Notation in Statistical Significance - Asterisks (***, **, *) indicate levels of statistical significance in tables or results: 1.​ One Asterisk (*): ○​ p < 0.05 (Significant) ○​ Minimum level of statistical confidence. 2.​ Two Asterisks ()**: ○​ p < 0.01 (Highly Significant) ○​ Stronger evidence of significance. 3.​ Three Asterisks (*)**: ○​ p < 0.001 (Very Highly Significant) ○​ Highest level of statistical confidence Self-Report Questionnaires - ​ Definition: A common method to measure personality, relying on individuals to self-report responses.  ​ Format: Participants typically rate statements (e.g., True/False, Likert scales).​ Example:​ 1 = Strongly Disagree, 5 = Strongly Agree ○​ Example items: 1.​ I like cats. 2.​ Cats freak me out. Benefits of Self-Report Questionnaires - 1.​ Ease of Administration: Simple for researchers to distribute. 2.​ Convenience: Requires minimal effort from participants. 3.​ Efficiency: Gather large amounts of data quickly. 4.​ Scoring Simplicity: Straightforward to analyze results. 5.​ Random Sampling: Facilitates gathering data from diverse populations. Drawbacks of Self-Report Questionnaires - 1.​ Social Desirability Bias: Participants may respond in socially favorable ways. 2.​ Acquiescence Response Set: Tendency to agree with all statements. 3.​ Reverse-Scored Items: Helps counteract response biases but can confuse participants Alternatives to Self-Report Questionnaires - 1.​ Informant Reports: Useful for traits difficult to self-assess (e.g., ADHD behavior scales). 2.​ Clinical Interviews: Structured or semi-structured; useful for extreme traits, not everyday assessments. 3.​ Behavioral Measurements: Observed actions, either self-reported or recorded by an observer. 4.​ Archival/Life Outcomes Data: Data from historical records (e.g., school or medical records). 5.​ Projective Tests: Use ambiguous stimuli to assess implicit motives. 6.​ Physiological Measures: Bodily responses reveal underlying reactions (e.g., fMRI, cortisol, heart rate). Evaluating Personality Scales 1.​ Reliability: Consistency of the measure. ○​ Internal Reliability: Items on the scale measure the same concept. ​ Cronbach’s Alpha: Ideal range 0.6–0.7 (too high >0.9 indicates narrowness). ○​ Test-Retest Reliability: Similar results over time. ​ Measured by Pearson’s correlation coefficient (r) and p-value. ○​ Intercoder (Interrater) Reliability: Agreement between independent coders. ​ Cohen’s Kappa: Ideal rates: ​ Moderate correlation: 0.40–0.59 ​ High correlation: 0.60–0.79 ​ Alpha should be ~0.6 or above, representing a high correlation 0.9 and greater, means that we are no longer capturing distinct aspects of e.g., traits, and instead measuring something very narrow regarding traits in general, e.g., 2 questions designed to measure extroversion, if the correlation is 0.9 or higher, means the questions are not really different from each other 2.​ Validity: Scale measures what it’s supposed to measure. ○​ Face Validity: Appears to measure the intended concept.  ○​ Predictive Validity: Predicts relevant outcomes (e.g., conscientiousness predicts academic performance). ○​ Convergent Validity: Correlates with other measures of the same construct. ​ Example: Extraversion in two different personality inventories. ○​ Discriminant Validity: Does not correlate with unrelated constructs. Triangulation ​ Combining multiple methods to measure personality. ​ Ensures convergent validity by correlating results across different measures. Study Design 1.​ Correlational Studies: ○​ Examines relationships between variables. ○​ Limitations: Cannot establish causation; susceptible to confounding variables. 2.​ Experimental Studies: ○​ Identifies cause-and-effect relationships. ○​ Key Features: ​ Random Assignment: Ensures groups are comparable. ​ Independent Variables (IVs) and Dependent Variables (DVs): Allows testing effects. ○​ Example: Testing catharsis vs. cognitive neoassociation theory. Descriptive Statistics - ​ Mean: The average of all data points. ​ Median: The middle value in a data set. ​ Mode: The most frequently occurring value Academic badges - Open Data: downloadable original data can be personally viewed and analyzed Open Materials: copies of measurement tools are publicly available Pre-Registered: data in a database so a study can be duplicated