2022 Introduction to Statistics in Research Mitchell 2nd ed

I N T R O T O R E S E A R C H : D A T A V I S U A L I Z A T I O N & C O M M O N S T A T T E S T S

DMAIC (Six Sigma) approach to solving a problem includes the following steps: define, measure, analyze, improve, and control. There are a variety of problem-solving methodologies in the corporate setting that are used to drive quality. 4) Prescriptive Analytics - In business, the question is asked, “What should be done?” as part of the optimal recommendation for a decision. In business, the researcher is looking for the best possible action based on the descriptive statistics and the variety of choices. In the use of Power Business Intelligence (Power BI), prescriptive analysis is used in the oil and gas management because prices fluctuate based on political, environmental policy, and demand. Healthcare practitioners use prescriptive analysis to improve clinical care. Prescriptive analytics recommends actions you can take to affect those outcomes. Techniques include simulation, graph analysis, algorithms, and machine learning. 5) Diagnostic analytics helps you understand why something happened in the past. Questions here might include, “Why did it happen?” Techniques include drill-down, data discovery, data mining, and correlations. As you might expect, the healthcare industry uses this analytical model every day. A sudden spike in the emergency room count could be pointing to an infectious agent that has broken out in your area (think of COVID-19!). 6) Predictive Analytics – predicts what is most likely to happen (future), so you can expect a question, “What will happen?” Some of the techniques used include data mining, modeling, and artificial intelligence. The most likely use is by insurance and financial companies, but a growing industry that uses predictive analytics includes marketing and healthcare (COVID-19 algorithms. 7) Exploratory Data Analytics – gaining key insights by focusing on analyzing patterns to recognize potential relationships. This is generally one of the first steps in gaining an understanding of your data. In Power BI, once you upload your data, Power BI will review it for insights and create graphs automatically. Essentially, Power BI looks for subsets of your data. The types of insights include: 1) outliers; 2) change point in a time series; 3) correlation; 4) low variance; 5) majority (attributing to a single factor); 6) overall trends in time series; 7) seasonality in time series; 8) The greatest value of a picture is when it forces us to notice what we never expected to see. ----John Tukey

steady share. Part of becoming a good researcher is to be able to communicate your findings; so always use every tool you can to really understand your data.

8) Causal Analytics - this type of analytics focuses on the “why” something occurred. It is important to understand the difference between correlation and causation. Causation indicates that one event is the result of the occurrence of another event. (cause = effect). Correlation tests for a relationship between two variables. But seeing the two moving together does not necessarily mean one variable causes the other. “ Distinguishing between what does or does not provide causal evidence is a key piece of data literacy” ( Statistical Knowledge Portal: a Free Online Introduction to statistics, 2020, para 2).

In this Analytic Journey illustration pay attention to the value Vs the difficulty axis. Never go straight to the hypothesis test; always take the time to investigate the data using descriptive statistics, charts and

111