Here are my (non-academic) articles that discuss statistics.
1. How to use statistics to tackle a real-world problem? Part 1. Problem Formulation
- Date: 2020/10/28
This article will discuss the basic concepts of statistical consulting — how to solve a real-world problem. Most of the statistical training focuses on theoretical and analytic skills; however, the duty of a statistician is much more than analyzing the data. Instead, statisticians solve the problem through scientific strategies, most likely through/using inductive reasoning. Box (1976) says that the statistician’s job requires “The wit to comprehend complicated scientific problems, the patience to listen, the penetration to ask the right questions, and the wisdom to see what is, and what is not, important.” In other words, statisticians need to be equipped with good communication skills to obtain sufficient background knowledge of the problem and then deliver solutions to the clients efficiently and precisely.
As a statistician, how do you solve a research problem? Statisticians can implement a problem-solving procedure. After fully understanding the problem, we formulate the problem statistically. If necessary, we should design a study to collect data to obtain more scientific evidence. Based on fundamental statistical theories, we assess the structure and quality of the dataset by examining the initial data pattern. Then, we propose appropriate statistical methods and models to perform data analysis and future prediction. Finally, we interpret the empirical data. The report communicates the results using data visualization skills, makes inferences for generalization, draws conclusions cautiously, and proposes pragmatic suggestions. When applicable, we compare the findings with previous research results. This extra step might help us find potential issues or alternative methodology about the dataset.
Problem formulation is the most crucial step in an analytic project. Per the golden rule of applied mathematics, we know that an approximate answer to the right question is more valuable than a precise answer to the wrong question. Giving the right answer to the wrong question is also called a type III error (Kimball, 1957). Albert Einstein mentioned that “The formulation of a problem is often more essential than its solution, which is merely a matter of mathematical or experimental skill.” Therefore, make sure that you have already fully understood the problem before considering solutions. Some researchers underestimate the time and effort required to understand and formulate the correct problem. Usually, it takes more time to find the right question than to find the right answer…