Alternatives to Experimental Research

Chapter Objectives

  • Describe and explain alternatives to Experimental Research: Non-experimental designs, Surveys and Interviews, and Correlational and Quasi-experimental designs.

  • Provide examples and deeper understanding of the alternatives.

Non-Experimental Designs

  • Definition: A method used to study behaviors in natural settings rather than controlled environments.

  • Purpose: Useful for exploring unique or rare occurrences and sampling personal information.

  • External Validity: These designs tend to have higher external validity, making their findings more generalizable to real-world situations.

Key Types of Non-Experimental Designs:

  1. Phenomenology

    • Focuses on describing individuals' immediate experiences.

    • Characteristics: Low manipulations, low impositions.

    • Limitations: Can be subjective and may not lead to generalizable findings.

  2. Case Studies

    • A detailed descriptive record of an individual’s experiences and behaviors kept by an outside observer.

    • Purposes:

      • Generate inferences and hypotheses.

      • Develop therapy techniques.

      • Study rare phenomena.

      • Provide counterinstances to established theories.

      • Serve motivational and persuasive purposes.

  3. Field Studies

    • Conducted outside of laboratory settings in natural environments.

    • Method:

      1. Make observations.

      2. Formulate a question.

      3. Create a hypothesis.

      4. Make predictions.

      5. Test hypothesis through experiments.

      6. Analyze collected data.

  4. Qualitative Studies

    • Based on narrative rather than numerical data, focusing on personal thoughts and feelings.

    • Increasingly significant in psychology, pushing for a paradigm shift.

    • Importance of replicability in qualitative research.

Surveys and Interviews

  • Purpose: Useful for gathering opinions, attitudes, and self-reports about experiences which are hard to observe.

  • Key Aspects of Survey Research:

    • Efficiently collects large amounts of data through methods like questionnaires and interviews (face-to-face, telephone, or focus groups).

    • Generalizability depends on participant sampling procedures.

Constructing Surveys

  • Crafting an effective survey instrument involves careful consideration of content, wording, format, and question placement to avoid bias and improve clarity.

  • Considerations for Survey Items:

    • Start with engaging questions.

    • Avoid sensitive questions at the beginning.

    • Mind wording and context effects which can lead to response bias.

    • Pretesting the survey is crucial to ascertain clarity and flow.

Data Collection Methods

  • Various methods include written questionnaires, mail surveys, telephone surveys, and interviews.

  • Self-Administered Questionnaires: Should ensure anonymity.

  • Differentiate between manifest (literal meaning) and latent (implied meaning) content in responses.

Sampling Methods

  1. Probability Sampling: Ensures each member of the population has a known chance of being selected.

    • Types:

      • Simple Random Sampling

      • Systematic Random Sampling

      • Stratified Random Sampling

      • Cluster Sampling

  2. Non-Probability Sampling: No random selection leading to potential bias.

    • Types:

      • Quota Sampling

      • Convenience Sampling

      • Purposive Sampling

      • Snowball Sampling

Correlational and Quasi-Experimental Designs

  • Correlational Designs: Examine the relationships between preexisting behaviors without manipulating variables, allowing predictions but not causal inferences.

  • Quasi-Experimental Designs: Lack randomization, examining groups based on pre-existing characteristics.

  • Key Types:

    1. Ex Post Facto Studies

    2. Non-equivalent Groups Design

    3. Longitudinal Design

    4. Cross-Sectional Study

    5. Pretest/Posttest

Correlation Statistics

  • Pearson’s Correlation Coefficient (r): Measures strength and direction of relationships between two variables.

  • Coefficient of Determination (r²): Indicates how much variability in one variable can be explained by another.

  • Considerations: Correlation does not imply causation; outliers and range restrictions can distort results.

Review Summary:
Non-experimental, survey, and correlational research methods serve significant roles in gathering and analyzing psychological data without traditional experimental manipulation. Each method offers different advantages and limitations that must be carefully considered in any research design.

References: Myers, A., & Hansen, C. H. (2011). Experimental Psychology. Wadsworth Publishing Company.