Chapter 2 – Psychological Research (OpenStax)

Why is Research Important

• Importance of research in psychology: informs understanding of behavior and mental processes; guides decision-making in education, policy, and practice.

• Use of research information: evidence-based conclusions, better public policies, and improved interventions.

• The Process of Scientific Research: knowledge is advanced through a cycle of proposing hypotheses, conducting studies, analyzing results, and creating or modifying theories based on findings; this often involves iterative refinement and replication to establish reliability.

Approaches to Research

• Clinical or Case Studies: involve detailed study of one person or a small group; provide rich, in-depth information but have limited generalizability.

• Naturalistic Observations: observing subjects in their natural environment without intervention; example: Jane Goodall’s chimpanzee studies; highlights real-world behavior but less control over variables.

• Surveys: collect self-reported data from a sample; useful for broad patterns but rely on honest reporting; issues include sampling and wording effects.

• Archival Research: examining existing records or data; described as the simplest form of research approach.

• Longitudinal vs. Cross-sectional Research:

  • Longitudinal: follows the same group over years or more; strengths include observing development and change over time; weaknesses include attrition/drop-out and time requirements.

  • Cross-sectional: compares different age groups at one point in time; faster and cheaper but may be confounded by cohort effects.

The Scientific Method and Hypotheses

• Scientific method in psychology: proposing hypotheses, conducting empirical investigations, and building or revising theories based on results.

• A good hypothesis should be:

  • testable and falsifiable (able to be proven wrong)

  • often stated as an if-then statement

  • Freuds’ concepts of id, ego, and superego are not falsifiable, illustrating why falsifiability matters.

• Example of a testable hypothesis in psychology: H: ext{If } X ext{ is manipulated, then } Y ext{ will change in a specified direction.}

Inductive and Deductive Reasoning in Research

• Inductive reasoning: derives general conclusions from specific observations.

• Deductive reasoning: tests a theory by deriving specific predictions from it.

• Circular relationship: inductive and deductive processes reinforce each other in scientific inquiry; theories guide observations, which in turn refine theories.

• Figure references: a cycle showing how hypotheses lead to research, which leads to results that inform and modify theories.

Examples and Figures Mentioned

• Figure 2.2 Trephination: historical practice illustrating early approaches to understanding behavior and brain function.

• Figure 2.3 D.A.R.E. program: popular in schools but research suggests it is ineffective; prompts discussion on translating findings into public policy.

• Figure 2.4 Psychological research relies on both inductive and deductive reasoning (circular relationship).

• Figure 2.5 Scientific knowledge is advanced by the scientific method (proposing hypotheses, conducting research, modifying theories).

Hypotheses and Concepts of Testing

• What is a good hypothesis? Should be testable and falsifiable; Freud’s id/ego/superego example shows non-falsifiability.

• Hypotheses are often formulated as if-then statements to specify expected relations.

Approaches to Research: Detailed Designs

• Clinical or Case Studies: provide detailed information about a single case; not generalizable but can generate hypotheses.

• Naturalistic Observations: observe natural behavior (Jane Goodall example); high ecological validity but less control over variables.

• Surveys:

  • Random Sampling: participants should be randomly sampled from the target population to ensure representativeness; large samples improve generalizability.

  • Wording Effects: subtle changes in wording can shift responses (e.g., government prohibition vs. government forbiddance).

  • Dewey Defeats Truman (1948): historical example illustrating how sampling or question phrasing can mislead results.

• Archival Research: examining existing records; described as the simplest form of archival data collection.

Longitudinal vs Cross-Sectional Studies: Pros and Cons

• Longitudinal studies: track the same individuals over time; strengths include analyzing development and change; drawbacks include attrition and longer durations.

• Cross-sectional studies: compare different ages/groups at one time; quicker and cheaper but susceptible to cohort effects.

Analyzing Findings: Correlation and Causation

• Correlational Research: examines how two variables vary together; correlation does not imply causation.

• Illusory Correlations: perceiving a relationship where none exists, often reinforced by confirmation bias.

• Causality in research requires experimental manipulation and control of confounding variables.

• The Experimental Hypothesis: a specific testable prediction about how manipulating one or more factors will affect behavior or mental processes.

• Variables in experiments:

  • Independent Variable (IV): the factor that is manipulated; the presumed cause.

  • Dependent Variable (DV): the factor that is measured; the presumed effect.

  • Confounding Variable: another factor that could influence the DV, potentially biasing results.

• Selecting and Assigning Experimental Participants: ensuring groups are comparable and that assignment minimizes preexisting differences.

• Interpreting Experimental Findings: use statistical analysis to determine whether group differences are meaningful beyond chance.

• Reliability and Validity: reliability = consistency of results; validity = whether the measurement actually measures what it intends to measure. Cultural differences may affect validity.

• Relationship: Validity implies reliability; however, reliability does not necessarily imply validity. (In other words, a measure can be consistently wrong.)

Correlation and Its Visualization

• Correlation Coefficient R: measures the strength and direction of the relationship between two variables.

  • Range: R \in [-1.00, +1.00]

  • R = 0 indicates no linear relationship; R = +1 is a perfect positive relationship; R = -1 is a perfect negative relationship.

• Scatterplots: graphical representation of the strength and direction of correlations; data points closer to a straight line indicate stronger relationships.

  • Examples:

  • Positive correlation: weight and height (as one increases, the other tends to increase).

  • Negative correlation: tiredness and hours of sleep (more sleep, less tired).

  • No correlation: shoe size and hours of sleep.

• Correlation ≠ Causation: association does not reveal which variable causes the other.

Illusory Correlations and Confirmation Bias

• Illusory correlations: perceived relationships that do not exist; common with selective attention or biased interpretation.

• Confirmation bias: tendency to search for, interpret, and recall information that confirms preconceptions while ignoring contrary evidence.

Conducting and Interpreting Experiments

• Experiment: researchers vary one or more factors (IV) to observe effects on behavior or mental processes (DV); aim to control other factors by random assignment.

• Manipulating factors of interest and holding others constant helps isolate effects of the IV.

• Example focus: determining which child behaviors increase after exposure to violent television content.

What hypothesis are you going to test? Experimental Formulation

• Hypotheses can come from careful observation or a literature review.

• Example: Are children more likely to display violent behaviors after viewing violent TV programming?

• Experimental Hypothesis: a specific, testable prediction about the effect of the IV on the DV.

Experimental Design: Groups and Definitions

• Experimental Group: receives the treatment (e.g., views violent TV program).

• Control Group: does not receive the treatment (e.g., views nonviolent TV program).

• Operational Definitions: explicit description of how variables are measured and manipulated (e.g., what counts as "violent behaviors": punching, toy guns, kicking).

Controlling Bias and Ensuring Rigor

• Experimenter Bias: researchers’ expectations may influence results.

• How to mitigate bias: single-blind or double-blind procedures.

• Double-Blind Procedure: both participants and experimenters are unaware of who receives the treatment or placebo.

• Placebo: an inactive substance or condition given to control group.

• Placebo Effect: observed effects due to participants’ expectations rather than the treatment.

Experimental Variables: Definitions and Design Considerations

• Independent Variable: factor manipulated to observe its effect.

• Dependent Variable: measured outcome.

• Confounding Variable: extraneous variable that could influence results.

• In an experiment, manipulations of the IV are expected to cause changes in the DV.

Practice and Application

• Example practice item: Can people taste the difference between red and yellow Starburst during a blind taste test when their nose is closed? (Illustrates hypothesis formation, groups, and variables.)

Selecting and Assigning Experimental Participants

• Participants are the subjects of psychological research.

• Population: all individuals to be studied (e.g., all FSCJ students).

• Sample: a subset of the population; must be randomly selected and large enough to generalize results.

• Population vs Sample distinction is crucial for generalizability and external validity.

• Random Sampling: each person in the population has an equal chance of being selected.

• Random Assignment: after sampling, participants are assigned to experimental and control groups by chance to minimize preexisting differences.

Variability in Sampling and Group Assignment

• Large populations may require practical sampling strategies; samples should still be representative.

• Figure 2.18 summarizes that researchers may work with a large population or a sample group that is a subset of the population.

• Quasi-experimental designs: when random assignment is not possible or ethical (e.g., comparing smokers vs non-smokers); these designs have limitations for causal inference.

Analyzing Experimental Findings: Statistics and Inference

• Statistical Analysis: determines whether differences between groups are meaningful and not due to random chance.

• Confounds and limitations must be acknowledged; replication strengthens reliability of findings.

Reporting Research

• Peer-reviewed journal articles: critical review process ensures quality and replicability; reviews and potentially conducts related research to verify findings.

• Replication: essential for establishing reliability and generalizability of results; each replication adds evidence for the original findings.

• APA Style (2019): recommended for psychology majors; includes guidelines for writing and attribution; APA also endorses the use of the singular "they" as a gender-neutral pronoun.

• References: proper citation format (example shown):

  • AuthorLastName, FirstInitial., & AuthorLastName, FirstInitial. (Year). Title of article. Title of Journal, Volume(Issue), Page Number(s). https://doi.org/number

Ethical Considerations in Psychological Research

• Ethics: research must be ethical throughout design, conduct, and review; distinguishes human participants and animal subjects.

• IRB: Institutional Review Board; committee of administrators, scientists, and community members that reviews proposals for research involving human participants.

• Informed Consent: informs participants about what to expect, potential risks, and implications; participation must be voluntary; data handling is confidential.

• Deception: may be used in some studies to maintain integrity, but must be followed by full debriefing at study conclusion where participants receive complete and truthful information about the study.

• Debriefing: post-study explanation addressing deception, study purpose, and participants’ questions.

• Research involving animals: protected by IACUC (Institutional Animal Care and Use Committee); ensures humane treatment and welfare of animals.

Reliability, Validity, and Threats to Integrity

• Reliability: consistency of results under similar conditions; measured by statistical analyses such as inter-rater reliability (agreement among observers).

• Validity: accuracy of the instrument in measuring what it is intended to measure; cultural differences can affect validity.

• A measure that is valid is also reliable; however, a reliable measure need not be valid.

Special Topics and Real-World Relevance

• Vaccinations and public health: Figure 2.19 notes that some people still believe vaccines cause autism; research led to retractions of some studies due to financial conflicts of interest; emphasizes ethical and scientific standards in publishing and public policy.

Ethics in Animal Research and Human Research: Summary

• IRB oversees human research ethics; informed consent and risk disclosure are core components.

• Deception requires debriefing; participant welfare is prioritized.

• IACUC protects animal welfare in research settings.

Quick Reference: Key Terms and Concepts

• Population vs. Sample

• Random Sampling vs. Random Assignment

• Independent Variable (IV) vs. Dependent Variable (DV)

• Confounding Variable

• Operational Definition

• Reliability vs. Validity

• Inter-rater Reliability

• Correlation vs. Causation

• Illusory Correlations and Confirmation Bias

• Quasi-experimental Design

• Statistical Significance and Replication

• Peer Review and APA Style

• IRB and Informed Consent

• Debriefing and Deception

• IACUC and Animal Research

Notes on Figures and Examples from the Text

• Figure 2.12: Scatterplots illustrate strength and direction of correlations (positive, negative, none).

• Figure 2.17: In an experiment, IV manipulations are expected to yield DV changes.

• Figure 2.18: Illustration of population vs. sample in research.

• Figure 2.19: Vaccination-autism discussion and issues surrounding publication and retraction.

Formulas and Notation

• Correlation coefficient: R \in [-1, 1] with interpretation:

  • R = +1.00: perfect positive correlation

  • R = -1.00: perfect negative correlation

  • R = 0.00: no linear relationship

• Hypothesis form (example): H: \text{If } X \text{ is manipulated, then } Y \text{ will change (direction specified).}

• Key relationships:

  • Independence: manipulation of the IV to observe effect on the DV

  • Control variables: holding constant to isolate IV effects