Research Methods and Statistics in Introductory Psychology

Research Methods in Psychology

Overview of Psychological Research Methods

Psychological research utilizes a variety of methods to investigate questions about behavior, cognition, and emotion. Understanding these methods is essential for interpreting findings and designing studies.

• Descriptive Methods: Used to observe and describe behavior without manipulating variables.

• Experimental Methods: Involve manipulation of variables to determine cause-and-effect relationships.

• Correlational Methods: Examine relationships between variables without manipulation.

• Longitudinal Methods: Study the same participants over time to observe changes and long-term effects.

Experimental Methods

Experimental methods are designed to test hypotheses about causal relationships by manipulating an independent variable (IV) and measuring its effect on a dependent variable (DV).

• Independent Variable (IV): The variable that is manipulated by the researcher.

• Dependent Variable (DV): The variable that is observed or measured to assess the effect of the IV.

• Key Requirements for Experiments:

  • Random assignment to groups

  • Control condition (as similar to experimental condition as possible except for the IV)

  • No pre-existing differences between groups

• Importance: Without random assignment and control conditions, causality cannot be inferred.

Examples of Experimental Questions

• Does playing video games lead to aggressive behavior?

• Do people remember information better if they cram or space out their study sessions?

• Can children wait for a bigger reward if the reward is explained?

Experimental Methods: Validity and Bias

Validity refers to the accuracy and generalizability of research findings. Bias can distort results and must be controlled.

• Internal Validity: The degree to which changes in the DV are due to the IV and not other factors.

• External Validity: The degree to which findings generalize to the real world.

• Counter Blindness and Bias: Use of double-blind procedures to prevent experimenter and participant expectations from influencing results.

Longitudinal Methods

Longitudinal studies follow the same sample of people over time, measuring the strength and direction of relationships between variables.

• Can be correlational or involve experimental manipulation.

• Examples:

  • Does childhood self-control predict achievement, quality of relationships, and financial well-being over time?

  • How does stress affect health over time?

  • How long do anti-depressants work?

Examples of Research Methods in Practice

• Children with more self-control were found to be happier adults (Richards, 2011).

• People who think they know more about something know less (Dunning and Krueger).

• College students randomly assigned to give advice, rather than receive, spent more time on homework (Eskreis-Winkler, Fishbach, & Duckworth, 2018).

Strengths and Weaknesses of Research Methods

Comparison Table

The following table summarizes the strengths and weaknesses of major research methods in psychology:

Method	Strength	Weakness
Observational methods	Real-world behavior	Sample can be biased
Case study methods	In-depth analysis	Limited generalizability
Correlational methods	Flexible	Can't infer causality
Experimental methods	Causality!	Limited ecological validity
Longitudinal methods	Establish order	Attrition

Statistics in Psychology

Descriptive and Inferential Statistics

Statistics are mathematical methods for describing and interpreting data in psychological research.

• Descriptive Statistics: Summarize basic features of data from a given sample (e.g., mean, median, mode).

• Inferential Statistics: Draw conclusions about the meaning of the data and make predictions about populations.

Recognizing Patterns in Data

Patterns in data are often visualized using distributions and measures of variability.

• Variability: Refers to how spread out scores are around the mean.

• Low Variability: Scores are clustered closely around the mean.

• High Variability: Scores are spread out from the mean.

Normal Distribution and the 68-95-99.7 Rule

Many psychological variables are normally distributed, meaning scores are symmetrically spread around the mean.

• 68-95-99.7 Rule: In a normal distribution:

  • 68% of scores fall within 1 standard deviation (SD) of the mean

  • 95% within 2 SDs

  • 99.7% within 3 SDs

Measures of Central Tendency

• Mean: Arithmetic average of values

• Mode: Most commonly occurring value

• Median: Midpoint value when data points are ranked (50th percentile)

Measures of Variability

• Range: Span of scores from lowest to highest

• Variance: Average squared deviation from the mean $\text{Variance} = \frac{\sum (x_i - \bar{x})^2}{N}$

• Standard Deviation (SD): Average deviation of scores from the mean (in same unit as mean) $\text{SD} = \sqrt{\text{Variance}}$

Inferential Statistics

Correlations

Correlational methods assess the relationship between two variables, measured by Pearson's r.

• Pearson's r: Measures the strength and direction of a linear relationship between two variables. $r$ ranges from -1 (perfect negative) to +1 (perfect positive), with 0 indicating no correlation.

• Effect Size: Small = .10, Medium = .30, Large = .50

Comparing Groups: t-test and ANOVA

• t-test: Compares means of two groups relative to their standard deviation. $t = \frac{\bar{x}_1 - \bar{x}_2}{\sqrt{\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}}}$

• ANOVA (Analysis of Variance): Compares means of more than two groups. One-way ANOVA compares means of >2 groups relative to their variance.

• Cohen's d: Measures effect size between two groups. $d = \frac{M_1 - M_2}{SD_{pooled}}$ Small effect < .20, Medium = .50, Large > .80

Hypothesis Significance Testing and p-values

Significance testing helps researchers assess whether observed effects are likely to be true in the population.

• Null Hypothesis (H0): Assumes no effect or relationship (e.g., population correlation = 0).

• p-value: Probability of obtaining a result as extreme as the observed one, assuming the null hypothesis is true. If $p < .05$, the result is considered statistically significant and the null hypothesis is rejected.

• Interpretation: A p-value of .03 means there is a 3% chance of observing the result if the null hypothesis is true.

Key Takeaways

• Statistical significance does not guarantee practical significance.

• Results may still be wrong due to sampling error or bias.

• Always consider effect size and validity when interpreting findings.

Additional info:

• Some examples and references (Richards, Dunning and Krueger, Eskreis-Winkler et al.) were inferred to illustrate application of methods.

• Validity definitions and bias control (double-blind procedures) were expanded for clarity.

• Equations for variance, standard deviation, t-test, and Cohen's d were added for completeness.