Psychological Research Methods Vocabulary

Foundations of Psychological Research and the Scientific Method

• Goal of Scientific Research in Psychology:

  • Psychology aims to understand human behavior alongside underlying mental and physical processes.
  • A primary objective is constructing statistical predictive models to anticipate human behavior.
  • Scientific models are dynamic and subject to ongoing modification or expansion as new empirical data emerges.

• Facts vs. Opinions:

  • Fact: An objective, observable, and verifiable reality established through empirical evidence.
  • Opinion: A personal judgment, preference, or conclusion not necessarily supported by empirical evidence.
  • Empirical Evaluation of Early Intervention Programs:
    • Head Start: Extensively researched early childhood intervention targeting low-income children. Empirical findings demonstrate short-term benefits (e.g., temporary closure of the academic achievement gap), but long-term sustained gains are limited, as the achievement gap re-expands as children progress through school.
    • Georgia Pre-K: Funded statewide by the Georgia Lottery and open to all residents regardless of income level. Operates during standard school hours, requiring families to fund post-school childcare independently. Empirical evaluations reveal mixed efficacy—showing significant effectiveness in structured, well-funded implementations, but limited impact in under-resourced settings.
    • Critical Evaluation: Objective decision-making regarding public programs requires critical analysis of research methodology, investigator credentials, structural differences between effective and ineffective models, and avoiding reliance on personal anecdotes (e.g., assuming universal program efficacy solely because one's own child entered advanced content classes).

• The Scientific Method and Scientific Reasoning:

  • Theory: A well-developed, structured set of ideas that explains observed phenomena.
  • Hypothesis: A specific, testable prediction derived from a theory, formulated as a research question regarding expected experimental outcomes.
  • Falsifiability: A fundamental criterion for scientific hypotheses and theories requiring that a proposition can be disproven by empirical evidence.
    • Non-falsifiable Examples: Evolutionary psychological theories that cannot be conclusively tested due to temporal constraints spanning eons; Freudian psychoanalytic concepts centered on the unobservable unconscious mind.
  • Scientific Reasoning Cycle:
    • Deductive Reasoning: Process of predicting specific empirical outcomes based on general premises or established theoretical principles ($\text{General Premise} \rightarrow \text{Specific Empirical Prediction}$).
    • Empirical Observation: Conducting structured studies to collect quantitative or qualitative data.
    • Inductive Reasoning: Process of deriving general conclusions and theoretical updates from specific empirical observations ($\text{Specific Data Points} \rightarrow \text{General Theoretical Conclusion}$).
    • Theory Revision: Data analysis leads to confirmation or modification of the existing theory, initiating a new cycle of hypothesis formulation and testing.
  • Research Focus: Researchers often dedicate entire careers to a narrow specialization (e.g., cognitive development in three-year-olds) to iteratively refine theoretical models.

Descriptive and Observational Research Methodologies

• Clinical or Case Studies:

  • Definition: An in-depth investigation focusing on a single individual or a small, unique group possessing rare physical or psychological characteristics.
  • Krista and Tatiana Hogan Case Study: Conjoined twins who share a thalamus (the brain structure regulating and relaying sensory information, excluding the sense of smell).
    • Empirical Testing: Tactile or thermal stimulation (e.g., tickling or applying heat) applied to the hand of one twin enables sensory detection in the other twin.
    • Limitations: Impossible to replicate due to the rarity of the anatomical arrangement, leading to near-zero sample scalability.
  • Phineas Gage Case Study: Historical case study demonstrating that traumatic damage to the frontal lobe drastically alters personality and social behavior.
  • Generalizability: The extent to which research findings from a specific sample can be applied to the broader population. Case studies generally lack generalizability unless assessing universal human characteristics (e.g., facial expressions of emotion or the Big Five personality traits).

• Naturalistic Observation:

  • Definition: Observing and recording behavior in its natural setting without experimental manipulation or direct researcher intervention.
  • Jane Goodall Case Study: Conducted a landmark five-year naturalistic observation study of chimpanzee behavior in Africa.
  • Methodological Requirements: Researchers must remain unobtrusive so their presence does not alter natural subject behavior (e.g., observing playground dynamics from a distant park bench versus standing amidst children with a clipboard).
  • Observer Bias: Distortion of observational records caused by a researcher's unconscious expectations or theoretical biases.
  • Mitigation Strategies: Formulating explicit operational definitions paired with an exhaustive behavioral checklist (e.g., defining aggressive playground acts strictly as pinching, pulling, poking, or hitting).
  • Inter-Rater Reliability: A metric evaluating the degree of agreement and consistency between independent observers coding the exact same behavior based on standardized criteria.

• Survey Research:

  • Definition: A structured data collection instrument (administered via paper, electronic platforms, or direct interviews) designed to efficiently gather self-reported data from large sample sizes.
  • Population: The overall defined group of individuals targeted for investigation (e.g., the entire student body at Chattahoochee Technical College, spanning ages $16$ to $90$, including dual-enrolled, traditional, and non-traditional students).
  • Sample: A subset of individuals selected from the overarching population. To ensure validity, the sample must mirror population demographics (e.g., sampling only dual-enrolled students introduces sampling bias regarding general campus services).
  • Census: A comprehensive study that measures and counts every single individual within a defined population (e.g., the United States Census).

• Archival Research:

  • Definition: Examining pre-existing datasets, institutional records, or historical documents to discover patterns, trends, or relationships.
  • Examples: Analyzing collegiate academic databases to assess course loads (e.g., identifying that taking $4$ courses per semester, excluding summer terms, yields higher graduation rates); evaluating standardized ACT and SAT scores to validate their predictive utility for freshman-year college GPA.

• Developmental Research Designs:

  • Longitudinal Research: Assessing the exact same cohort of subjects repeatedly over extended time intervals (e.g., testing infants at $2$, $4$, $6$, $8$, $10$, and $12$ months, or tracking cohorts across several decades).
    • Advantages: Provides clean data by controlling for generational cohort differences.
    • Disadvantages: Highly expensive, time-consuming, and subject to participant attrition (the progressive loss of subjects over time due to relocation, death, or voluntary withdrawal).
  • Cross-Sectional Research: Comparing distinct age groups or population segments simultaneously at a single point in time (e.g., comparing groups of $3$-, $5$-, and $7$-year-olds, or evaluating sexual partner frequency across Gen Z, Millennials, Gen X, and Baby Boomers).
    • Advantages: Fast, cost-effective, and immune to participant attrition.
    • Disadvantages: Susceptible to cohort effects and age-related lifespan confounders.

Correlational Research and Analysis

• Correlational Principles:

  • Correlation: A statistical relationship measuring the extent to which two or more variables co-vary.
  • Correlation Coefficient ($r$): A numerical statistic ranging strictly from $-1.00$ to $+1.00$ indicating the strength and direction of a relationship.
    • $+1.00$: Perfect positive correlation.
    • $-1.00$: Perfect negative correlation.
    • $0.00$: Absolute absence of a linear statistical association.

• Types of Correlation:

  • Positive Correlation: Variables move in the exact same direction; as variable $X$ increases, variable $Y$ increases (e.g., human height and body weight).
  • Negative Correlation: Variables move in opposite directions; as variable $X$ increases, variable $Y$ decreases (e.g., hours of restful sleep and self-reported fatigue).
  • Zero Correlation: Absence of any systematic relationship, represented by a completely random scatter plot (e.g., shoe size and hours of sleep).

• Directionality and Strength:

  • The mathematical sign ($+$ or $-$) denotes direction only, not strength.
  • The absolute numerical value indicates strength; values closer to $1.00$ (whether $-1.00$ or $+1.00$) represent stronger predictive relationships.

• Correlation vs. Causation:

  • A statistical correlation between two variables does not prove or imply a direct cause-and-effect relationship.
  • Confounding Variable: An unmeasured, underlying third variable that directly causes observed changes in both correlated variables.
    • Example 1: Positive correlation between the number of churches and liquor stores in towns (Confounding Variable: total population growth).
    • Example 2: Positive correlation between ice cream sales and drowning rates (Confounding Variable: outdoor ambient temperature increases).

• Cognitive Biases in Empirical Interpretation:

  • Illusory Correlation: Perceiving a meaningful statistical relationship between two variables when no actual relationship exists.
  • Confirmation Bias: The tendency to selectively search for, interpret, and recall information that confirms pre-existing beliefs while disregarding disconfirming empirical evidence (e.g., maintaining claims linking vaccines to autism based on a discredited $12$-subject study while ignoring vast experimental literature disproving the link).

Experimental Methodology and Control Procedures

• Establishing Causation:

  • Controlled experiments represent the sole methodology capable of establishing definitive cause-and-effect ($\text{Cause } A \rightarrow \text{Effect } B$) relationships.

• Experimental Group Structure:

  • Experimental Group: The group of subjects exposed to the experimental manipulation or treatment variable.
  • Control Group: The group maintained under identical conditions as the experimental group, except they do not receive the experimental manipulation. Serves as a baseline comparison.

• Placebo Effect:

  • Psychological or physiological responses caused solely by participant expectations or beliefs regarding a treatment, rather than active therapeutic agents (e.g., spontaneous symptom reduction in psychiatric depression trials).

• Operational Definitions and Variables:

  • Operational Definition: A precise, standardized explanation specifying exact procedures, measurements, and conditions used to define experimental variables.
  • Independent Variable (IV): The variable systematically manipulated or controlled by the experimenter.
  • Dependent Variable (DV): The variable measured by the experimenter to assess the effect of the independent variable.
  • Note-Taking Methodology Study ("The Pen is Mightier Than the Keyboard"):
    • Independent Variable: Note-taking medium (handwritten notes vs. laptop note-taking).
    • Dependent Variable: Learning efficacy, factual recall accuracy, and conceptual comprehension scores.

• Sources of Experimental Bias:

  • Experimenter Bias: Unconscious cues or behaviors by the researcher (e.g., subtle nodding during participant assessments) that alter participant responses or distort data recording to align with experimental hypotheses.

• Blinding Protocols:

  • Single-Blind Study: Research design where participants are unaware of whether they are in the experimental or control group, while researchers remain aware.
  • Double-Blind Study: Research design where both participants and the experimental staff interacting with them are kept unaware of group assignments. Prevents both experimenter bias and placebo effects; essential standard in clinical trials.

Sampling, Experimental Controls, and Publication Standards

• Sampling and Group Assignment:

  • Participants: Human subjects engaged in scientific research (preferred term over "subjects").
  • Random Sample: Sampling procedure where every member of a defined population has an equal chance of being selected (achievable in small populations like all students at Chattahoochee Technical College).
  • Random Assignment: Procedural method placing participants into experimental or control groups such that each individual has an equal probability of assignment to either group (e.g., coin flipping or shuffling unlabelled plain folders). Equalizes baseline differences across groups.

• Quasi-Experimental Limitations:

  • Quasi-Experimental Variables: Subject traits that cannot be ethically or physically manipulated or randomly assigned by researchers (e.g., biological sex, chronological age, geographic origin, generational cohort).

• Statistical Analysis:

  • Selecting appropriate statistical tests based on design: Correlational analyses for survey descriptions; $t$-tests or Analysis of Variance (ANOVA) for experimental group comparisons to generate probability metrics ($p$-values).

• Peer Review and Replication:

  • Peer-Reviewed Journal Articles: Standard scientific publishing process where submitted manuscripts undergo anonymous evaluation by independent experts in the field prior to publication.
  • Replication: Independent repetition of an published experimental design using new sample populations to establish the reliability and stability of original findings.
  • Scientific Retractions: Official retraction of published studies when findings are proven unreplicable, methodologically flawed, or fraudulent.
    • Example 1: Retraction of the fraudulent autism-vaccine paper originally published in Science, leading to the revocation of the primary author's medical license.
    • Example 2: Institutional termination and research expulsion resulting from fabricated datasets (e.g., major fraud cases at Harvard University).

Psychological Measurement and Ethical Governance

• Psychometric Standards:

  • Reliability: The consistency, stability, and reproducibility of a measure or experimental result across repeated trials.
  • Validity: The accuracy of a test or instrument in measuring the exact construct it was designed to evaluate (e.g., using a math test to assess reading comprehension lacks validity).

• Ethical Governance and Human Protections:

  • Institutional Review Board (IRB): An administrative committee comprising scientists, medical professionals, attorneys, and non-scientific community members tasked with reviewing human research protocols to prevent harm.
  • Informed Consent: A detailed written agreement explaining study objectives, potential risks, benefits, voluntary participation terms, explicit right to withdraw at any time without penalty, and primary investigator contact information.
  • Prohibition of Coercion: Participation must be strictly voluntary. When research participation is encouraged (e.g., extra credit in college courses), equal non-research alternatives must be provided (e.g., reading a research paper on implicit bias and writing a summary).
  • Deception Protocols: Withholding true research hypotheses is permissible only when essential to prevent study bias, provided no physical or psychological harm occurs.
  • Debriefing: Mandatory post-experimental session where participants are informed of the complete truth, underlying purpose, details of any deception used, and provided psychological remediation if needed.

• Historical Ethical Failure:

  • Tuskegee Syphilis Study: Unethical medical study tracking rural African American men who were deceptively told they were receiving treatment for syphilis when active treatment (penicillin) was intentionally withheld to observe natural disease progression. Directly catalyzed modern federal IRB regulations and human subject protections.

Animal Research in Psychology

• Governance and Animal Oversight:

  • Institutional Animal Care and Use Committee (IACUC): An ethical review board composed of administrators, scientists, veterinarians, and community representatives that evaluates animal research proposals to enforce humane treatment standards.

• Rationale for Animal Models:

  • Genetic and Biological Homology: Humans share substantial genetic material with other mammals (e.g., approximately $90\%$ shared DNA with most mammals, $98\%$ shared DNA with chimpanzees).
  • Neurological Similarity: Shared basic neural structures across species, such as brain reward centers in rats, enabling research into the addictive properties of analgesic medications.
  • Invasive Research Methodologies: Allows necessary biological procedures that are ethically prohibited in human participants (e.g., post-experimental euthanasia for surgical brain tissue examination).

• Legal Protections for Non-Human Primates:

  • Chimpanzees were legally granted personhood status in New York State, legally barring them from invasive research due to their inability to provide informed consent.
  • Primate research is legally restricted to non-harmful observational or non-invasive cognitive designs (e.g., language development studies at the Great Ape Trust in Iowa).