Designing a Research Study

Description of Study and Design to Be Used in Future Research Project

Topic 1. Describe a vulnerable population in terms of research. Why the population is considered vulnerable? What other groups might also be considered vulnerable populations?

A vulnerable population in terms of research is a population that is often overlooked by the academic community or a population that does not fit in with the mainstream sample. Waisel (2013) states that: “vulnerable populations include patients who are racial or ethnic minorities, children, elderly, socioeconomically disadvantaged, underinsured or those with certain medical conditions,” and that these populations typically “have health conditions that are exacerbated by unnecessarily inadequate healthcare” (p. 186). Asian-Americans with type 2 diabetes are a vulnerable population in the U.S. They are a racial and ethnic minority on whom there is scant medical research in terms of how best to provide quality care for them (Walton-Moss, Samuel, Nguyen, Commodore-Mensah, Hayat & Szanton, 2014).

Other groups that might also be considered vulnerable populations would include other minority racial and ethnic groups as well as groups that minority in terms of gender identity, socio-economic status, or groups with disabilities, and so on. These could include homosexual populations, HIV-infected populations, drug abusers, and many others.

Topic 2. Assume you want to do a project that compares the survey results before an intervention to those after an intervention in the same sample (the same people will take both surveys). You plan to use a paired t-test to analyze your results. Using the \\\"Determining the Appropriate Sample Size,\\\" Topic Material, perform a sample size calculation to determine how large your sample should be. Justify your sample size calculation with citations. Discuss how your sample size may affect the validity of your study.

A sample calculation of how to determine the size of a sample population will depend on the confidence level Z score that the researcher wants to achieve. If the confidence level aimed at is 90%, the Z score would be 1.645. This score is then plugged into another equation:

Necessary Sample Size = (Z-score)2 * StdDev*(1-StdDev) / (margin of error)2

0.5 is normal standard of deviation and 0.05 is normal margin of error acceptance, so, the equation that is used would look like this:

Necessary Sample Size = ((1.96)2 x .5(.5)) / (.05)2

(3.8416 x .25) / .0025

385 respondents are required.

As Kadam and Bhalerao (2010) note, a sample is not an entire population but rather on a representational grouping of it—“a further subset of the target population which we would like to include in the study. Thus a ‘sample’ is a portion, piece, or segment that is representative of a whole” (p. 55). For the sample to be effective, it has to have significance, and the level of significance is the “p” value, and typically the range for “p” to have significance is anywhere below 0.05 or p < 0.05.