Exam 8: Experiments for Evaluation Research

Correct Answer:

Verified

See pages 211-216. Use Box 8.3 to help you choose a study design. Effects on validity:
No control group: cannot rule out plausible alternatives in determining causality, and can't know the true size of effects
Naturally occurring groups: bias in the composition of groups, the effect may be exaggerated or minimized, and can't know the extent of the effect of this bias on the results
No pretest: least damaging to validity
The more your design deviates from the classical experiment (combining these), the less valid the results

Correct Answer:

Answered by Examlex AI Copilot

Threats to internal validity refer to factors that can affect the accuracy of the results of an experiment or study. Some common threats to internal validity include history, maturation, testing, instrumentation, regression to the mean, selection bias, and experimental mortality.

To minimize these threats, researchers can employ various strategies. For example, they can use random assignment to ensure that participants are equally distributed across different conditions, which can help minimize selection bias. They can also use control groups to account for the effects of history and maturation, and counterbalancing to control for the effects of testing and instrumentation. Additionally, researchers can use statistical techniques such as analysis of covariance to control for potential confounding variables and reduce the impact of regression to the mean.

Overall, minimizing threats to internal validity requires careful planning, rigorous methodology, and thoughtful consideration of potential confounding variables. By addressing these threats proactively, researchers can increase the internal validity of their studies and enhance the reliability of their findings.

Correct Answer:

Answered by Examlex AI Copilot

Assigning subjects to research groups can be done in a variety of ways, but the most common methods include random assignment, matched pairs, and stratified random sampling. Each approach has its own advantages and disadvantages, and the best method will depend on the specific research question and the characteristics of the subjects being studied.

Random assignment is often considered the gold standard for assigning subjects to research groups because it helps to ensure that the groups are comparable at the outset of the study. This approach minimizes the potential for bias and allows researchers to make causal inferences about the effects of the independent variable on the dependent variable. However, random assignment may not always be feasible or ethical, especially in cases where certain subjects may be at higher risk for harm or where it is not possible to control for all relevant variables.

Matched pairs and stratified random sampling are alternative approaches that can help to address some of the limitations of random assignment. Matched pairs involves pairing subjects based on specific characteristics and then randomly assigning one member of each pair to the treatment group and the other to the control group. This approach can help to ensure that the groups are balanced with respect to important variables, but it may be more time-consuming and difficult to implement than simple random assignment. Stratified random sampling involves dividing the population into subgroups based on certain characteristics and then randomly selecting subjects from each subgroup to ensure that the groups are representative of the population as a whole. This approach can help to ensure that the groups are balanced with respect to important variables, but it may also be more complex and resource-intensive than simple random assignment.

In conclusion, the best approach for assigning subjects to research groups will depend on the specific research question and the characteristics of the subjects being studied. Random assignment is often considered the gold standard, but matched pairs and stratified random sampling can also be valuable alternatives in certain situations. Ultimately, researchers must carefully consider the pros and cons of each approach and choose the method that is most appropriate for their study.