Confirmation Bias and Basic Research

Natural world

The real, touchable world

Empirical questions

Questions answerable by observation

Systematic observation

Produces quantitative and qualitative data

Quantitative data

Numbers, graphs, statistics

Parsimonious explanation

Simplest accurate explanation

Provisional findings

Initial findings

Protoscience

Fringe science at the edges of current scientific understanding

Nonscience

Applies systematic techniques but doesn't use the scientific method

Confirmation bias

Only seeking information that confirms beliefs and ignoring contradictory evidence

Basic research

Focuses on advancing knowledge

Applied research

Focuses on solving a specific problem

Commonsense explanation

Limited explanation based on observed event and previous experience

Scientific explanation

Constructed using the scientific method, generalizable and addresses observable questions

Science

Activities aimed at producing systematic knowledge about the natural world

Scientific method

Multiple methods followed by scientists, not a mechanical set of steps

Pseudoscience

Ideas presented as science but depart from the scientific method

Confirmational strategy

Looking for evidence to confirm predictions from a theory, should also use disconfirmational strategy

Characteristics of scientific explanations

Empirical, rational, testable, parsimonious, and general

Circular explanation

Informal fallacy where the conclusion is not materially different from an assumed premise

Tautology

A statement always true by virtue of its terms, with no propositional content

Method of authority

Involves consulting an authoritative source, not always providing valid answers

Rational method

Involves doubting beliefs, searching for 'self-evident' truths, and deducing conclusions

Theory

Well-developed set of ideas proposing an explanation for observed phenomena

Laws

Widely accepted theories with no significant rivals in explaining facts within their domain

Hypothesis

Tentative and testable statement based on a larger theory, predicts behavior if theory is correct and is falsifiable

Model

Representation of a concept or basic processes

Constructs

Explanatory model based on verifiable and measurable events or inferred processes

Mechanistic explanation

Assumption that psychological processes can be understood like mechanical or physiological processes

Functional explanation

Explains the existence of an entity or process by the functions it carries out

Quantitative theory

Study and development of methods and techniques to measure human behavior and attributes

Qualitative theory

Involves qualitative data

Domain of a theory

The overall genre of a theory

Scope of a theory

The number and diversity of phenomena explained or interpreted by a theory

Testability

The degree to which a hypothesis or theory can be evaluated empirically

Prediction of novel events

Predicting events not previously observed

Parsimony

The preference for the simplest explanation of an event or observation

Logically valid argument

An argument that makes it impossible for the premises to be true and the conclusion false

Deduction

The process of inferential reasoning

Modus ponens

Deductive argument form and rule of inference: P implies Q, P is true, therefore Q is true

Modus tollens

If P implies Q, and Q is false, then P is false

Affirming the consequent

Drawing a conclusion that if the consequent is true, then the antecedent must also be true

Auxiliary assumptions

Assumptions about the non-occurrence or opposite effect in data, indicating a problem with the hypothesis or the assumptions

Induction

Drawing general principles from specific observations and cases

Inductively strong

If true premises make the conclusion highly unlikely to be false

Falsifiability

The logical possibility that an assertion, hypothesis, or theory can be shown false by observation or experiment

Positive analogy

Comparing two alike things, e.g., tree is to leaf as flower is to petal

Negative analogy

Comparing two things by their dissimilarities

Neutral analogy

Reflects ignorance about features at both levels

Strong inference

Testing a hypothesis by attempting to demonstrate its falsity and modifying the theory based on research outcomes

Nuremberg code

Ethical guidelines for human research, including voluntary participation and informed consent

Declaration of Helsinki

Ethical guidelines for medical research with human subjects, based on relevant scientific literature and independent review

Belmont Report

Basic ethical principles and guidelines for resolving ethical problems in research with human subjects

Respect for persons

Requires treating individuals as autonomous human beings and not using them as a means to an end

Beneficence

Minimize harm and maximize potential benefits of research

Justice

Design research for equitable sharing of burdens and benefits

APA ethical guidelines

Includes beneficence, nonmalfeasance, fidelity, integrity, justice, and respect

Institutional review board

Committee reviewing research consistency with Belmont report

Risk/ratio benefit

Subjective evaluation of risks and benefits for research projects

Informed consent

Voluntary agreement based on understanding of procedure, benefits, risks

Deception

Withholding or misinforming participants, violating informed consent

Animal research

Research involving animals

Researcher misconduct

Fraudulent behavior eroding public confidence in scientific research

Data fabrication

Making up data or results and reporting them

Falsification

Manipulating research materials or changing/omitting data

Plagiarism

Appropriating another person's ideas or processes

Research question

Questions with operational definitions, theoretical interest

Unsystematic observation

Observation of everyday behavior to discover general research ideas

Operational definition

Precisely described definition of a concept in terms of operations or measures

Primary source

Full report of a research study, including rationale

Secondary source

Summarizes info from a primary source, may be biased or incomplete

Refereed journal

Papers undergo review by two or more reviewers

Nonrefereed journal

Papers do not undergo review

File drawer phenomenon

Unpublished studies due to failure to obtain positive results

Peer review

Evaluation of scientific work by other qualified professionals

Title/title page

Informative title about the study or theory, includes researchers

Abstract

Short summary of the article's key points, including problem, method, findings, conclusion

Introduction

States authors' plan, problem importance, review of related research, theoretical context

Method

Describes operations, materials, variables, design, and control features of the study

Results

Summarizes important results in tables and figures, directs reader to relevant data

Discussion

Relates results to hypotheses, addresses other explanations and future steps

References

List of sources used in the study

Experiments

Systematic research study varying variables and observing results

Causation

Empirical relation between events or variables causing change in each other

Joint method of agreement and difference

Factor A causes outcome X when A leads to X and absence of A leads to absence of X

Ceiling effect

High scores make it impossible to determine differences between groups

Floor effect

Low scores due to task difficulty, making it impossible to find differences

Control variable

Independent variable held constant during an experiment

Confound

Two variables combine, making the effects of one indistinguishable from the other

Third-variable problem

Unmeasured variable causing changes in observed variables

Directionality problem

Difficulty specifying the direction of a causal arrow

Independent variable

Variable changed by the experimenter

Dependent variable

Variable changed by the independent variable

Internal validity

Methodological soundness and confound-free study allowing strong causal statements

Testing intact groups

Building a confound by testing intact groups

Random assignment

Solution to confounds by assigning subjects randomly

History

External events between pre and post test

Maturation

Internal changes between pre and post test

Regression to the mean

Predicted scores for extreme x scores regress to the mean

External validity

Generalizability of study findings to other populations, settings, and times

Demonstrations

Expose subjects to treatment condition, not showing causal relationship