Study Notes on Internal Validity and Related Research Methods

Evaluating Internal Validity of Experiments

General Overview

  • Internal validity refers to the degree to which causal conclusions can be drawn from a research study. It determines whether the results of an experiment can be attributed to the independent variable rather than other factors.

  • Key focus is on eliminating confounding variables and ensuring that the alteration in the dependent variable is a result of the manipulation of the independent variable.

Research Methods in Psychology

  • Refers to the methodologies employed in psychology to ensure rigorous testing of hypotheses and theories.

Events and Opportunities at Florida Atlantic University

Internship Workshop

  • Date: February 3rd, 2026

  • Time: 6:00 PM - 7:30 PM

  • Location: SR149

  • Hosted by: Student Government and the Career Center

  • Purpose: Interactive workshop to assist with finding internships using platforms such as Handshake. Participants will receive essential tips and information.

Career Week Activities

Pop-Up Closet

  • Date: February 4th, 2026

  • Time: 10:30 AM - 2:00 PM

  • Description: Opportunity to discover a professional wardrobe and obtain a high-quality headshot.

Networking Mastery Event

  • Date: February 5th, 2026

  • Time: 2:00 PM - 4:00 PM

  • Description: A fair to connect with industry professionals and learn about job, research, and internship opportunities.

Lunch and Learn Series

  • Event: AI Companions and the Future of Relationships

  • Hosted by: Harriet L. Wilkes Honors College

  • Date: February 25th, 2026

  • Time: 12:30 PM - 2:45 PM

  • Free lunch provided.

Author Series featuring Alexander Chee

  • Date: February 27th, 2026

  • Time: 4:00 PM

  • Description: Discussion on Chee's insights, his works, and career in writing.

Internal Validity: Threats and Concepts

Types of Threats to Internal Validity

  1. Between Subjects Designs

    • Selection Effect: Participants are systematically different across conditions.

    • Example: Different numbers of participants in each condition resulting in bias.

  2. Within Subjects Designs

    • Order Effects: The sequence of conditions may influence the results.

  3. All Designs

    • Design Confounds: A flaw in design where other variables vary alongside the independent variable.

    • Observer Bias: Observers may see what they expect to see, impacting construct validity.

    • Demand Characteristics: Participants alter their behavior based on what they believe is expected of them.

    • Placebo Effect: Improvement resulting from participants' beliefs about receiving treatment.

Evaluating Claims

  • Types of claims: Frequency, Association, and Causal.

  • Evidence Sources include:

    • Observational studies or polls.

    • Experiments: Quasi-experiments and correlational studies.

  • Evaluation of claims is done through the lens of four validities: Internal, Construct, External, and Statistical validity.

Design Confounds

Description

  • A design confound occurs when an extraneous variable varies alongside the independent variable, making it possible to mistakenly attribute effects to the independent variable.

  • Example of design confounds in between-subjects: Unequal distribution of participant experience levels.

Solutions to Design Confounds

  • Control Variables: Any variable that is intentionally kept constant across conditions to minimize variability.

  • Implementing random assignment or matched groups to evenly distribute participant characteristics, which mitigates systematic variability.

Selection Effects in Between Subjects

Definition & Examples

  • Occur when participants are different between experimental conditions, potentially skewing results.

  • Example of systematic variability where participants’ backgrounds differ significantly across groups.

Mitigation Strategies

  • Random Assignment: Allocating participants to conditions randomly to eliminate biases.

  • Matched Groups: Creating groups that are equivalent on certain characteristics.

Addressing Order Effects

Order Effects Defined

  • Refers to the impact that the sequence of conditions has on participant performance (e.g., practice, fatigue).

Counterbalancing

  • Counterbalancing involves presenting different sequences of conditions to control for order effects.

  • Helps ensure each condition has the same likelihood of appearing first or last in the sequence, thus managing practice and fatigue effects.

Observer Bias and Demand Characteristics

Observer Bias

  • Occurs when the researcher's expectations interfere with the observation process, impacting the integrity of collected data.

Solutions

  • Implement a Masked/Blind Design to minimize expectation effects.

  • Create Codebooks and follow strict observation protocols to standardize data collection.

Demand Characteristics

  • When participants alter their responses due to their perceptions of the study's purpose.

Solutions

  • Utilize double-blind designs to obscure purpose from both participants and researchers.

Placebo Effects

  • Participants may report improvement due to their belief in the efficacy of a treatment even when none is actually administered.

  • Solution: Implement a placebo control group to better assess the true efficacy of the intervention through comparison.

Upcoming Assignments & Reflections

Assignment 1

  • Focus: Introduce and summarize threats to internal validity, continuing from the last class.

Assignment 2

  • Participants analyze a given scenario about educational games and identify possible confounding variables.

Individual Reflection

  • Compare systematic and unsystematic variability, focusing on implications for internal validity.

Concluding Notes

  • Students are encouraged to familiarize themselves with various internal validity threats and consider potential solutions to ensure rigorous research designs in their studies.

  • Ongoing discussions and assignments will refine understanding of these research methodologies and their practical implications.