📊Experimental Design Unit 2 – Principles of Experimental Design

Key Concepts and Terminology

• Experimental design involves planning and conducting experiments to test hypotheses and draw conclusions
• Independent variable (IV) manipulated by the researcher to observe its effect on the dependent variable (DV)
• Dependent variable (DV) measured or observed to determine the effect of the independent variable
• Control group does not receive the experimental treatment, serving as a baseline for comparison
• Experimental group receives the treatment or intervention being tested
• Confounding variables extraneous factors that may influence the dependent variable, potentially distorting the results
  • Confounding variables should be controlled for through randomization, matching, or statistical techniques
• Randomization assigns participants to groups by chance, helping to distribute potential confounds evenly across groups
• Blinding conceals group assignment from participants (single-blind) or both participants and researchers (double-blind) to minimize bias

Types of Experimental Designs

• Between-subjects design compares different groups of participants, each exposed to a different level of the independent variable
• Within-subjects design exposes each participant to all levels of the independent variable, allowing for comparisons within individuals
• Factorial design manipulates two or more independent variables simultaneously to examine main effects and interactions
  • Main effect influence of one independent variable on the dependent variable, ignoring other variables
  • Interaction effect occurs when the effect of one independent variable depends on the level of another independent variable
• Repeated measures design assesses the same participants under different conditions or at multiple time points
• Counterbalancing varies the order of conditions to control for potential order effects
• Quasi-experimental design lacks random assignment but still manipulates an independent variable (pre-existing groups)
• Natural experiment takes advantage of naturally occurring events or circumstances to study their effects

Variables and Their Roles

• Independent variable (IV) manipulated by the researcher to observe its effect on the dependent variable
• Dependent variable (DV) measured or observed to determine the effect of the independent variable
• Confounding variables extraneous factors that may influence the dependent variable, potentially distorting the results
• Moderator variables influence the strength or direction of the relationship between the independent and dependent variables
  • Moderators can be categorical (gender) or continuous (age)
• Mediator variables explain the mechanism or process through which the independent variable affects the dependent variable
• Control variables held constant to minimize their influence on the dependent variable
• Operationalization defines variables in measurable terms, specifying how they will be manipulated or assessed

Sampling Techniques

• Population entire group of individuals or objects of interest for the research question
• Sample subset of the population selected for the study, ideally representative of the larger group
• Simple random sampling selects participants from the population with equal probability
• Stratified random sampling divides the population into subgroups (strata) and then randomly samples from each stratum
  • Ensures representation of key subgroups in the sample
• Cluster sampling divides the population into clusters (naturally occurring groups) and then randomly selects clusters to include in the sample
• Convenience sampling selects participants based on their availability and willingness to participate
• Purposive sampling selects participants based on specific characteristics or criteria relevant to the research question
• Sample size number of participants included in the study, influencing statistical power and generalizability

Controlling for Bias and Confounds

• Randomization assigns participants to groups by chance, helping to distribute potential confounds evenly across groups
• Blinding conceals group assignment from participants (single-blind) or both participants and researchers (double-blind) to minimize bias
• Placebo effect occurs when participants' expectations or beliefs about treatment lead to improvements, even with an inactive substance
• Hawthorne effect occurs when participants' behavior changes due to awareness of being observed
• Demand characteristics cues that may influence participants' behavior to align with perceived expectations of the study
• Experimenter bias occurs when researchers' expectations or beliefs unintentionally influence the study's outcomes
  • Using standardized protocols and double-blinding can help minimize experimenter bias
• Confounding variables extraneous factors that may influence the dependent variable, potentially distorting the results
  • Confounding variables should be controlled for through randomization, matching, or statistical techniques

Data Collection Methods

• Surveys and questionnaires collect self-reported data from participants using a set of predetermined questions
• Interviews gather in-depth information through direct conversation with participants
  • Interviews can be structured (fixed questions), semi-structured (mix of fixed and open-ended questions), or unstructured (open-ended, exploratory)
• Observations involve systematically watching and recording behavior or events in a natural or controlled setting
• Physiological measures assess biological or physical responses (heart rate, brain activity, skin conductance)
• Behavioral measures quantify observable actions or responses (reaction time, accuracy, frequency)
• Archival data pre-existing data collected for other purposes (medical records, census data, public documents)
• Triangulation uses multiple methods or sources to corroborate findings and enhance validity

Statistical Analysis Basics

• Descriptive statistics summarize and describe the main features of a dataset (mean, median, mode, standard deviation)
• Inferential statistics use sample data to make inferences or draw conclusions about the larger population
• Hypothesis testing evaluates the likelihood of observed results occurring by chance, assuming the null hypothesis is true
  • Null hypothesis ($H_0$) states that there is no significant effect or relationship between variables
  • Alternative hypothesis ($H_1$ or $H_a$) states that there is a significant effect or relationship between variables
• $p$-value probability of observing results as extreme as those in the sample, assuming the null hypothesis is true
  • Smaller $p$-values provide stronger evidence against the null hypothesis
• Statistical significance occurs when the $p$-value falls below a predetermined threshold (usually $\alpha = .05$), indicating a low likelihood of results occurring by chance
• Effect size quantifies the magnitude or strength of the relationship between variables (Cohen's $d$, Pearson's $r$)
• Confidence intervals range of plausible values for a population parameter, based on the sample data and desired level of confidence

Ethical Considerations in Experiments

• Informed consent ensures that participants understand the nature, purpose, and potential risks of the study before agreeing to participate
• Confidentiality protects participants' identities and personal information from being disclosed without their permission
• Anonymity collects data without any identifying information, so individual responses cannot be linked to specific participants
• Debriefing informs participants about the true nature and purpose of the study after their participation
  • Debriefing is especially important when deception has been used
• Deception intentionally withholding information or providing false information to participants to maintain the integrity of the study
  • Deception should only be used when necessary and with appropriate safeguards in place
• Minimizing harm researchers must take steps to ensure that participants are not exposed to undue physical, psychological, or emotional risks
• Balancing risks and benefits potential benefits of the research should outweigh any risks to participants
• Institutional Review Board (IRB) reviews and approves research proposals to ensure they meet ethical standards and protect participants' rights and welfare