This paper is aimed at defining the appropriate sampling strategy and sample size for the study concerning relationship between HIV treatment compliance and social support among African American women with HIV. The paper consists of an introduction, the estimation of the population group, the analysis of the most appropriate sampling strategy, and the reference list. An overview of relevant resources concerning the HIV issue and the sampling strategies is provided. It is recommended that the population will consist of the African American women who were diagnosed with HIV in 2011. Since the total amount of the women is too big for conducting an effective study, it is recommended that the systematic random sampling strategy should be applied as it relies on random choice and prevents bias in sampling through a simple formula. The calculations of the desired population are conducted through an online calculator.
specifically for you
for only $16.05 $11/page
Stratified studies of the distribution of HIV and AIDS among various social categories state that the amount of African American women living with HIV has significantly decreased since 2008. Nevertheless, the up-to-date data suggests that the newly infected African American females are twenty times as numerous as white females. More than 80% of the infections are the result of contacts with infected males; besides, African American women also tend to have sexual intercourse with the representatives of their own ethnic group, which makes puts them under a greater risk of being infected and further spreading the disease (HIV Among African Americans, 2015). Other cases encompass drug usage as a result of the sex trade; it is suggested that this is an under-researched tendency in African American females which form a victimizable stratum (Dunne et al., 2014). Another vulnerable group is formed by African American women living in a poor neighborhood and having a low income level (Sikkema et al., 2014). A cohort study of African American women who have had HIV for a decade or more shows that, upon learning of their positive HIV status, they undergo all stages of grief. The study emphasizes the possibility of their recovery if the women are supported socially (Smith, 2015). Nevertheless, the actual relationship between HIV treatment adherence and social support is unknown, nor is the most optimal strategy of sampling. The purpose of the present paper is to find out the best sampling strategy and estimate the sample size for the study of the relationships between HIV medical compliance and social support for African American females.
As it was stated, the relationship between HIV medical adherence and social support of African American women remains understudied, to-date. To estimate such relationship, it is needed to consider HIV-infected African American women as a population group. The total number of newly-infected African American females in 2011 was 6100 which is about a third of all new infections among African American people. It is needed to estimate their HIV status and whether they have received any social support. Social support can be classified into several subgroups not necessarily related to public aid: social support networks, relatives, partners, friends, and acquaintances. It can be predicted that there will be an amount of women who have not received any support. These women are to be categorized separately. After questioning the group, it will be possible to draw a parallel between the support they have and their medical compliance. The main question is whether there is a correspondence between how – and whether – a woman is supported and her attitude towards medical adherence. Our hypothesis is that there is a difference between supported and unsupported women’s adherence. As a null hypothesis, we will take the absence of that difference.
Sampling methods and sample size
To conduct a high-quality research on the subject, it is needed to be able to operate with the basic notions and ideas of sampling. It is stated that inappropriate techniques used in research are likely to lead to inconsistent and one-sided samples (Devane et al., 2004). When the research is related to health care practices, too little number of the members of the sample group lacks credibility while too big a number can prove senseless when the answer is obvious, or there was a mistake in the hypotheses (Noordzij et al., 2011). Although there are many formulae, it is necessary to take some universals into account: the α-error, the β-error, and the standard deviation or SD. The first one consists in taking into account a nonexistent difference between groups under study. The second one subsumes not seeing the existing difference. The SD is the variability of the sample data (Gogtay, 2010). These are the variables that depend on the investigator and that we can include into the formula of an electronic sample size calculator. With the SD of 1, we set the α-error confidence at 5% and β-error at 20%. After the calculation, the desired sample size amounts to 471 (Researcher’s Toolkit, 2015).
As it was mentioned above, considering the probability of the mistakes, sometimes grounding the research upon too large a population number is a waste of time and funds. This is the reason we should calculate an amount of accessible persons sharing equal characteristics, which is just a smaller group of people (Weathington et al., 2010). The key word here is “accessible”. The most optimal choice of the method here is the systematic random sample since it ensures that every member of the population is chosen by an inclusion, which provides a non-biased choice (Frankfort-Nachmias & Leon-Guerrero, 2014). Using a simple formula, it is possible to define a sample interval of systematic random sampling that does not require as significant resources as the initial number. The ratio equals the size of the whole population divided by the sample size. It is stated that, today, African American females amounted up to 24% of all newly infected people of this origin. The actual number is 6100, and this number we are going to use in the equation since we cannot afford to question all of them:
Where K is the sample interval. Thus, it is best to choose every thirteenth woman in the population.
Thus, for the paper concerning the relationship between HIV treatment compliance and social support among African American women with HIV, it is recommended that the population will consist of the women in question. Since the total amount of newly-infected females is too big, it is necessary to apply the systematic random sampling strategy that simultaneously relies on random choice and prevents bias in sampling. Of the total population of 6100 HIV-infected African American women it is recommended to leave 471, which is more accomplishable, and do so with the usage of the abovementioned strategy.
100% original paper
on any topic
done in as little as
Devane, D., Begley, C. M., & Clarke, M. (2004). How many do I need? Basic principles of sample size estimation. Journal of Advanced Nursing, 47(3): 297-302.
Dunne, E. M., Dyer, T. P., Khan, M. R., Cavanaugh, C. E., Melnikov, A., & Latimer, W. W. (2014). HIV Prevalence and Risk Behaviors Among African American Women Who Trade Sex for Drugs Versus Economic Resources. AIDS and Behavior, 18(7), 1288-1292.
Frankfort-Nachmias, C., & Leon-Guerrero, A. (2014). Social Statistics for a Diverse Society. Thousand Oaks, California: SAGE Publications.
Gogtay, N. J. (2010). Principles of sample size calculation. Indian Journal of Ophthalmology, 58(6): 517–518.
HIV Among African Americans. (2015). Web.
Noordzij, M., Dekker, F. W., Zoccali, C., & Jager, K. J. (2011). Sample size calculations. Nephron Clinical Practice, 118(4): 319-323.
Researcher’s Toolkit. (2015). Web.
Sikkema, K. J., Heckman, T. G., & Kelly, J. A. (2014). HIV Risk Behaviors Among Inner-City African American Women. Black Women’s Health: A Special Double Issue of Women’s Health: Research on Gender, Behavior, and Policy, 3(4), 349-366.
Smith, M. K. (2015). Struggles and resilience of African American women living with HIV or AIDS: A qualitative study. Journal of Social Work, 15(4), 409-424.
Weathington, B. L., Cunningham, C. J. L., & Pittenger, D. J. (2010). Research Methods for the Behavioral and Social Sciences. New York, New York: John Wiley & Sons.