Research Methods, Statistics, and Ethics in Psychology

Experimental Variables and Research Scenarios

• Independent Variable (IV): The variable that is manipulated or categorized by the researcher to observe its effect on an outcome.

• Dependent Variable (DV): The variable that is measured by the researcher to assess the effect of the independent variable.

• Experimental Conditions: The specific environments, levels, or treatments of the independent variable to which participants are exposed.

• Class Time Scenario:

  • Independent Variable: Class time.

  • Conditions: Early or daytime classes versus nighttime classes.

  • Dependent Variable: Exam scores.

• Child Violence Exposure Scenario:

  • Independent Variable: Level of violent media exposure shown on television.

  • Conditions:

  • Non-violent programming (e.g., Sesame Street).

  • Mild or harmless violence.

  • Intense, life-threatening, or fatal violence.

  • Dependent Variable: The degree of aggressive behavior exhibited by the children following media exposure.

  • Observational Case Study on Aggression: Exposure to media characters can directly trigger behavioral imitation in young children. For example, a child who watched Spidey and Friends began displaying whining behavior copied from a character named Rhino. After receiving figurines of characters including Spidey, Spin, Hulk, Gwen (Ghost-Spider), and Rhino for Christmas, the child picked up the Hulk figurine, shouted "Hulk, smash!", and physically struck an adult on the leg. This outcome prompted the immediate reduction of the child's viewing access to the program to curb violent behavioral mimicry.

• Rat Memory Scenario:

  • Independent Variable: Solution administered prior to testing (Alcohol solution versus Saline solution).

  • Dependent Variable: Maze memory performance.

• Study Method Scenario:

  • Independent Variable: Method of studying.

  • Conditions:

  • Recite-recite-review strategy.

  • Rote memorization through repetition.

  • Dependent Variable: Memory retention and test performance.

Sampling Methods and Quasi-Experimental Designs

• Population: The complete, overarching group of individuals or cases about which a researcher wishes to draw conclusions.

• Sample: A subset of individuals selected from the larger population to participate in a research study.

• Random Sample:

  • A sampling procedure designed to gather participants as randomly as possible from the target population.

  • Ensures that the sample serves as an accurate representative subset of the larger population, which is necessary because researchers cannot access an entire population.

• Random Assignment:

  • An experimental procedure in which every participant has an equal mathematical chance of being placed into any condition or group within a study.

  • Eliminates selection bias and confounding participant attributes (e.g., preventing researchers from systematically placing friends or specific individuals into preferred conditions).

  • Practical Implementation Methods: Drawing participant names out of a hat, or having participants blindly pick colored tokens (e.g., choosing a red chip versus a blue chip) from a container.

• Quasi-Experimental Designs:

  • Research designs utilized when true random assignment of the independent variable is physically, biologically, or ethically impossible.

  • Fixed Subject Characteristics: Variables such as chronological age, biological eye color, or natural hair color cannot be randomly assigned or manipulated (e.g., researchers cannot use a time machine to alter a participant's age or force structural biological changes).

  • Ethical Constraints: Certain behaviors cannot be experimentally assigned due to severe ethical violations, such as forcing non-smokers to smoke cigarettes or forcing non-caffeinated individuals to consume high doses of caffeine.

  • Causal Limitations: Because participants bring pre-existing differences to the study without random assignment, quasi-experiments cannot definitively establish true cause-and-effect relationships.

  • Scientific Value: Quasi-experiments remain vital, highly valuable, and essential tools in psychological research, particularly when combined with existing empirical literature to approximate causal inferences across demographic groups.

Statistical Analysis and APA Journal Publication

• Statistical Significance:

  • Statistical analyses are conducted to determine whether observed differences in outcomes across conditions (e.g., exam performance differences between daytime and nighttime conditions) represent true effects or random variance.

  • Significance standardly relies on probability testing ($p$-values).

  • Alpha Threshold: The universal standard threshold for statistical significance is $p \le 0.05$ ($5\%$ probability that the result occurred by random chance). Results exceeding $p = 0.05$ are designated as statistically non-significant.

• Quantitative vs. Qualitative Analysis:

  • Quantitative Analysis: Empirical evaluation relying on numerical values, measurement scales, and probability testing.

  • Qualitative Analysis: Descriptive evaluation relying on participant interviews, response categorization, and thematic analysis, which is characteristically labor-intensive and time-consuming.

• American Psychological Association (APA) Standards:

  • Provides standardized rules for manuscript formatting, citations, paper organization, and structural components across psychological literature.

  • Standardized guidelines ensure consistency, clarity, and ease of reading for researchers reviewing published literature.

• Peer-Reviewed Journal Publication:

  • Peer-reviewed journal articles represent the supreme "gold standard" for scholarly dissemination in psychology and STEM disciplines (contrasting with book publications prioritized in humanities disciplines like history or English).

  • Unpaid Academic Labor System: Scholarly publishing relies on unpaid, voluntary labor from expert researchers who conduct peer reviews, while commercial journal publishers monetize the resulting publications.

The Peer Review Process and Research Replication

• Role of the Journal Editor:

  • Receives manuscript submissions and performs an initial screening for topical alignment with the journal's domain (e.g., rejecting a molecular biology paper submitted to a developmental psychology journal).

  • Dispatches qualified manuscripts to expert reviewers for critical evaluation.

• Role of Peer Reviewers:

  • Field experts who evaluate submitted manuscripts regarding theoretical rigor, methodological soundness, and analytical accuracy.

  • Journals typically recruit $3$ reviewers (or a minimum of $2$) per manuscript.

  • Reviewers evaluate weaknesses and provide feedback ranging from harsh criticism to constructive guidance. Methodological flaws exist in all published papers, as no journal article is flawless.

• Editorial Decision Outcomes:

  • Reject: The manuscript is declined for publication, often after a review period lasting $3\text{ to }6\text{ months}$.

  • Immediate Accept: Direct acceptance upon initial submission is exceptionally rare and serves as a potential red flag regarding journal quality.

  • Revise and Resubmit (R&R): The primary target outcome for authors.

  • Minor Revisions: Involves small textual or structural adjustments.

  • Major Revisions: Requires extensive manuscript overhaul, including re-running statistical analyses or re-evaluating theoretical frameworks.

  • Resubmission: Revised manuscripts are returned to the editor and resubmitted to the original or new reviewers for final evaluation.

• Scientific Replication:

  • Methodological descriptions in published research must contain sufficient, granular detail to allow independent scientists to duplicate the study.

  • Embodied Cognition Example: Embodied cognition theory posits that bodily states alter spatial perception (e.g., carrying a heavy backpack causes a distance of $1\,mile$ to feel significantly farther than walking unburdened). A skepticism-driven replication study evaluating this exact backpack protocol yielded no empirical support for the phenomenon, demonstrating the necessity of replication.

  • Replication Crisis: Academic publishing structures disincentivize exact replications, as journals systematically prioritize novel findings or "replication and extension" studies over pure replication attempts.

Measurement Reliability and Validity

• Reliability (Consistency):

  • The consistency and stability of a measuring instrument over time and across test items.

  • Internal Consistency: Evaluates whether different items within the same scale or survey yield consistent measurements of the same construct. The inclusion of multiple, similarly phrased questions within a single survey is an intentional structural method to verify scale reliability.

• Validity (Accuracy):

  • The extent to which a measurement tool accurately measures the specific variable it was designed to evaluate.

  • Ecological Validity: The extent to which experimental findings can be generalized and applied to real-world settings and everyday environments.

  • Construct Validity: The degree to which a test measures the intended theoretical construct rather than extraneous variables (e.g., measuring sleep quality by evaluating sleep duration and mattress comfort rather than unrelated emotional states).

  • Face Validity: The surface-level appearance of a measurement tool; whether the items appear relevant and valid to an external evaluator upon initial inspection.

• Hierarchy of Reliability and Validity:

  • Reliability and validity are fundamentally linked complementary concepts.

  • Validity is hierarchically more important than reliability: a test can be highly reliable (consistently producing identical results) while remaining entirely invalid (measuring the wrong target variable accurately).

Ethical Standards and the Institutional Review Board (IRB)

• Historical Ethical Failures in Psychology:

  • Stanford Prison Experiment: Induced severe emotional trauma and psychological distress in human participants.

  • Milgram Shock Study: Subjected participants to high levels of psychological stress and deception regarding the administration of lethal electrical shocks.

  • Tuskegee Syphilis Study: Economically disadvantaged participants were subjected to painful, non-therapeutic spinal taps under the guise of free medical care. When penicillin was discovered as the effective cure for syphilis, treatment was intentionally withheld from participants to observe the natural progression of the disease.

• Regulatory Frameworks: Ethical violations led to the creation of the Belmont Report, which formed the foundation for the ethical guidelines adopted by the American Psychological Association (APA).

• Institutional Review Board (IRB):

  • An administrative committee formally established within research institutions to review, approve, modify, or reject human subject research protocols prior to data collection.

  • Board Structure: Low-risk studies (such as simple surveys) may be managed by an IRB chair, whereas protocols involving elevated risk require evaluation by a full committee board meeting monthly.

  • Conflict Management: IRB chairs and board members cross-review proposals to eliminate self-approval conflicts of interest.

  • Strict Mandate: No human subject research may commence without explicit, formal IRB approval.

• Mandatory Ethical Protocols:

  • Informed Consent: Mandatory documented agreement informing human participants of all study procedures, potential risks, and rights prior to participation.

  • Deception: Must be avoided whenever possible. Deception is permitted exclusively when necessary to preserve research validity (e.g., studying social prejudice, where disclosing the true hypothesis would cause participants to adopt defensive or socially desirable response biases).

  • Debriefing: A mandatory post-experimental session where researchers explain the true purpose of the study, disclose any deception used, and minimize lingering participant distress.

Animal Research and Institutional Animal Care and Use Committee (IACUC)

• Rationale for Animal Models:

  • Rodent Models: Rats and mice represent the primary animal subjects in psychological and pharmacological research.

  • Methodological Advantages: Rodents possess rapid reproductive cycles, small physical sizes allowing efficient laboratory housing, lower ethical barriers compared to larger mammals (e.g., horses), and physiological/pharmacological system similarities to humans.

  • Substitutive Function: Animal models serve as essential substitutes for invasive biological manipulations or high-risk drug trials that are strictly unethical to perform on human participants.

  • Generalization Limits: Findings from animal research cannot be directly applied to human populations without independent verification through human clinical testing.

• Institutional Animal Care and Use Committee (IACUC):

  • The dedicated governing oversight committee for research involving non-human animal subjects, serving as the regulatory representative for animal welfare.

  • Regulatory Duties: Mandates pre-experimental proposal review and conducts unannounced, semi-annual or annual facility inspections to ensure lab sanitation, proper housing, and ethical care.

  • Strict Mandate: No animal research protocol can begin or continue without full IACUC approval.