A research proposal consists of many integral parts that work together to outline a viable project and produce a path to any answer (Babbie 2011). With each element of a proposal scrutinized, the sample design is a commonly discussed component. A sample is defined as a process that asserts that a portion of an appropriate size, in a relevant setting yields results that can be applied to a larger population as a whole (Lohr 2009). Sampling is needed in order to contain cost components of a project that may otherwise make it prohibitive. This definition defines the reach of the project and the scope of the evidence, making the sample subjects crucial to a properly balanced research effort.
Sampling can be broken down into probability processes that include systematic random sampling, stratified random sampling and simple random sampling (Lohr 2009). Each of these processes holds unique strengths and weaknesses that create the need to determine the projects requirements in order to use the best method. Conversely, non-probability sampling is when an effort is implemented that does not ensure that all elements have equal opportunities of being included (Lohr 2009). With methods of non-probability research including, accidental sampling, purposive sampling and quota sampling there is a direct correlation to the categories assumed to exist within any organisation or society.
With many sampling procedures to choose from ones such as the convenience sampling procedure commonly provide a method for the research to create the easiest means to gather relevant evidence (Lohr 2009). Throughout the research process, there must be an effort to ensure a balanced sampling effort in order to ensure that the end results do not tend to bias in favour of the researchers predetermined assumptions (Lohr 2009).
A sampling design is the process that the research proposes to create the evidence regarding the target population (Qunilan 2011). Elements including sample size must be determined in order to provide the proper ratio to the general population in question (Quinlan 2011). This leads to the creation of a confidence interval, or CI, that measures the reliability of any sampling estimate (Lohr 2009). This element varies from sample to sample and is determined by the unique needs of the project at hand. With higher per cent age values denoting better confidence in the results the CI, the numbers have become an important sampling index tool (Lohr 2009). Further, the population in question must be assessed in terms of geographical, regional, age and gender needs as related to the project. In order to properly these assessments sampling frames are created that define the research’s population of interest, specifying the aspects of the study that are in question (Quinlan 2011). The process of creating sampling frame serves to define the associated elements from which the study selects a sample of the target population. It is due to the common condition that there is no access to an entire population of interest creating the need to rely upon a sampling frame that represents each of the elements of the population (Quinlan 2011). These sampling frames are divided into two types that include: list and non-list. List frames examples may include registered voters, residents listed in the telephone directory, or a roster of students (Quinlan 2011). Nonlist sampling can be created at events or incorporations of large populations.
Research requires accurate and timely evidence that is created through the sampling process, creating the need to specify the nature of creation and collection. In the end, any study is built on the foundation of its evidence making the sampling a critical element of every effort.
References
Babbie, E. (2011). The practice of social research (1st ed.). Belmont, Calif.: Wadsworth Pub. Co.
Lohr, S. (2009). Sampling (1st ed.). Pacific Grove, CA: Duxbury Press.
Quinlan, C. (2011). Business research methods (1st ed.). Andover, Hampshire, UK: South-Western Cengage Learning.