Answer

Sampling Techniques and Study Design

Sampling frame is often defined as a set of information that can be used in the identification of a sample population under study. Similarly, it is viewed as the procedure that a researcher follows as the actual basis for sampling (Wolf et. al., 2005).  The sampling frame may include details such as numerical identifiers for each individual in the sample in addition to other identifying information that pertains the characteristics of each individual to be studied allowing for in-depth analysis. The sampling frame is indeed the most essential element in a research’s sampling plan. As Benedetti, Piersimoni, & Postiglione (2015) observes, the type of frame used in the selection of units from a population have the greatest influence on both the quality of the findings as well as the cost of the research. According to the scholars, this because the frame is the only means through which the statistical units to be tallied in the study are identified.  Similarly, studies indicate that faulty sampling frames are likely to be the major cause non-sampling error or under-coverage of significant subgroups within the organization.  For instance, Yin (2013) indicates that since the primary aim of any survey is to effectively give accurate inferences about the target population, a well-designed sampling frame permits adequate representativeness of the population in the sample by bringing in all the qualitative characteristics of that population, therefore, ensuring that all the divisions within the population such as geographical coverage are considered. 

A poorly specified sampling frame that is either inaccurate or incomplete could hinder the research process in a variety of ways. First, it is likely to lead to a sampling error (Wolf et al., 2005). That is, a situation in which the sample selected does not adequately represent its population. This might affect not only the validity of the study findings, but also lead to a huge loss of money on the side of the researcher.   As Yin, (2013) observes, the core benefit for a surveyor is to obtain maximum information from a population and this often involves huge financial investments and significant decisions. However, a poorly estimated sampling will render such investment unprofitable due to inadequate coverage of the study population leading to inaccurate findings. Additional problem that might arise from a poorly developed sampling frame in increased amount of bias. Fortune, Reid & Miller (2013) define bias as the act of deviating in one direction from the true value of the concept being measured leading to conclusions or results that systematically differ from the truth. Besides, the poorly constructed sampling frame may only favor the selection of a particular units within a data set with given characteristic to be included in the sample. This limits the chances of other units appearing in the sample as well as the representativeness of the sample. Other problems that might arise from inaccurate sample frame are the coverage error. These refer to the omission of some units – under coverage – or wrongly including units into the sample also known as over coverage (Yin, 2013).  As the scholar indicates, the coverage errors often lead to biased results in addition to affecting the variance of the results.

Convenience sampling, also commonly referred to availability sampling or accidental sampling, refer to a non-probability sampling technique that involves selecting subjects based on their convenient accessibility or closeness to the researcher (Burns, Grove, & Gray 2011). Some of the benefits associated with convenience sampling include the ease and the simplicity of research since data collection can be easily facilitated within a short period time.  For instance, an individual may send a link to an online questionnaire to friends on his mobile phone contact list or through social networking website such as Facebook, Google+, Twitter polls, or LinkedIn. However, the studies that employ convenience sampling could be viewed as invalid. This is because, it is highly viable to selection bias. According to Warner (2013), the use of convenient sampling technique may result into selection bias especially if the researcher is personally involved into the study. This is because, the researcher may be consciously or unconsciously develop predetermined ideas regarding the research. However, the scholar indicates that the only way the bias can be reduced is to randomly assign the subjects into groups once they have been recruited.  Convenience sampling is also viewed to possess high levels of sampling error. This is because, its sampling frame is not known and its subjects are not chosen randomly, this reduces the ability of the sample to be a representative of the population under study therefore giving the sampling technique little credibility.

External validity refers to the confidence that a researcher may have in generalizing the study findings across a variety of situations or people who were not included in the study (Statistics Learning Center, 2012). The generalization of the results often requires that the sample included for the study truly represent the population to which the findings are to be generalized. Of the sampling techniques presented in the module, the simple random sampling seems to optimize external validity. This is because; the technique ensures that each member of the target population gets an equal chance of inclusion into the sample due to its stringent requirement (Statistics Learning Center 2012).  For instance, one has to develop an effective sampling frame that would guarantee that every member of the population to be investigated gets an equal chance of being included in the study. This is often achieved by calculating the response rate as well as the percentage of items or subjects contacted as a way of evaluating how well the simple random sample represents the population so as to ensure the most representative sample is obtained.