Making Decisions with Data: Understanding Hypothesis Testing & Statistical Significance - Article Review Example

“Making Decisions with Data: Understanding Hypothesis Testing and Statistical Significance” is a late 2019 Project Management Journal Article by Robert A. Cooper. Cooper (2019) wrote about the impeccability of making the right decisions with data. The author discusses the importance of statistical methods in the practice of science. Most importantly, the article has talked about the concepts that researchers associate with hypothesis testing and the application of independent samples using the ‘t-test’ methodology to determine variability and dependence. The application of this knowledge is an important connection for the suitability of this article to the project management.

Is your time best spent reading someone else’s essay? Get a 100% original essay FROM A CERTIFIED WRITER!

Save 25%

Cooper (2019) argued that students should be given research opportunities to analyze data and make inferences based on the information they get. By taking the “the lady testing tea” research, Cooper (2019) stated that there are some probabilities that are relatively easy to predict. However, project management entails the understanding of both simple and complex project structures with many constraints and variables. Project managers should be able to work with complex systems and conditions and work with the probability of events based on whatever information is present at the time.

Statistics provide for data organization for interpretation and dealing with variation (Cooper, 2019). Without statistics, it will be difficult for project managers to conduct scientific management of their work schedules and project deliverables. Errors are bound to occur, especially prediction and sampling error. Due to inconsistencies in data collection or human error in data recording and validation, errors might occur and affect the outcome of the analysis. Thus researchers have the provision of the null hypothesis to guide the course of the research activities.

Further, the article gives a clear indication of the sources of variation in hypothesis data and data analysis. There are real and induced variations, and both affect the outcome of the research process. Cooper (2019) stated that the outcome of events partly depends on the biases of the participants. Hypothesis testing works to eliminate doubt and bias that researchers might have concerning a particular outcome of the event. In the article, it has been recounted the story of the tea testing lady, who could tell whether tea or milk was poured in the cup first.

The Bristol Lady did identify all the cups correctly, but by luck or ingenuity or sheer probability, people will never know. However, the most significant element from this activity is the probability of events and sample size. The ‘tea testing’ test used eight samples to determine the outcome of the research. Ronald Fischer, who was present at the tea party, and a great scientist of the time, evaluated the probability of getting her answers correct to 1 guess in 70 chances, or 0.014 (Cooper, 2019). In a randomized trial of various repetitions, the probability that the lady would get all her answers correct came to 51.4%. This percent is more than half the possibilities, meaning that the lady could get more than 4 cups correct. The null hypothesis already inferred in this situation is proved to be true. The Boston tea lady correctly identified which cup had tea poured the first wand, which had milk in the cup first.

The author defined statistically significant outcomes as the unlikely outcomes to occur by chance (Cooper, 2019). A statistically significant outcome is analyzed by a chi-square test or test of significance using independent samples and t-test. The article stated that one should consider the outcome of an experiment before experimenting. Also, one should project all possible outcomes and be able to interpret them. The former part concerns hypothesis testing and the creation of both the null hypothesis and alternative hypothesis. At the same time, the latter is the test of significance to come up with the scientific probability of the event occurring. Possible interpretations from the data should as well be already identified.

Cooper (2019) has also utilized independent samples ‘t-test’ to contrast hypothesis testing to the t-test in a controlled experiment. It is reaffirmed in the article that the null hypothesis is what the researcher expects to be the outcome of the experiment. The alternative hypothesis disapproves of the null hypothesis. These hypotheses work with the probability of events. In turn, probability works with errors and confidence levels. From sampling error to measurement error, probability outcomes require some element of confidence and t-test score to be valid and dependable. The author stated that a confidence level of alpha 0.05 is acceptable in most situations (Cooper, 2019). The data sets from the experiment are used as evidence to support the inferences made from research results.

Conclusion

This article has a significant realization to the connection between the null hypothesis, alternative hypothesis, and the significance level of data items. In entirety, Cooper (2019) provided important insight into the study of probability and decision making. Project managers have to constantly make decisions that affect the timely completion of the project. Being faced with such daunting choices necessitates the application of proven scientific methods to effectively and efficiently achieve the project goals. Project managers should, therefore, understand how probability works, how to work with constraints and competing priorities as well as how to gauge the test of significance of any given variable. This article is a worthy read to all persons who desire to have in-depth knowledge of the working of hypothesis testing and the significance of data analysis to decision making.

Reference

Cooper, R. A. (2019). Making decisions with data: understanding hypothesis testing & statistical significance. The American Biology Teacher, 81(8), 535-542. https://doi.org/10.1525/abt.2019.81.8.535