Comparison of Experimental and Non-Experimental Factorial Designs

Example of Causal Limitations in a Non-Experimental Factorial Design

In a 2 x 2 factorial study predicting behavior based on measured mood (positive vs. negative) and self-esteem (high vs. low), why can't the researchers establish a causal relationship?

In a study predicting individuals' willingness to have unprotected sex based on their measured mood (positive vs. negative) and measured self-esteem (high vs. low), match each element of the study to its correct methodological description.

A psychologist conducts a 2 x 2 factorial study to investigate how measured optimism (high vs. low) and measured sleep quality (good vs. poor) relate to cognitive performance. If the results show a significant difference in performance scores, the psychologist can safely conclude that the participants' level of optimism caused the change in their performance.

Arrange the steps in the logical analysis of a research design that evaluates the relationship between participants' measured mood (positive vs. negative) and measured self-esteem (high vs. low).

In the hypothetical study provided as an example of a $2 \times 2$ non-experimental factorial design, which two variables serve as the non-manipulated factors?

Based on the research example involving measured mood and self-esteem, arrange these design components in order from the most general organizational structure to the most specific experimental element.

A researcher reviews a study that utilizes a $2 \times 2$ factorial design to examine how measured mood and measured self-esteem relate to individuals' willingness to engage in unprotected sex. In evaluating the scientific merit of the conclusion that 'negative mood causes riskier behavior,' a critic would argue that the study lacks the _____ of independent variables required to establish a causal relationship.

A researcher designs a study on mood and academic motivation. She randomly assigns half of the participants to watch an uplifting video clip (positive mood condition) and the other half to watch a neutral clip (neutral mood condition), then measures each participant's pre-existing self-esteem level to create a $2 \times 2$ design. This study is best classified as a fully non-experimental factorial design, just like the hypothetical mood × self-esteem willingness study in which neither variable is manipulated.

In the hypothetical $2 \times 2$ factorial study predicting willingness to have unprotected sex from participants' measured mood (positive vs. negative) and self-esteem (high vs. low), match each specific feature of the study to the methodological concept it best illustrates.

A public health researcher reads the hypothetical $2 \times 2$ factorial study on mood, self-esteem, and willingness to engage in unprotected sex, then concludes that mood is a proven causal risk factor and proposes mood-based intervention programs. A peer reviewer argues this conclusion is unjustified because the non-experimental design cannot rule out _____ variables — unmeasured factors that could independently influence both participants' mood and their willingness — as alternative explanations for the observed association.

Based on the example of the hypothetical study predicting willingness to have unprotected sex, describe its design by identifying the independent variables, their levels, and their classification. What is the primary causal limitation of this design?

Diagnose the error in the researcher's conclusion. Explain how the non-experimental nature of the factorial design in this study limits her ability to make this causal claim.

Suppose a researcher wants to modify the hypothetical study to determine if mood has a causal effect on willingness to have unprotected sex. Applying your knowledge of experimental methods, how should the researcher change the implementation of the mood variable?