JR

Chapter 1-6 Psychology: Introduction to Research Methods (Vocabulary)

Correlations

  • A correlation is a statistical measure of the strength of a relationship between two variables.

  • Two key directions of relationship:

    • Positive correlation: as one variable increases, the other also increases, or they move in the same direction.

    • Negative correlation: as one variable increases, the other decreases, or they move in opposite directions.

  • Zero correlation: no relationship between the variables.

  • Correlation coefficient (r): a value on a scale from -1 to 1 that quantifies the strength and direction of the relationship.

    • The symbol used is r with -1 \le r \le 1 .

    • The closer |r| is to 1, the stronger the relationship; the closer r is to 0, the weaker the relationship.

  • Data visualization: correlations are often displayed on a scatter plot.

  • Uses in psychology:

    • Allows looking for patterns or relationships without manipulating variables (unlike experiments).

    • Useful when ethical or practical constraints prevent experimentation.

  • Data collection methods for correlational studies: surveys, naturalistic observation, archival data.

  • Reasons for using correlational methods:

    • Some questions cannot be ethically studied with experiments (e.g., effects of divorce on children’s self-esteem).

    • Some variables cannot be directly manipulated (e.g., height).

    • Large datasets can make correlation analysis efficient and reveal patterns.

  • Example patterns:

    • Positive: more study time → higher grade; fewer absences → fewer discipline referrals.

    • Negative: more absences → lower grade; less sleep → more car accidents.

  • Important caution: correlation does not imply causation.

    • Example where both variables rise together does not mean one causes the other (ice cream sales and violent crime example).

Correlation vs Causation and Related Issues

  • Illusory correlations: perceiving a relationship where none exists, a bias that can be reinforced by selective observation.

  • Regression toward the mean: extreme scores tend to move toward the average on subsequent measurements, which can create the illusion of a relationship.

  • These phenomena can create or exaggerate perceived links between variables.

Experiments: Core Concepts

  • Experiments are the only research method that can establish a cause-and-effect relationship.

  • Definition: an experiment is a research method in which an investigator manipulates one or two factors (variables) to observe the effects on behavior or mental processes.

  • Experimental findings are often generalizable to a larger population, though experiments can be time-consuming and expensive and can be influenced by extraneous factors (confounding variables).

  • Key terms:

    • Hypothesis: a prediction about the relationship between variables.

    • Independent variable (IV): the factor that is manipulated or controlled by the researcher.

    • Dependent variable (DV): the outcome measured or observed.

    • Control group: participants who do not receive the manipulated variable.

    • Experimental group: participants who do receive the manipulated variable.

  • Example to identify IV and DV:

    • If testing whether a new medication affects hair growth in men, the DV is the amount of hair growth, and the IV is the medication.

  • Random assignment vs random sampling:

    • Random assignment: participants are assigned to control or experimental groups randomly, used in experiments.

    • Random sampling: participants are randomly selected from a population for a study, used in surveys.

Experimental Design and Control of Biases

  • Random assignment helps ensure groups are comparable and reduces bias.

  • Control vs experimental groups enable comparison to determine the effect of the IV.

  • Blinding:

    • Single-blind study: participants do not know whether they are in the control or experimental group.

    • Double-blind study: neither participants nor researchers know who has received the placebo, to reduce expectations influencing results.

  • Placebo and placebo effect:

    • Placebo: a fake treatment (e.g., decaf coffee when testing caffeine) used to control for expectations.

    • Placebo effect: participants’ changes in behavior due to belief they received an active treatment.

Confounding Variables, Bias, and Validity

  • Extraneous/Confounding variables: factors other than the IV that can influence the DV.

    • Example: heat as a confounding variable in a study linking ice cream sales to violent crime.

  • Experimental bias: researcher bias that supports the hypothesis (cherry-picking data or methods).

  • Validity: the extent to which a test or experiment measures or predicts what it is supposed to.

  • Hawthorne effect (not in notes, but mentioned): participants alter behavior simply because they know they are being observed.

  • Barnum effect (not in notes, but mentioned): people believe vague personality descriptions apply uniquely to them (e.g., astrology).

Non-Descriptive (Descriptive/Non-Experimental) Methods

  • These methods describe behavior and effects but do not manipulate variables.

  • Examples mentioned: case studies, naturalistic observation.

  • Purpose: observe and record behavior and its effects without experimental manipulation.

Qualitative vs Quantitative Research

  • Quantitative research: relies on numerical data (e.g., survey results, test scores).

  • Qualitative research: relies on in-depth narrative analysis (descriptions, interviews) that can be translated into numerical data.

  • The course will revisit these distinctions and how both can contribute to understanding.

Ethics in Psychological Research (APA Guidelines)

  • APA: American Psychological Association; provides guidelines for ethical research.

  • Voluntary participation: participants freely choose to participate.

  • Informed consent: participants are given enough information about the study to decide whether to participate.

  • Passive or partial disclosure is possible regarding purpose, but participants should understand procedures.

  • If participants are under 18, parental consent is required.

  • Confidentiality: participant information must be kept private and secure.

  • Debriefing: after participation, explain the study’s purpose and any deception used.

  • Deception and its disclosure: if deception occurred, debriefing must reveal it and justify its use.

  • No long-term harm: researchers must ensure no lasting physical or psychological harm to participants; animals must be well cared for.

  • IRB approval: research institutions require Institutional Review Board approval before performing experiments; this involves submitting a full proposal with purpose, methods, questions, and participant details.

  • Scientific integrity: reporting accurate, verified data and avoiding misrepresentation or fake science; citing reputable sources.

  • Citations: acknowledge others’ ideas when used; helps protect against errors and maintains scholarly integrity.

  • Practical note: computers may not be used during certain activities; prepare to engage with correlational activity as described in the notes and Canvas.

Relationships Between Methods and Applications

  • When planning research, choose method according to ethical considerations, feasibility, and what you aim to determine (relationship vs causation).

  • Correlational studies are valuable for identifying relationships, patterns, and hypotheses that can later be tested with controlled experiments.

  • Experiments provide stronger evidence for causation but require careful design to minimize biases and confounding factors.

Quick Reference: Key Formulas and Concepts

  • Correlation coefficient: r = \frac{\mathrm{cov}(X,Y)}{\sigmaX \sigmaY} = \frac{\sum (xi-\bar{x})(yi-\bar{y})}{\sqrt{\sum (xi-\bar{x})^2}\sqrt{\sum (yi-\bar{y})^2}}

  • r is bounded: -1 \le r \le 1

  • Interpretations (rules of thumb):

    • |r| close to 1: strong relationship; |r| close to 0: weak relationship.

    • Positive r: variables move in the same direction.

    • Negative r: variables move in opposite directions.

  • Examples from the lecture:

    • Positive: more study → higher grade; fewer absences → fewer referrals.

    • Negative: more absences → lower grade; less sleep → more car accidents.

    • Ice cream sales and violent crime: both rise with temperature but one does not cause the other.

  • Experimental structure: IV, DV, control and experimental groups, random assignment.

  • Blinding types: single-blind vs double-blind.

  • Placebo effect and placebo control.

  • Confounding variables and experimental bias must be minimized for valid results.

  • Hawthorne effect and Barnum effect as additional considerations in interpretation and communication of findings.

  • Ethics and integrity: voluntary participation, informed consent, confidentiality, debriefing, minimization of harm, IRB approval, animal welfare, and accurate reporting with proper citations.

Notes on Upcoming Activities

  • The course will return to experiments and their components in a couple of days.

  • There is an assignment to practice with correlational methods on Canvas.

  • Students will identify independent and dependent variables in practice questions.

Connections to Foundational Principles and Real-World Relevance

  • Correlational research supports hypothesis generation and ethical study of relationships when experiments are not feasible.

  • Understanding causation requires controlled manipulation and random assignment to rule out confounding variables.

  • Ethical guidelines (APA) protect participants and ensure the reliability and integrity of scientific findings.

  • Knowledge of biases and effects (Placebo, Hawthorne, etc.) improves experimental design and interpretation of results in real-world psychology research.

Common Misconceptions and Practical Tips

  • Do not infer causation from correlation alone; always consider possible confounds and alternative explanations.

  • Be mindful of regression toward the mean when interpreting extreme scores in repeated measurements.

  • Use random assignment to reduce selection bias and improve internal validity.

  • Use blinding when possible to reduce expectancy effects on participants and researchers.

  • Maintain transparency: preregister hypotheses, report methods, and cite sources to uphold scientific integrity.

Quick Summary for Exam

  • Correlation measures relationships; r ∈ [-1,1]; strength grows with |r|; correlation does not imply causation.

  • Experiments establish causation via manipulation of IV, measurement of DV, random assignment, control groups, and blinding.

  • Ethical guidelines (APA) govern voluntary participation, informed consent, confidentiality, debriefing, deception handling, and IRB approval.

  • Descriptive and non-experimental methods (case studies, naturalistic observation) describe phenomena but do not establish causation.

  • Distinguish quantitative (numerical data) from qualitative (narrative data translated to numbers).

  • Be aware of biases and effects (confounding variables, experimental bias, placebo effect, Hawthorne effect, Barnum effect) when designing and interpreting research.