PSYC217 Chapter 8 - Experimental Design Intro - The essence of experimental design is exposing participants to two different conditions that are identical except for the independent variable of interest, which you've manipulated. Confounding and Internal Validity - A critical advantage of experiments over non-experiments is that experiments can support inferences of causality (when conducted properly). - Experiments teach that altering one variable causes another variable to change. " Researchers manipulate the IV, creating groups that differ in the levels of that variable then examine the effect of manipulation on the DV > · All variables must be kept constant through various means including direct experimental control and random assignment (or an equivalent procedure). . If scores for the DV are different between groups (levels of the independent variable), the researcher concludes that the independent variable caused this difference in results. " Conclusion is based on the logic that if the only difference between the groups is the manipulated independent variable, then this difference must be the cause of any difference in the measured dependent variable. - Good experimental design allows researchers to make causal claims because competing, alternative explanations (e.g., confounding variables) are controlled or eliminated. · When the results of an experiment can confidently be attributed to the independent variable, the experiment is said to have high internal validity " To achieve high internal validity, the researcher must design and conduct the experiment so that only the independent variable can be the cause of the results. · Identifying and eliminating confounds and other threats to internal validity is often difficult. Planning a Basic Experiment - The simplest possible experimental design has two variables: 1. Independent variable " Has two levels or conditions (e.g., an experimental group and a control group), 2. Dependent variable. - Researchers must make every effort to ensure that the only difference between the two levels of the independent variable (the experimental group and the control group) is the manipulated variable. - There are also two broad classes of experiment. 1. Between-subjects design (also known as an independent groups design) · Different people experience the levels of the independent variable. · In this kind of experiment, the experimental and control groups are distributed between different people. 2. Within-subjects design (also known as a repeated measure design) · All the same people experience all levels of the independent variable. · For a within-subjects design, the experimental and control conditions are equally distributed within all participants.
- Any between-subjects experiment involves three broad steps: 1. Obtaining two approximately equivalent groups of participants 2. Introducing different levels of the independent variable to the participants · Steps 2 and 3 also hold true for within-subject designs. 3. Measuring the dependent variable. · Steps 2 and 3 also hold true for within-subject designs. - When there are only two levels of the independent variable, the between-subjects experimental design can be diagrammed like this: · Participants -> assigned to either experimental or control group (IV's) -> the DV is then measured.