¶ … power important? The issue of power is important because a study that has statistical power demonstrates the ability to detect an effect that is due to the intervention being measured. Power indicates that the effect being observed in a study results from the manipulation of the independent variable.
What is the relationship between level of significance and power?
If the results of a study indicate that the difference between the results of the control and intervention groups are statistically significant, then the researcher would logically conclude that the effect observed is due to the manipulation of the independent variable. The fact that the researcher was able to detect an effect demonstrates that the study has statistical power.
How is power tied to sample size?
If the sample size of the study is not large enough, the effect under investigation may not be observable, even though the intervention may in fact be effective. It is of the utmost importance for researchers to include an appropriate number of participants in the sample in order to accurately determine whether the intervention under investigation is truly effective or not.
Why is effect size often difficult to accurately estimate? What are the best means for making this estimate? Effect size is difficult to estimate because it requires detailed statistical information (in particular means and standard deviations) specifically related to the research question under investigation. The means of the control group, means of the intervention group and either pooled or conservative standard deviations are then used in an equation to calculate an expected effect size.
If no published data exists to calculate effect size, how will you calculate on? If there is no published data that directly corresponds to the research question under investigation, the researcher could look to studies related and similar to the current study in order to obtain the data necessary to calculate an effect size.
What strategies might be employed to increase the expected effect size? It is always optimal to have a large effect size. An increased expected effect size means that a smaller sample size will be required in order to observe the effect under investigation.
The best method for increasing expected effect size is dosing. Dosing involves increasing the amount or frequency of the intervention.
How can dosing be used to increase the expected effect size? A researcher can use dosing to increase and expected effect size by increasing the duration or frequency of the intervention for participants in the treatment condition. Increasing the dose of the intervention makes the effect under investigation more apparent and pronounced, thus giving it more statistical power.
Discuss how the two research questions / hypotheses presented in Module 1 could be "dosed-up" for a larger expected effect size.
You’re 79% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.