Empirical Approach, Scientific Goals, and Research Ethics in Psychology

Empiricism

Empiricism is the approach to knowledge that says we should rely on systematic observation and measurement of the world (data) rather than only on intuition, authority, or anecdote.

Importance of empirical approaches

They replace guesswork with evidence, reduce the influence of biases, produce quantifiable, falsifiable predictions, and enable cumulative knowledge.

Confirmation bias

Confirmation bias is the tendency to favor evidence that confirms what we believe.

Availability/recency bias

Availability/recency bias is the tendency to overweight vivid anecdotes.

Experimenter expectancy / demand characteristics

These are biases that can be reduced with blinding and standardized procedures.

Illusory correlations

Illusory correlations occur when a relationship is perceived when none exists.

Description (goal of scientific research)

Description involves defining and cataloging behavior or phenomena.

Example of Description

A researcher documents how many people on a college campus use a given social app and in what contexts.

Prediction (goal of scientific research)

Prediction involves identifying reliable associations to guess future behavior.

Example of Prediction

Finding that higher daily stress scores predict sleep problems the next night.

Determination of cause (causation)

This shows that changes in one variable produce changes in another.

Example of Determination of cause

Randomly assigning participants to receive either cognitive training or an active control, then seeing which group improves on reasoning tests.

Explanation (goal of scientific research)

Explanation involves developing theories that explain why and how something happens.

Example of Explanation

A theory that chronic stress causes immune dysregulation via HPA-axis activation.

Temporal precedence

Temporal precedence means the cause must occur before the effect.

Example of Temporal precedence

If increased study time causes better grades, study time must increase before the grades improve.

Covariation (association)

Covariation means when the cause changes, the effect also changes.

Example of Covariation

As nicotine exposure increases, lung cancer incidence increases.

Elimination of alternative explanations

This means other plausible causes must be ruled out.

Example of Elimination of alternative explanations

Controlling for temperature shows that ice cream doesn't cause drowning — temperature is a confound.

Basic research

Basic research is aimed at expanding fundamental knowledge or testing theory without immediate practical application in mind.

Applied research

Research aimed at solving practical problems or improving real-world outcomes. Example: testing an intervention to reduce medication nonadherence.

Contrast between Basic and Applied research

Basic: theory-driven, controlled conditions, seeks general principles. Applied: problem-driven, often in real-world settings, seeks solutions.

Contributions to trust in nonscientific claims

Why people believe bad claims and how to evaluate them.

Authority appeal

Someone presented as an 'expert' — celebrity endorsements.

Anecdotes & vivid stories

Emotionally compelling single cases.

Repetition

Hearing a claim many times increases believability.

Illusory correlations and pattern-seeking

Seeing relationships where none exist.

Jargon and pseudo-technical language

Sounds scientific.

Social proof and echo chambers

Many peers accept it.

Poor statistical literacy

Misunderstanding probability, base rates.

Practical checklist to assess and reduce trust in false claims

Look for replication and independent confirmation.

Peer-reviewed sources

Check peer-reviewed sources and the quality of the outlet.

Operational definitions

Ask for operational definitions and measurement details.

Effect sizes

Evaluate effect sizes, not just 'statistical significance'.

Conflicts of interest

Search for conflicts of interest (funding, vested parties).

Systematic reviews/meta-analyses

Prefer systematic reviews/meta-analyses over isolated studies.

Post hoc explanations

Beware of post hoc explanations (after-the-fact rationalizations).

Sources of research ideas

Common sense / folk psychology (may suggest hypotheses but can be misleading).

Practical problems

e.g., how to reduce relapse rates.

Observations of the world

Clinical observations, naturalistic observation.

Existing theories

Generate novel predictions.

Past research / literature gaps

Replications, extensions.

Serendipity

Unexpected findings.

Media / social trends

New behaviors to study.

Functions of a scientific theory

Organize known facts — a theory explains and organizes empirical findings into a coherent framework.

Generate testable predictions

Theories produce hypotheses that can be empirically tested (falsifiability).

Types of research reports

Three kinds of research reports: Literature reviews, Theory articles, Empirical research articles.

Typical sections of an empirical paper

Abstract — brief summary (background, methods, key results, main conclusion).

Introduction section

Background literature, theory, rationale, research question(s), hypotheses.

Method section

Detailed description so others can replicate: participants (who, how many, sampling), materials/measures (operational definitions), design, procedure, ethics.

Results section

Analyses, statistics, effect sizes, tables/figures; report exactly what was found (no interpretation).

Discussion section

Interpret results, relate to theory, discuss limitations, practical implications, and directions for future research.

APA 7th edition journal-article reference

Author, A. A., & Author, B. B. (Year). Title of the article: Subtitle if present. Title of Journal, volume number(issue number), page range.

Example of APA reference

Smith, J. A., & Lee, R. B. (2021). Social media use and sleep quality in college students. Journal of Behavioral Health, 37(2), 112-128.

Scientific method

Steps including asking a question, reviewing literature, formulating a hypothesis, designing the study, collecting data, analyzing data, interpreting conclusions, communicating results, and replicating findings.

Good research question

Qualities include being clear and focused, feasible, specific about population and variables, grounded in theory, and ethical.

Good hypothesis

Qualities include being testable and falsifiable, specific about variables and expected direction, operationalized, and predicting a pattern of results.

Peer review

The evaluation of research manuscripts by independent experts before publication to ensure quality, validity, and novelty.

Benefits of peer review

Quality control, methodological improvement, filtering out obviously flawed work.

Limitations of peer review

Potential reviewer bias, slow process, occasional failure to detect fraud, variability in review standards.

Predatory publishers

Publishers or journals that exploit authors for fees while providing little or no peer review or editorial standards.

Signs of predatory journals

Aggressive unsolicited emails, promises of fast peer review or guaranteed acceptance, vague editorial board names, low-quality websites, high article processing charges.

How to avoid predatory journals

Check journal indexing, confirm editorial board members, read sample articles for quality, check journal policies, consult tools like DOAJ.

Impact factor

A metric that averages citations per article, often used as a proxy for journal prestige but has limitations.

PsycINFO

A major bibliographic database indexing psychological literature, useful for literature searches.

Belmont Report

A foundational U.S. ethical statement identifying three principles: respect for persons, beneficence, and justice.

APA Ethics Code

Five principles: A: Beneficence & Nonmaleficence, B: Fidelity & Responsibility, C: Integrity, D: Justice, E: Respect for People's Rights & Dignity.

Informed consent

Participants must be given adequate information, understand it, and voluntarily agree.

Operationalize variables

The process of defining how variables will be measured or manipulated in a study.

Collect data

The step in the scientific method where information is gathered following ethical standards.

Analyze data

The step in the scientific method where appropriate statistics are applied to the collected data.

Interpret & draw conclusions

The step in the scientific method where findings are assessed within the bounds of validity.

Communicate results

The step in the scientific method involving sharing findings through papers or conferences.

Replicate / extend

The process of confirming findings through repeated studies or expanding upon them.

Empirical characteristics

Features of research that are based on observation and experimentation.

Systematic characteristics

Features of research that follow a structured and methodical approach.

Testable characteristics

Features of research that can be supported or refuted through experimentation.

Self-correcting characteristics

Features of research that allow for adjustments and improvements based on new evidence.

Communal characteristics

Features of research that involve peer review and replication by the scientific community.

Assent

agreement from minors or those who cannot give legal consent, combined with guardian permission.

Coercion

undue pressure or inducement — ethically problematic.

Deception

withholding or misrepresenting study details; allowed only when necessary, no alternative, and not causing harm.

Debriefing

explaining the true purpose & procedures after participation, correcting misconceptions, and offering support/referrals if needed.

Confidentiality

Protect personal data, anonymize when possible, securely store data, explain limits to confidentiality (e.g., mandatory reporting).

IRB (Institutional Review Board)

Function: review proposed human-subjects research to ensure ethical treatment and regulatory compliance.

Exempt review

minimal risk, standard educational practices or anonymous surveys.

Expedited review

minimal risk but falls under specific categories (biological samples, noninvasive measures).

Full review

greater-than-minimal risk, vulnerable populations, invasive procedures.

Minimal risk

Risk not greater than those encountered in daily life or during routine physical or psychological exams.

Research fraud

fabrication (making up data), falsification (manipulating data), selective reporting.

Plagiarism

presenting others' words/ideas as your own — includes verbatim copying and improper paraphrase.

Avoidance of plagiarism

careful citation, paraphrasing with attribution, keeping raw data, preregistration, open methods and data where appropriate.

Milgram obedience studies

demonstrated willingness to obey authority to administer apparent shocks; raised ethical issues about stress to participants and informed consent.

Asch conformity experiments

showed social pressure can lead people to give incorrect answers to conform with group.

Elevator experiment

demonstrates conformity in public spaces.

Tuskegee syphilis study

egregious violation — participants withheld effective treatment; central to modern research protections.

Omission

withholding information (e.g., not telling participants every detail).

Commission

actively providing false information (deception). Both have ethical implications; must justify and debrief.

Construct validity

The degree to which a test or measure actually measures the theoretical construct it claims to measure.