Public health sectors like the Centre for Disease Control and Prevention (CDC) rely on solid concepts based on current literature to make sound decisions. In modern healthcare, leaders rely on various secondary data to plan and execute interventions to increase efficiency, effectiveness, and overall performance. Before using data or information from any secondary source, it is paramount to understand basic concepts like weighting. Weighting is an essential component in survey sampling used to obtain estimates of population parameters of interest. Researchers use weighting to match the population profile of more than one variable to get a sample representative through the use of SPSS software ( Lavallée & Beaumont, 2015) . The weighting process involves three steps. The first is obtaining the design weight that accounts for sample selection. The second is adjusting weights to compensate for nonresponse and adjusting the weights to allow the estimate to coincide with the total population. Weighting provides researchers with several benefits and is the application in various areas of research.
Importance of weighting in secondary data
Weighting in a sample survey is important since they help in analyzing statistical complex survey data. In an informative sampling design, weighted estimators produce better and accurate results. Weights are applied after collecting data, allowing the researcher to correct errors during data collection to represent the population being studied accurately. Another importance of weighting adjustment is compensating for biased estimators produced by survey nonresponse ( Pike, 2008) . In such a situation, the major problem is the bias introduced when respondents have similar characteristics that differ from nonrespondents. Some individuals from the selected sample may fail to respond to a survey, especially those conducted through email, which might lead to bias (Fuller, 1974).
Delegate your assignment to our experts and they will do the rest.
The weighting method allows the researchers to test survey samples containing responses from various sample population. Weighting makes it possible to get accurate data concerning behavioral changes in the selected population. CDC is known for using a post-stratification statistical method to weight the Behavioral Risk Factor Surveillance System (BRFSS) of various demographics ( Pierannunzi et al., 2012 ). Another benefit of weighting is estimating the probability sample size since it enables the researcher to compensate for differential sampling ratios arising from under or over coverage. Using weighting makes it possible to balance the number by down weighting one variable and up weighting the other.
Application of Weighting.
During the research, the weighting will reduce bias resulting from nonrespondents by taking advantage of auxiliary information available for respondents and nonrespondents ( Pike, 2008; Fuller,1974) . Biased estimators occur when survey variables of respondents and nonrespondents differ from each other, and there is a difference in response rate across subgroups (Pierannunzi et al., 2012). I will apply weight adjustment to compensate for bias arising from various subgroups' response rates that differ on survey variables.
Another application of weighting during the study is to make adjustments to account for the population's over-coverage or under-coverage. During the research, I will use weighting to get the actual number of males enrolled in Medicaid out of the known size "N*" of the targeted population, which has different size "N." I will adjust the weight to ensure it sums up to N and not N*. Also, I will use weighting to correct any imbalance in sample profiles after data collection ( Lavallée & Beaumont, 2015) . For instance, in a survey conducted of a sample of 400 people, 300 are men and 100 are women. Hence, there is a need to correct the imbalance. By weighting, I will down weight sample size of men from 75% to 50% and the up weight proportion of women from 25% to 50%. Through this, a balance will be achieved, leading to accurate results.
References
Fuller, C. H. (1974). Weighting to adjust for survey nonresponse. Public Opinion Quarterly , 38(2), 239–246.
Lavallée, P., & Beaumont, J. F. (2015). Why We Should Put Some Weight on Weights. Survey Methods: Insights from the Field (SMIF) .
Pierannunzi, C., Town, M., Garvin, W., Shaw, F. E., & Balluz, L. (2012). Methodologic changes in the behavioral risk factor surveillance system in 2011 and potential effects on prevalence estimates. MMWR: Morbidity & Mortality Weekly Report , 61(22), 410–413.
Pike, G. R. (2008). Using weighting adjustments to compensate for survey nonresponse. Research in Higher Education, 49(2), 153–171.
