Lindsay Dodson 8:10pmMar 25 at 8:10pmManage Discussion Entry In the articleReaching the hard-to-reach: A probability sampling method for assessing the prevalence of driving under the influence after...

Lindsay Dodson

In the articleReaching the hard-to-reach: A probability sampling method for assessing the prevalence of driving under the influence after drinking in alcohol outlets, 2012. The sampling procedure that was used is probability. They collected the data from different outlets that sell alcohol and looked at the data of alcohol-related crashes as a result of that outlet. Then sample the probability of the hours they were driving vs. not driving. In this case, the probability sampling method did adequately represent the population because they had enough sample size to provide an accurate depiction of what was happening. They could have taken all the data, not just a random sample size. I would suggest keeping the data set in a specific location to be able to provide the best possible statistical information of a specific area.

On the other hand the non-probability sampling method article Data sampling strategies for disaster and emergency health research.The information for this study was not random like that of a probability sampling study, so therefore this was considered a non-probability study sample data set. They should include both types of sampling methods because with non-probability sampling it can create data bias when only taking samples from one specific location. In order to err on the side of caution or trying to set it up in a way that the best data is presented.

In my opinion, I believe probability sampling is the better of the two options. It gives you a better data set, without as many restrictions and bais. Cleaner data. Non-probability would be better suited for a very specific desired outcome for a small sample size. The problem with that though is it is not as scholarly accepted as the other type, it will not help a larger audience only the small intimate one it would be intended for will benefit.

References:

De Boni, R., Pedro Luis do, N. S., Bastos, F. I., Pechansky, F., & Mauricio Teixeira Leite, d. V. (2012). Reaching the hard-to-reach: A probability sampling method for assessing prevalence of driving under the influence after drinking in alcohol outlets.PLoS One, 7(4) doi:http://dx.doi.org.proxy-library.ashford.edu/10.1371/journal.pone.0034104(Links to an external site.)

Stratton, S. J. (2019). Data sampling strategies for disaster and emergency health research.Prehospital and Disaster Medicine, 34(3), 227-229. doi:http://dx.doi.org.proxy-library.ashford.edu/10.1017/S1049023X19004412

Daniel Hoogestraat

Sampling Techniques

When scientists conduct experiments, they select sample groups to capture a glimpse of what they believe to be an accurate representation of a population. There are different approaches as to how scientists select their sample groups for an experiment. One approach is the use of probability sampling, and the other is the use of non-probability selection. Specifically, probability and non-probability sampling are used to examine, detect, and quantify cause-and-effect relationships between test subjects and independent variables (Cozby & Bates, 2017). In probability sampling, every member of the population has the chance of being selected. It involves principle, randomization, or chance. In non-probability selection, however, members of the population have a zero or "unknown" chance of being selected (Cozby & Bates, 2017, p. 154).

Through examples provided by researchers Kamenidou, Stavrianea, Mamalis, and Mylona (2020) and Torten, Reaiche, & Caraballo (2016), the discussion below will demonstrate my understanding of probability and non-probability selecting methods with respects to quantitative research.

Probability Sampling Method

Researchers Ron Torten, Carmen Reaiche, and Ervin Caraballo are renowned professors in the fields of business and economics. The referenced article seeks to answer how personnel factors impact the performance of teleworkers. Specifically, the article's researchers explore how experience and training relate to teleworkers' satisfaction, performance, and productivity. The study was conducted using 400 (21 years of age or older) participants from a United States teleworking sample using quantitative instruments (i.e., canonical correlation) to gather employee job performance perception, job satisfaction, and productivity. Established by Qualtrics.com, the researchers used a simple random selecting method to nominate their subjects in terms of age, occupation, and other demographics. Specifically, the simple random selecting method was executed in the following manner:

1. Population: The population of U.S Teleworkers


2. Sampling Frame: Members of Qualtrics samples


3. Sampling Method: Simple random selection of Qualtrics samples (400 subjects)

In this specific research study, the researchers could have used the cluster sampling method instead of the simple random method to compare results. For instance, instead of randomly sampling from a list of teleworkers, the researchers could have identified "clusters" of individuals in specific occupations or industries to observe what specific telework occupations produce satisfied or dissatisfied employees.

Non-Probability Sampling Method

Researchers Kamenidou, Stavrianea, Mamalis, and Mylona (2020) provided a self-assessment to a generation Z population to investigate their knowledge and awareness level of COVID-19 symptoms. Using the social media platform Facebook, the researchers employed non-probability sampling (snowball sampling) to select their sample group and distribute questionnaires. Specifically, the snowball sampling method was executed in the following manner:

1. A small group of participants were recruited initially.


2. The sample size grew by asking the initial recruitments to recruit others which increased the sample group exponentially.


3. The secondary recruits were then asked to recruit others, providing a compounded sample group of 824 subjects.

In this specific research study, the researchers could have used the purposive sampling method instead of the snowballing method to compare results. The purposive sampling method is based on the judgment of the researcher. In other words, the researchers could have "purposely" selected their subjects based on typical criteria such as age, gender, education level, and occupation (Cozby & Bates, 2017). Being that part of the study investigated the differences between genders, a purposive sampling method could have made the selection process more rigorous.

Comparison

Comparing both the probability and non-probability research studies provided in this discussion, I feel that the probability method of simple random selection is the most rigorous because it ensures that everyone in the population has an equal opportunity to be selected. For example, "If the population has 1,000 members, each has one chance out of a thousand of being selected" (Cozby & Bates, 2017, p. 154). Whereas, in my opinion, non-random sampling does not provide an adequate representation of a population.

Concluding Remarks

Implementing selection methods within a research study is persuasive and beneficial in supporting claims because it helps decipher information more objectively and scientifically. In short, selecting, observing, and measuring data to determine variances, percentages, and behaviors validates research because it can be easily replicated or generalized for future applications.

References

Cozby, P. & Bates, S. (2017).Methods in behavioral research (13th ed). New York, NY: McGraw-Hill.

Kamenidou, I. “Eirini,” Stavrianea, A., Mamalis, S., & Mylona, I. (2020). Knowledge assessment of COVID-19 symptoms: Gender differences and communication routes for the generation Z cohort.International Journal of Environmental Research and Public Health,17(19), 1z. https://doi-org.proxy-library.ashford.edu/10.3390/ijerph17196964

Torten, R., Reaiche, C., & Caraballo, E. L. (2016). Teleworking in the new milleneum.Journal of Developing Areas,50(5), 317.