Answer

Sampling Bias/Error

In the quest to conduct studies and provide findings, researchers can sometimes make sampling biases and errors which ultimately affect the validity and reliability of their findings. Notably, according to Leung (2015), sampling bias; which is also known as the sample selection bias, refers to those errors which arise in research studies due to the researcher failing to select the research participants properly. When researchers collect a sample in such a way that all the members are not likely to be included, then a sampling bias can be found (Singh & Masuku, 2014). For instance, the use of a sampling method which does not allow for an equal opportunity for selection, then a sampling bias occurs. This paper discusses the presence of sampling bias of the new article by Epstein (2019) which reported on how Google shifts votes.

Epstein (2019) reports in his commentary that Google may have in one way or another skewed the results of the 2018 midterms in the US elections my millions of votes. The assertions of the author were based on a study that he did with about 130 anonymous “field agents” in the Orange County area of California. According to Epstein (2019), when compared to the other search engine such as Yahoo, Google had a significant liberal approach as its results in all the ten days leading up to the election day and in 10 positions of search results were all liberal. In this case, the sampling bias arose from the fact that Epstein (2019) only selected the anonymous field agents from orange county in California and thus the study could not be representative enough of the study population (voters). Additionally, the findings could not be generalized as Orange County experienced extreme cases of bias and thus the conclusions could not be applied to the entire US elections.

I think the issue of bias in the selection of the research participants can be resolved by using a simple random sampling. In specific, the use of simple random sampling would ensure that the samples are strictly chosen by chance (Leung, 2015). As such, every member of the study population will have an equal odd to be selected as a study participant. The sampling bias evident in the study by Epstein (2019) affects the credibility of the results and implies that they cannot be generalized. As such, the other way to fix the bias is ensuring that the sample is obtained from a larger population to ensure that the sample is representative of the study population. Taking sample from a broader population increased the generalizability of the findings.

In conclusion, a sampling bias; which involves an error in obtaining a sample, adversely affects the credibility and reliability of the research findings. The article by Epstein on how Google search may have shifted the 2018 US midterms is riddled with sampling bias. In specific, the author failed to deploy an appropriate sampling method which led to the sample not being representative enough of the study population. To fix this problem, I will use a simple random sampling to make sure that every individual in the study population has an equal chance of being selected as a study participant.