Personality and Individual Differences: Psychological Measurement 1

Introduction to Psychological Measurement

Joachim Stover introduces the scientific study of personality and individual differences, focusing on Chapters 1 and 2 of the Ashton (2018) textbook. The lecture supplements the book, emphasizing the importance of reading the assigned chapters carefully.

Core Concepts

This section will cover basic concepts of psychological measurement, simple statistical ideas, assessing the quality of measurement (reliability and validity), and methods of measurement.

Why is Psychological Measurement Important?

Psychological measurement is crucial for quantifying personality traits, relating these measurements to each other, ensuring the measurement of meaningful characteristics, and comparing individuals meaningfully.

Quantitative Nature of Psychology

Psychology relies on quantitative data to relate numbers and understand relationships between variables.

Meaningful Characteristics

Measurements should focus on characteristics that are relevant and widely applicable, not idiosyncratic.

Accurate Measurement

Ensure that the measurement accurately captures the intended trait (e.g., measuring extraversion and not neuroticism).

Meaningful Comparisons

The goal is to make meaningful comparisons among individuals (e.g., Frank is more extroverted than Jake, who is more extroverted than Lisa).

Statistical Analysis

Measurements enable the calculation of relationships between variables and the strength of these relationships (e.g., the correlation between extraversion and positive emotions).

Measurement Peculiarities in Personality

Unique considerations when measuring personality traits.

Absence of Meaningful Zero Level

Personality traits typically lack a true zero point; individuals do not have "zero extraversion."

No Absolute Amounts

There are no absolute units for personality traits; a score of 50 for extraversion is meaningful only in relation to other scores.

Ratios are Not Used

Ratios are generally avoided; we don't say someone is "twice as extroverted" as another person.

Interval Data Approximation

Well-designed personality measures aim to achieve approximately equal intervals between scores, corresponding to similar differences in the measured construct (e.g., a 20-point difference should represent roughly the same difference in the trait level).

Statistical Ideas in Measurement

Overview of essential statistical concepts for understanding personality and individual differences.

Levels of Measurement

Four basic levels of measurement in statistics:

Nominal

Data can only be categorized (e.g., smoker vs. non-smoker, resident of Kent vs. Sussex). No meaningful comparisons.

Ordinal

Data can be ranked (e.g., first place, second place, third place). The magnitude of differences is unknown.

Interval

Equal intervals between measurements correspond to approximately equal differences in the construct measured. This is the target in personality measurement.

Ratio

Data can be ranked, evenly spaced, and have a natural zero (e.g., money). Rare in personality measurement.

Application to Personality Data

Data from personality scales are generally between ordinal and interval levels. Statistical analyses typically treat them as interval data for approximation purposes.

Standard Scores

Converting raw scores into standard scores for interpretation.

Need for Standardization

Raw scores from personality scales must be converted into standard scores to compare them to others.

Method

Subtract the mean from the raw score and divide by the standard deviation.

Interpretation

The resulting score indicates how far a person is from the mean, allowing placement on a normal distribution.

Example: IQ Scores

Standardized IQ tests have a mean of 100 and a standard deviation of 15. A score of 130 is two standard deviations above the mean, placing the person in the top 2%.

Normal Distribution

Most personality scores approximate a normal distribution, with most people having medium scores and fewer at the extremes.

Correlation Coefficient

Understanding relationships between variables.

Definition

Correlation coefficient (r) indicates the strength and direction of the relationship between two variables.

Types of Correlations

Positive Correlation

As one variable increases, the other also increases.

Negative Correlation

As one variable increases, the other decreases.

Range

Correlations range from -1.00 to +1.00.

r = 1.00

Indicates a perfect positive correlation (identical variables).

r = -1.00

Indicates a perfect negative correlation (opposite variables).

Scatter Plots

Visual representation of correlations, with each dot representing a person's scores on two variables.

Examples

Positive Correlation

Extraversion and positive emotion (e.g., r = 0.5).

Negative Correlation

Neuroticism and self-esteem (e.g., r = -0.5).

Zero Correlation

No relationship between variables.

Benchmarks for Interpretation

Guidelines for interpreting the magnitude of correlations.

Large/Strong Correlations

|r| \geq 0.40

Moderate Correlations

0.20 \leq |r| < 0.40

Small/Weak Correlations

0 < |r| < 0.20

No Correlation

r \approx 0

Importance of Weak Correlations

Even small correlations can be significant, depending on the variables (e.g., conscientiousness and health).

Sample Representativeness and Size

Considerations for interpreting empirical findings.

Sample Representativeness

Samples should reasonably represent the population of interest (e.g., studying car salespeople requires studying car salespeople).

Common Issues

Reliance on Psychology Undergrad Students

Many findings are based on psychology students, limiting generalizability.

WEIRD Samples

Western, Educated, Industrialized, Rich, Democratic societies—studies often based on these samples.

Restricted Variance

Ensuring a wide range of scores in the constructs being studied.

Sample Size

A large sample is needed for statistical power to detect correlations.

General Guideline

Samples of 250 people or more provide stable estimates and sufficient statistical power.

Assessing Quality of Measurement

Evaluating reliability and validity.

Reliability

The extent to which a measure is consistent and exact.

Internal Consistency Reliability

The extent to which items in a measure assess similar things. Measured by Cronbach's alpha ([alpha]), which should be above 0.70.

Inter-Rater Reliability

The extent to which different observers or raters agree.

Test-Retest Reliability

The extent to which scores are consistent over time.

Example

The Relationship Between the Number of Items and Internal Consistency
This demonstrates that more items and items that show positive intercorrelations lead to higher reliability of the measure.

Validity

The extent to which a test measures what it claims to measure.

Content Validity

The extent to which the measure represents all aspects of the construct.

Construct Validity
Convergent Validity

The measure correlates highly with other measures of the same construct.

Discriminant Validity

The measure does not correlate with measures of different constructs.

Criterion Validity

The measure predicts relevant behaviors.

Summary of Key Concepts

  • Understanding psychological measurement and basic statistics.

  • Levels of measurement, standard scores, and normal distribution.

  • Correlations, sample representativeness, and sample size.

  • Reliability and validity.