The normal goal of a survey is usually to describe a population. Another goal that may need an extremely different study design is to make connections between groups (Thayer-hart et al., 2010). Usually surveys are used to get responses from a certain group of individuals and to learn about their desires and needs. The next step in designing a survey is to determine the target population. If the target population is small, then it is generally doable to survey the whole population. Yet, when the population is large or scattered and the survey is a part of a difficult research project, a “sampling” process is usually used to choose and survey a subset of the entire target population one is interested in (Thayer-hart et al., 2010). If a sampling process would strengthen the results of the survey, one will desire to use the services of a sampling statistician to direct oneself. One major benefit of surveying a sample rather than an entire population is that with a lower sample size, one can incorporate more follow through participants to promote responses. As a result, cooperation and the quality of data may be enhanced. To reach the target population, one must ensure that they have correct and up to date contact information. Getting this information may be expensive and may take up a lot of time, particularly when the target population is very …show more content…
Choices made during the planning process “about what data are needed and the format in which it will be collected will pay dividends when it comes time to analyze the data” (Thayer-hart et al., 2010). A rational process for organizing and analyzing the data of the survey should be implemented before obtaining the first responses, with those involved getting sufficient training. Scaled responses are normally changed to numerical quantities for easier analysis. A template or mask can assist coders with converting the data more correctly and faster. Careful consideration “needs to be given to coding responses to open-ended items” (Thayer-hart et al., 2010). This process normally deals with observing some preliminary data to identify potential categories, and then testing to assure how consistently the categories are selected by different coders. To analyze responses to open-ended questions, one can copy the comments onto single cards and put similar comments together. This will allow one to examine the most recurring ideas. For analysis of the data, visual demonstrations can also help with understanding the data. If multiple choice questions are involved, one can create a frequency histogram of the responses for each question to depict the the difference in responses. A bar graph could also be used to show the percent of respondents choosing specific responses. If the results of the survey are written for publication, an advance