Notes on Research Designs

Abstract

• The transcript appears to begin with an overview of research designs in psychology, emphasizing that psychologists test research questions using a variety of methods.

• Key contrast: correlations (observe naturally occurring variables) vs. experiments (manipulate one variable and observe effects on another).

• Other methods include longitudinal and quasi-experimental designs.

• Practical constraints (time, money, resources) influence method choice.

• Note: There is no explicit abstract text provided in the transcript itself; this section is referenced as part of the module structure.

Learning Objectives

• Articulate the difference between correlational and experimental designs.

• Understand how to interpret correlations.

• Understand how experiments help us infer causality.

• Understand how surveys relate to correlational and experimental research.

• Explain what a longitudinal study is.

• List a strength and weakness of different research designs.

Research Designs (overview)

• Psychology uses multiple designs to test hypotheses; no single method fits all questions.

• Two broad types: experimental and correlational research.

• Experimental research involves manipulation of an independent variable (IV) and measurement of a dependent variable (DV).

• Correlational research involves measuring variables as they naturally occur and examining relationships; cannot infer causality from correlation alone.

• Other designs mentioned: longitudinal studies, quasi-experiments, qualitative designs (participant observation, case studies, narrative analysis).

• Practical constraints often necessitate a mix of methods (e.g., surveys vs. longitudinal tracking).

• Analogy: Show me the data (psychology emphasizes data and methodological rigor to support conclusions).

Experimental Research

• Example prompt: Dunn’s happiness study examined whether spending $20 on self vs. others affects happiness.

• Independent Variable (IV): who the money is spent on (self vs. others).

• Dependent Variable (DV): happiness (as measured by self-report).

• Random assignment: crucial for causal inference; participants are assigned to conditions by chance (e.g., coin flip).

• Purpose of random assignment: to make groups similar on all characteristics aside from the manipulated IV, enabling causal attribution to the IV.

• Illustrative analogy: random assignment in sports teams yields roughly equal team quality.

• Other considerations to avoid confounds: control for extraneous variables that may influence DV.

• Double-blind procedure: both participants and experimenters are unaware of condition assignment to prevent placebo effects, participant demand, and experimenter expectancy from biasing results.

• Placebo effect: participants’ beliefs about receiving a treatment can influence outcomes.

• Participant demand: participants behave in ways they think the experimenter expects.

• Experimenter expectations: researchers’ beliefs can influence observations (observer bias).

Correlational Designs

• Correlational research: scientists passively observe and measure phenomena without manipulating variables.

• Purpose: identify patterns/relationships between variables; not to infer causality.

• Scope: typically examines two variables at a time.

• Dunn’s correlational study: spending on others positively correlated with happiness; more money spent on others associated with higher happiness.

• Visual aid: scatterplots are used to depict the relationship between two variables.

• Data point: each dot represents a unit (e.g., a person) with coordinates (x, y) corresponding to two measures (e.g., past-month rating and happiness).

• Correlation coefficient (r): summarizes direction and strength of association.

  • Positive correlation: as one variable increases, the other also increases. Example: Dunn’s study, r = $+0.81$.

  • Negative correlation: as one variable increases, the other decreases. Example: average male height vs pathogen prevalence, r = $-0.83$.

  • Scatterplot pattern: positive correlation forms an upward-right pattern; negative correlation forms a downward-right pattern.

• Strength of the correlation depends on the absolute value |r| (sign indicates direction, magnitude indicates strength):

  • Strong correlation: large |r| (e.g., $|r| ext{ is large}$).

  • Weak correlation: smaller |r| (e.g., $|r| ext{ around } 0.3 ext{–}0.4$).

  • Uncorrelated: r ≈ 0.

• Important caveat: correlation does not imply causation. There can be third variables or bidirectional causality (e.g., happiness might influence generosity, or a third variable like wealth could influence both).

• Perfect or near-perfect correlations can be observed (e.g., age vs year of birth shows a near-perfect negative relation in the example given).

• Visual and interpretive notes:

  • Strong correlation example: Dunn’s happiness vs spending on others (r = $+0.81$) shows a clear, tight pattern.

  • Weak correlation example: value of happiness vs GPA (r = $-0.32$) shows more scatter and exceptions.

  • Negative correlation example: height vs pathogen prevalence (r = $-0.83$) shows a clear inverse pattern.

• Limitations of correlational designs:

  • Cannot determine direction of causality.

  • Possibility of third-variable explanations (confounds).

  • Outliers or restricted ranges can affect r.

Qualitative Designs

• Qualitative designs allow study topics difficult to manipulate experimentally (e.g., income effects) or require in-depth understanding of individuals.

• Participant observation: researcher embeds in a group to study dynamics; example from Festinger, Riecken, and Schachter (1956) studying a cult by posing as members.

  • Note: participants often know the researcher is studying them.

• Case studies: intensive examination of a single case or context; useful for rare or unique cases (e.g., brain injuries).

  • Limitation: generalizability may be limited; findings may not apply to broader populations.

• Narrative analysis: study of stories/personal accounts to examine themes, structure, and dialogue; data can be written, audio, or video; emphasizes how people convey their experiences.

• Qualitative methods provide rich, contextual insights but may face questions about generalizability and objectivity.

Quasi-Experimental Designs

• Quasi-experimental designs study variables like marriage effects on happiness without random assignment.

• Key distinction: independent variable exists (e.g., marital status) but there is no random assignment to conditions.

• Implication: causal inferences are weaker because preexisting differences between groups may confound results.

• Example: comparing happiness levels of married vs. single individuals using existing groups rather than randomly assigning people to marry or stay single.

• Real-world classroom analogy: comparing two professors where students self-select into courses; possible confounds like prior ability or motivation.

• Conceptual difference: experimental designs include random assignment to conditions; quasi-experimental designs do not, making causal claims less robust.

Longitudinal Studies

• Longitudinal studies track the same individuals over time, sometimes for many years or decades.

• Strengths: provide valuable evidence for testing theories, observe temporal sequences, and study development/trends.

• Example: a long-running study of more than 20,000 Germans followed for two decades; Rich Lucas (2003) found that marriage is associated with higher happiness over time.

• Costs/limits: can be very costly and time-consuming; participant attrition is a concern; data management is complex.

Surveys

• Surveys enable data collection from large samples at relatively low costs and time commitments.

• Can be used for correlational research and can also be embedded in experiments.

• Examples from transcript:

  • King and Napa (1998): survey stimuli portraying happy vs. unhappy individuals; results: happy individuals were judged more likely to go to heaven (IV = happiness portrayal; DV = likelihood of heaven).

  • Harker and Keltner (2001): smiling in women’s yearbook photos correlated with being married 10 years later.

• Surveys are versatile but do not guarantee causality; they often rely on self-report and can be susceptible to social desirability biases.

Tradeoffs in Research

• Practical considerations influence method choice: time, money, and resource availability.

• Researchers may start with surveys for broad data and then follow up with longitudinal or experimental methods if needed and feasible.

• Ethical considerations are crucial when planning research (e.g., cannot experimentally manipulate brain injuries or traumatic experiences).

• Some topics (war, long-term isolation, abusive parenting, prolonged drug use) pose ethical challenges for experimental manipulation.

• The ethics of a study shapes design decisions and feasibility of causal inferences.

Research Methods: Why You Need Them

• Media coverage of research can misrepresent methodology; correlational findings are often misinterpreted as causal.

• The strength of a scientific finding lies in the rigor of its methodology.

• Understanding research methods enables critical consumption of information and better problem solving in various domains.

Outside Resources

• Harker & Keltner study of yearbook photographs and marriage: http://psycnet.apa.org/journals/psp/80/1/112/

• Rich Lucas’s longitudinal study on the effects of marriage on happiness: http://psycnet.apa.org/journals/psp/84/3/527/

• Dunn, Aknin, & Norton (Spending money on others promotes happiness): https://www.sciencemag.org/content/319/5870/1687.abstract

• What makes a life good? (King & Napa/PSP): http://psycnet.apa.org/journals/psp/75/1/156/

Discussion Questions

• What are some key differences between experimental and correlational research?

• Why might researchers sometimes use methods other than experiments?

• How do surveys relate to correlational and experimental designs?

Vocabulary

• Confounds: Factors that undermine the ability to draw causal inferences from an experiment.

• Correlation: Measures the association between two variables, or how they go together.

• Dependent variable: The variable the researcher measures but does not manipulate in an experiment.

• Experimenter expectations: When the experimenter’s expectations influence the outcome of a study.

• Independent variable: The variable the researcher manipulates and controls in an experiment.

• Longitudinal study: A study that follows the same group of individuals over time.

• Operational definitions: How researchers specifically measure a concept.

• Participant demand: When participants behave in a way they think the experimenter wants them to behave.

• Placebo effect: When receiving special treatment or something new affects human behavior.

• Quasi-experimental design: An experiment that does not require random assignment to conditions.

• Random assignment: Assigning participants to receive different conditions of an experiment by chance.

References

• Chiao, J. (2009). Culture–gene coevolution of individualism – collectivism and the serotonin transporter gene. Proceedings of the Royal Society B, 277, 529-537. doi: 10.1098/rspb.2009.1650

• Dunn, E. W., Aknin, L. B., & Norton, M. I. (2008). Spending money on others promotes happiness. Science, 319(5870), 1687–1688. doi: 10.1126/science.1150952

• Festinger, L., Riecken, H.W., & Schachter, S. (1956). When prophecy fails. Minneapolis, MN: University of Minnesota Press.

• Harker, L. A., & Keltner, D. (2001). Expressions of positive emotion in women's college yearbook pictures and their relationship to personality and life outcomes across adulthood. Journal of Personality and Social Psychology, 80, 112–124.

• King, L. A., & Napa, C. K. (1998). What makes a life good? Journal of Personality and Social Psychology, 75, 156–165.

• Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (2003). Re-examining adaptation and the setpoint model of happiness: Reactions to changes in marital status. Journal of Personality and Social Psychology, 84, 527–539.

Authors

• Christie Napa Scollon: Associate professor of psychology at Singapore Management University; Ph.D. in social/personality psychology from UIUC; research on cultural differences in emotions and life satisfaction.

License and Citations

• Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

• How to cite: Scollon, C. N. (2025). Research designs. In R. Biswas-Diener & E. Diener (Eds), Noba textbook series: Psychology. Champaign, IL: DEF publishers. Retrieved from http://noba.to/acxb2thy

Appendix: Section Map (from Page 20)

• Abstract

• Learning Objectives

• Research Designs

  • Experimental Research

  • Other considerations

• Correlational Designs

• Qualitative Designs

• Quasi-Experimental Designs

• Longitudinal Studies

• Surveys

• Tradeoffs in Research

• Research Methods: Why You Need Them

• Outside Resources

• Discussion Questions

• Vocabulary

• References

• Authors

• Creative Commons License

• About

• Privacy

• Terms

• Licensing

• Contact

• © 2025 Diener Education Fund