How To Make Decisions Using Data

Are you thinking about what you would like your next role to be at your company, or if you should make a career switch, or buy a new house? If you are looking to the future and planning your next career move or big life decision, it can feel challenging to make a decision. Information paralysis creeps in when you have so much information that it feels difficult to make any decision. You may even find the language of statistics frustrating. P-values, normal distributions, and R-Squared tests may seem overwhelming if you haven’t brushed up on your high school statistics book. However, even for the most data illiterate, having a simple yet confident understanding of statistics can help you make the right decision. Understanding what data is, how to challenge data analysis, and how to make decisions with that analysis can help you to make your next career decision.

Xander Watkins is an Orr Fellow at Resultant where he is a data scientist. Watkins said, “data science is storytelling with data.” If you are presented with thousands of data points in a spreadsheet, it can be impossible to see any relationship between those data. So, data science is “taking that information and we’ll try and figure out what story we can tell using this information and how we can translate that into something that someone who’s not a data scientist can also understand.”

Darell Huff wrote How to Lie With Statistics in 1954. Huff’s book is still a beginner’s authority to understanding errors in statistical analysis. For example, Huff cautions the reader to be skeptical of correlation. You may have heard your high school statistics teacher say “causation is not correlation.” Watkins put it this way: “When ice cream sales go up, murder rate goes up at the same time. You could present this fact and say ‘OK, well, we should stop selling ice cream because it causes murders. . . But it’s because it’s during the summertime when it’s hotter out. . . that’s an example of how you can present data to be super misleading. And it looks like there’s a strong correlation.” In a situation like this, it may be that there is a third variable, the warm weather.

Curt Merlau is the director of Resultant’s education practice, where he helps educational institutions understand the data they have and what to make of it. “The real power of data is in the conversations that happen around it, or because of it. And so you have to make sure to summarize, that there’s the right kind of environment and mindset around data before you just go start collecting it or visualizing it,” Merlau said. When you are looking at data, it is important to understand the origin of the data and how to use it. It does not necessarily take a data scientist to interpret the strength of data.

When you are considering looking at your future career, one reputable source is the U.S. Bureau of Labor Statistics. On the BLS website you can look up almost any career path in the “Occupational Handbook.” This tool provides great data on future outlooks, like expected salary, growth rate, and work experience. While this is a great tool, it is helpful to think of its limitations. For one, this data does not take into account individual factors. Is your career choice a good fit for you? Also, it does not do a great job of accounting for regional factors. Are software engineers needed more in Silicon Valley or New York City? But, as Merlau said, this data is a great place to start having conversations about your future.

Whether you are a seasoned data expert or a data novice, it is important to understand the basics of data analysis so that when you are confronted with it, you can evaluate the argument for yourself. Start with understanding the basics of causation and correlation, and strengthen that understanding by having conversations with mentors about what the data means.