Study Guide for PSYC 200: Methods in Psychological Science - Exam 2
PSYC 200: Methods in Psychological Science Study Guide – Exam 2
Frequency Distribution
- Definition: A frequency distribution is a graphical representation used to visualize the frequency of specific values or categories occurring in a dataset.
- What to look for in a frequency distribution:
- Shape: Can be bimodal, symmetrical, or unimodal.
- Outliers: Identifies unusual or extreme values in the data.
- Types of Variables:
- Quantitative: Numerical values that can be measured.
- Categorical: Qualitative descriptors that classify observations.
- Skewness Statistic:
- Negative Skew: Tail is longer on the left side; mode < median < mean.
- Positive Skew: Tail is longer on the right side; mean > median > mode.
- Zero Skew: Symmetrical distribution; mean = median = mode (unimodal).
Measures of Central Tendency
- Measures Defined:
- Mean: The average calculated by adding all values and dividing by the number of values.
- Formula: extMean=extNumberofScoresextSumofScores
- Median: The middle value when all scores are arranged in order.
- If there is an even number of scores, median is the average of the two middle scores.
- Mode: The most frequently occurring value in a dataset.
- Calculating the Mean:
- For weighted means with two sample scores, the calculation involves:
extWeightedMean=n1+n2M(n1)+M(n2),
where n1 and n2 are the sample sizes from the two groups.
Variability
- Definition: Variability refers to how spread out the values are in a dataset. It is important because measures of central tendency alone may not reflect the true nature of the data, particularly if outliers are present.
- Conceptual Understanding:
- High variability indicates that data points are widely spread, while low variability indicates that they are clustered closely around the mean.
- Error Bars Relation to Variability:
- Error bars on graphs indicate the confidence interval; larger error bars reflect greater uncertainty in the mean.
- Certainty in Results:
- Results from studies with low variability are considered more certain, especially in treatment vs. control scenarios.
- Calculating Variability:
- Find the mean.
- Subtract the mean from each score.
- Square those deviations.
- Calculate the average of the squared deviations (variance).
Observational Research
- Concept: Observational studies involve observing subjects in their natural environment without manipulation to develop theories before experimentation.
- Pros and Cons:
- Pros: High external validity; findings may generalize better to real-world settings.
- Cons: Potential biases such as observer bias, expectancy effects, and reactivity (e.g., demand effects, reactance).
- Examples of Observational Research:
- Chair experiment highlights natural behavior under observation.
- Bias Issues: Researchers must be cautious not to interpret data to fit preconceived ideas; interrater reliability can be a method to counteract observer bias.
- Ethics: Observational studies must be conducted in public spaces to respect privacy.
- Physical Traces: Refers to using physical remnants (e.g., discarded soda cans) to gather data.
- Archival Records Issues:
- Selective Deposit: Not all data gets recorded, potentially skewing results.
- Selective Survival: Data may survive while some data may not be accessible.
- Notable Study:
- Rice vs. Wheat Farming (Talhelm et al., 2018): The research was observational, analyzing how individualism correlates with wheat farming, demonstrating both pros and cons of observational studies. The researchers took steps to counteract biases in their findings.
- Self-report vs. Observation:
- Discrepancies may arise (e.g., individuals may act differently than they report). An example includes historical prejudice where self-reported attitudes differ from observed behaviors.
Correlational Research
- Assessing Associative Claims: Important to examine various types of validity when evaluating correlational research.
- Construct Validity: Evaluates whether a test or tool accurately measures the concept intended. Includes:
- Convergent Validity: Test should correlate with similar constructs.
- Face Validity: The measure appears to be valid at face value.
- Concurrent Validity: Correlation with other measures taken at the same time.
- Statistical Validity: Assesses the appropriateness of statistical conclusions. Principles include:
- Effect Size: The magnitude of the relationship between variables.
- Precision: How accurately the data reflects the true relationship.
- Replication: Consistency across studies and sample populations.
- Restriction of Range: Limited data may skew results.
- Curvilinearity: Non-linear relationships may exist.
- Internal Validity: Focuses on the extent to which causal conclusions can be drawn. Essential criteria include:
- Covariance: Demonstrates the relationship between variables.
- Temporal Precedence: The cause must precede the effect.
- Internal Consistency: Eliminates confounding variables that could affect results.
- Examining Third Variable Problems: Must be able to identify third variables that may impact both primary variables in a correlation.
- Scatterplot Analysis: Be prepared to identify third variables visually represented in scatterplots.
- External Validity: Addresses the generalizability of findings beyond the study sample.
- Moderators: Variables that may affect the strength or direction of the relationship between the independent and dependent variable.
- Identifying Moderating Effects: Consider demographic factors like gender that may influence the results.
Experimental Research
- Key Concepts:
- Random Assignment: Participants are randomly assigned to groups to reduce bias. Essential for supporting causal claims.
- Between-Subjects Design: Different subjects are assigned to different groups, giving pros and cons relating to sample size and variability management.
- Within-Subjects Design: The same subjects participate in all conditions, which can be effective but also has its own methodological challenges.
- Post-Test vs. Pre-Test/Post-Test Designs: Understand the differences and when to apply each method.