Pysch Lecutre 3

Key Concepts and Fundamental Definitions

• Scientific Method in Psychology

  • The establishment of psychology as an empirical discipline relies on scientific practices and structured research methods.

  • Research practices elevate empirical work above speculation, providing concrete evidence to test hypotheses and theories.

  • Research methods allow researchers to separate verified facts from subjective assertions, anecdotal claims, or pop-psychology assumptions.

• Core Methodological Definitions

  • Variable: Any measurable condition, event, characteristic, or behavior that is controlled or observed in a study.

  • Operational Definition: A definition that clearly indicates or describes the exact actions, operations, or procedures used to measure or control a specific variable. Operational definitions prevent misinterpretation by making abstract constructs concrete and measurable.

  • Experiment: A research method in which an investigator manipulates a variable under carefully controlled conditions and observes whether any changes occur in a second variable as a result.

  • Independent Variable (IV): The variable that is directly manipulated by the experimenter to test its effects on an outcome.

  • Dependent Variable (DV): The variable that is measured and is thought to be affected by or changed as a direct product of manipulating the independent variable.

  • Control Group: A group consisting of similar subjects who do not receive the manipulated independent variable (or experimental treatment). Control groups are utilized to ensure that confounding variables do not account for the observed results.

  • Correlation: A numerical or structural indication of a relationship between two variables that is explicitly non-causal.

Operational Definitions in Clinical Practice

• Application in High-Intensity Clinical Work

  • Operational definitions are essential in clinical settings, particularly when managing severe client behaviors involving suicidality, homicidality, and high-intensity aggression.

  • Vague clinical descriptors presented by clients or parents (e.g., "my kid is angry all the time" or "he gets aggressive") represent broad variables that mean different things to different individuals.

  • Concrete Scenario — Deconstructing "Fights":

  • A report stating a child was sent home for getting into a "fight" can represent drastically different behaviors:

    • Lower intensity: Yelling at a peer or taking a card back from a classmate.

    • Extreme intensity: Physical violence resulting in breaking someone's nose.

  • Establishing precise operational definitions allows clinicians to evaluate whether a child's behavior meets the specific threshold required for high-intensity specialized practice or if the child should be referred to appropriate colleagues better suited for lower-intensity behavioral struggles.

Case Study: Evolutionary Psychology and Infidelity Research

• Background and Context

  • Research began during an Introduction to Evolutionary Psychology course in the second semester of freshman year in New York.

  • The topic built upon a seminal $1992$ study examining sex differences in distress reactions to infidelity using an evolutionary psychology framework.

• Theoretical Framework of the $1992$ Study

  • Biological Females: Evolutionary theory posits that biological females experience significantly higher distress in response to emotional infidelity by male partners. This stems from an evolutionary fear related to pregnancy risk, specifically that male partners might allocate critical survival resources away from them and their offspring to another individual.

  • Biological Males: Evolutionary theory posits that biological males experience significantly higher distress in response to physical infidelity by female partners. This stems from paternity uncertainty and the biological necessity of differentiating whether an offspring carries their genetic material.

• The Infidelity Scenarios & Thought Experiment

  • Scenario A (Emotional Infidelity): A partner confesses to going out to dinner with someone else. It was an enjoyable date with good mood and food, but entirely non-physical, with absolute reassurance that it will not happen again.

  • Scenario B (Physical Infidelity): A partner confesses to a strictly physical encounter with someone else involving no emotional connection whatsoever, with absolute reassurance that it will not happen again.

  • Findings confirm that biological females report significantly higher distress regarding emotional infidelity, whereas biological males report significantly higher distress regarding physical infidelity.

• Hypothesis Refinement and Publication Path

  • Critical Research Question: What happens to distress patterns when the fear of pregnancy or resource allocation is controlled or removed? For example:

  • If a male partner engages in physical infidelity with another male entity, eliminating pregnancy risk.

  • If a male partner engages in emotional infidelity with another male, requiring resource allocation elsewhere.

  • Outcome: The inquiry led to an invitation to office hours, a crash course in research methods, and collaboration with a lab master's student named Julie. The project resulted in a peer-reviewed publication at the beginning of sophomore year.

Structure of Scientific Theories and Hypotheses

• Theories vs. Hypotheses

  • Theory: A system of interrelated ideas used to explain a set of observations.

  • Theories are not mere speculation or informal guesses; hard sciences rely on theories backed by extensive empirical evidence.

  • Heliocentrism: A theory explaining observations of planetary and stellar movements around the sun.

  • Gravity: A theory explaining physical observations, such as apples falling or astronauts jumping higher on the moon due to the moon's smaller mass compared to Earth.

  • Hypothesis: A tentative statement about the relationship between two or more variables. Hypotheses must be testable and formulated so variables can be objectively measured.

• Avoidance of the Term "Cause"

  • Rigorous empirical researchers avoid stating that one variable "causes" another.

  • Direct causal assertions require strict methodological rigor to account for all alternative confounding variables.

• Historical Links to School of Thought

  • The emphasis on objective, verifiable measurement of variables aligns directly with the Behaviorist school of psychological thought.

Step-by-Step Experimental Research Design: "Opposites Attract"

• Step 1: Formulating a Testable Hypothesis

  • Research question explored: Do opposites attract, or do similar personalities attract?

  • Operational Definition of Attraction: The willingness to engage in a physical or emotional/romantic relationship.

  • Operational Definition of Personality: The qualities of an individual that form their identity, as measured by standardized, validated psychological tests such as the Myers-Briggs Type Indicator (MBTI) or the Minnesota Multiphasic Personality Inventory (MMPI).

  • Formal Hypothesis: Individuals are more likely to experience feelings of attraction (willingness to engage in physical or emotional relationships) toward individuals with differing personalities (as measured by the MMPI).

• Step 2: Selecting Method and Design

  • Correlational Design Approach: Subjects take a personality assessment and administer the assessment to any prospective real-world romantic or physical partners. Researchers observe natural interactions without manipulating variables.

  • Experimental Design Approach: Researchers directly manipulate personality traits presented in hypothetical vignettes (e.g., profiles for fake individuals named John or Jake displaying traits like extroversion or conscientiousness opposite to or matching the subject's profile) while keeping physical appearance constant using standard photographs.

  • Independent Variable (IV): The personality traits described in the vignette (manipulated to be similar or opposite).

  • Dependent Variable (DV): The reported level of attraction to the vignette profile.

• Step 3: Identifying Extraneous and Confounding Variables

  • Confounding Variable: Any variable not actively measured or controlled by the experimenter that systemically impacts the dependent variable.

  • Example: In a study assessing female attraction to a male profile named John, a subject's sexual orientation (e.g., identifying as gay) represents a confounding variable that alters attraction independent of the manipulated personality traits.

• Step 4: Selecting Data Collection Techniques

  • Psychological tests (e.g., validated personality instruments) measure the independent variable.

  • Questionnaires and vignettes measure the dependent variable (e.g., rating attraction on a scale from $1$ to $10$).

• Step 5: Defining Population and Sample

  • Population: The larger collection of individuals from which a sample is drawn (e.g., all college students in the United States).

  • Sample: The specific collection of subjects selected for active observation in the study (e.g., college students attending a public university in the Northeastern United States).

Statistical Analysis: Correlation Coefficients

• Correlational Analysis Principles

  • Correlation: Measures the degree of a non-causal relationship between two variables.

  • Correlation Coefficient ($r$): A numerical index of the direction and strength of a relationship between two variables, bounded strictly between $-1.00$ and $+1.00$.

• Direction of Correlation

  • Positive Correlation: Two variables covary or change together in the same direction.

  • Example: Increased study hours associated with higher test grades.

  • Example: Decreased sleep hours associated with lower test scores (both variables move downward together, maintaining a positive correlation).

  • Negative Correlation: Two variables move in opposite directions; as one variable increases, the other decreases.

  • Example: Increased active listening in class associated with a decrease in required exam study time.

• Strength of Correlation

  • Strength is determined solely by how far the coefficient is from $0.00$, approaching either $-1.00$ or $+1.00$.

  • A correlation coefficient of $r = -0.50$ possesses identical strength to a correlation coefficient of $r = +0.50$.

  • Strong Correlation: Magnitude between $0.70$ and $1.00$ (or $-0.70$ and $-1.00$).

  • Medium / Moderate Correlation: Magnitude between $0.40$ and $0.70$ (or $-0.40$ and $-0.70$).

  • Weak Correlation: Magnitude between $0.00$ and $0.40$ (or $0.00$ and $-0.40$).

• Evaluated Examples

  • Time spent basketball practicing vs. points scored: Positive and strong.

  • Time spent basketball practicing vs. academic test scores: Negative and weak-to-medium.

  • Time spent gardening vs. plant height at end of season ($r = +0.75$): Positive and strong.

  • Money lost gambling vs. subjective happiness ($r = -0.50$): Negative and medium/moderate.

• Correlation vs. Causation

  • Correlation does not equal causation. Two variables being strongly correlated does not demonstrate that one causes the other.

  • Classic Third-Variable Example: Ice cream sales and drowning deaths show an extremely strong positive correlation ($r = +0.82$). Ice cream consumption does not cause drowning; both variables fluctuate due to an unaccounted third variable (warm weather in summer leads to both increased ice cream purchases and increased swimming activity).

Spurious Correlations

• Definition: Statistically robust correlations occurring between variables that have no logical, functional, or causal relationship.

• Documented Examples of Spurious Correlations:

  • Popularity of the "I'm on a Boat" internet meme correlates strongly with the number of executive administrative assistants in the state of Alabama.

  • Total associate degrees awarded in history correlate strongly with Centene stock prices.

  • Number of mobile heavy equipment mechanics in Maine correlates strongly with Verizon customer satisfaction ratings.

  • Number of movies starring Will Smith correlates strongly with solar energy generated in the United States.

  • UFO sightings in Utah correlate strongly with US patents granted to parents.

  • Popularity of the first name "Lane" correlates strongly with the number of merchandise displayers and window trimmers in Alaska.

  • Popularity of the first name "Stevie" correlates strongly with Lululemon stock performance/sales.

  • Associate degrees awarded in engineering correlate strongly with Google search volume for Daylight Savings Time.

Personal Background on Mathematics and Statistics

• Failed high school trigonometry twice; passed the New York Regents exam on the third attempt in junior year with a score of $66$

• Completed college statistics in high school to fulfill undergraduate math requirements without taking undergraduate psych statistics.

• Encountered Inferential Statistics 102 in the second year of a doctoral program, mastering computational statistics via the R programming language.

Methodological Biases and Controls

• Sampling Bias

  • Occurs when a sample is not representative of the overall population from which it is drawn.

  • Example: Evaluating instructional efficacy based on optional final exam scores. If an optional final exam policy allows students to drop their lowest grade, only a small subset (e.g., $20\%$ of a class, or $50$ out of $300$ to $350$ students) will take it. This subset represents struggling or stressed students, creating sampling bias.

  • Example: Assessing perceptions of race across US colleges by sampling exclusively an elite, predominantly white institution.

• Placebo Effect

  • Occurs when subjects' expectations lead them to experience physiological or psychological change despite receiving a fake, inactive, or non-empirical treatment.

• Social Desirability Bias

  • The tendency of survey respondents to answer questions in a manner that will be viewed favorably by others or align with social norms (e.g., self-reporting as a "kind person" on self-report instruments regardless of true behavioral tendencies).

• Experimenter Bias

  • Occurs when a researcher's expectations or hypotheses regarding study outcomes unintentionally influence the treatment of subjects or the interpretation of data.

• Double-Blind Procedure

  • A control procedure in which neither the subjects nor the researchers interacting with them know which participants are in the experimental group or the control group. Frequently used in oncological (cancer) clinical trials to eliminate bias.

Peer-Review Process and Scientific Journals

• Scientific Journals

  • Periodicals that publish peer-reviewed technical and scholarly articles within highly specific academic disciplines.

• Peer-Review Workflow

  • Manuscripts submitted to a journal undergo initial editorial screening (desk review/rejection).

  • Papers passing initial screening are assigned to three independent domain experts who volunteer to critique the study design, methodology, data analysis, and empirical rigor.

  • Reviewers issue feedback, requiring authors to make iterative revisions before a manuscript achieves final acceptance for publication.

• Replication

  • High methodological rigor and precise operational definitions enable independent researchers to replicate studies exactly, verifying whether findings hold true across diverse contexts.