Statistics for Business and Economics Discussion Responses – Description
Two classmates discussion responses
One: In my opinion, the most useful statistical test covered in the course is the one-sample t-test. The test identifies if the mean of the sample population can be compared to the standard (Bougie & Sekaran, 2020). By the way the statistic is designed, the result of the small sample can help make an inference about a total sample (McClave et al., 2021). However, the t-test, and therefore, t-statistic, is more variable than the corresponding z-test as there are more random variables in the t-test, and the corresponding curve is not a perfect bell shape like z-statistics (McClave et al., 2021). Nonetheless, not all data falls on a normal distribution, so the t-statistics fill that gap for imperfect data.
Conversely, one of the most challenging, albeit fun, concepts to grasp was proposing a research question that fits the test studied for a given week in the course. While the discussion posts were thought-provoking, there was still a shadow of a doubt if the research question fit the test. In a way, it was a type of inversion thinking, as typically, the research question comes first before deciding which tests to use. Regardless, as the weeks progressed and upon learning from other students, this type of thinking became easier. I have really enjoyed working on the final paper to put it all together.
References
Bougie, R., & Sekaran, U. (2020). Research methods for business: A skill-building approach (8th ed). John Wiley & Sons.
McClave, J., Benson, P. G., & Sincich, T. (2021). Statistics for Business and Economics (14th ed.). Pearson.
Two: I have been going back and forth on whether I should conduct a quantitative or qualitative study for my dissertation. It may be appropriate to conduct a phenomenological study involving experienced participants in that I may be covering a topic involving faculty attrition and mass shootings. However, these sorts of studies can be difficult to acquire participants; therefore, a quantitative study may be a more feasible route. Additionally, a deductive study does sound appealing. Nonetheless, the media and its influence will somehow be intertwined in the study. Based on the readings, I believe using linear regression to explain the correlation between the media (x) and school shootings (y) would be an effective technique. Media is a convoluted topic that may be covered in more than one variable; therefore, having that broad scope to determine how it can influence the outcome variable could be helpful.
Initially, it was difficult for me to grasp a z-statistic and a t-statistic; tests conducted to determine inferences relating to a sampled population (Gravetteret al., 2011). These concepts were challenging because I vaguely recall learning about them in the past and it took longer on my part to understand how and why these tests are used. However, the course was seemingly organized so that every week one part complemented the next; therefore, these concepts began to make sense once everything was put together. I understood the concept of normal distribution and sampling and how that can be used to make inferences about a greater population. This part of the course I retained relatively easily.
References
Gravetter, F. J., & Wallnau, L. B. (2011). Essentials of statistics for the behavioral science (8th ed.). Wadsworth.
The post Statistics for Business and Economics Discussion Responses first appeared on .
