Your third-grade science teacher introduced you to what may be the best strategy for picking up new skills and knowledge than any you'll ever find. You may not have realized it at the time or even remember it now, but it's been in use for hundreds of years: the scientific method.
As a former chemist, I realize I'm biased, but the reason scientists in every field stick with this foundational approach is because it allows them to make measurable, replicable observations that link cause and effect. Of course, humans have their own idiosyncrasies, so while much of the research on human behavior holds true among many people, it'll always be an approximation. To understand who you are and how you function—and then develop your own skills—you need to research and experiment on yourself. Here's how.
Start by assessing the status quo: How do you currently see yourself and the world? What do you notice about the way you react to and interact with others? What aspects or patterns in your life do you want to change? For instance:
- "When I work more hours, my career prospects improve, but my family life suffers."
- "My pants don't fit me anymore."
- "My mother-in-law hates me."
Your initial observations are the data you start with. Looking at those, identify what you'd like to learn more about, improve, or resolve. What do you want explained, changed, or fixed?
The beauty of science is that you choose the problem or question you want answered:
- "How can I succeed professionally while keeping my family happy?"
- "What can I do to lose 15 pounds?"
- "How can I improve my relationship with my mother-in-law?"
Now that you've distilled your observations into a challenge to tackle or question to answer, you can formulate a hypothesis. That's your best, educated guess about what you think may happen if you make the most reasonable changes you can envision. Like any good scientist, draw on your past experiences—the existing body of data—to predict what you think that might be. You can also look at what others have tried to solve a similar problem to the problem you've pinpointed.
Whatever your hypothesis is, it's the prediction that defines what your experiment will need to be designed to test. Here are a few example hypotheses:
- If I leave work early to spend time with family and bring some work home to do at night, I can work toward both career and family goals.
- If I remove sugary drinks and fried foods from my diet, I can lose at least 15 pounds in four months.
- If I make a point of spending more quality time with my mother-in-law, I can get to know her better and learn how to improve our relationship.
Once you decide on your hypothesis, it’s time to design and run your experiment. How will you show whether your hypothesis is right or wrong? What will you do to isolate other factors that may affect the outcome?
For instance, if you're set on losing 15 pounds by switching up your diet, you'll need to know that it's those factors that led to any weight loss and not something else, like any changes in your exercise regimen. In other words, you'll need to do a controlled experiment. That may mean logging both your exercise and diet habits for a month before starting on your diet change.
Armed with that knowledge, you can then control the exercise variable. By exercising the same amount before and during the experiment, that will no longer be a factor in the outcome you're looking to measure.
Good experimental design means having to make a range of other decisions you might not otherwise consider when you're trying to learn a new skill or change a habit. With the scientific method, though, you need to be rigorous. For instance:
- How long your experiment will be—will you stop at four months or when you lose 15 pounds?
- How and how often will you measure your weight?
- What data will you log—food and physical activity, or also mood, work hours, sleep, etc.?
Scientists keep meticulous journals, and so should you. The success of any experiment rests on how replicable it is. If you can lose 15 pounds every time using this approach, then you might have learned something really fundamental about yourself.
But you aren't done yet.
To know for sure, you need to take the time to understand what your results actually say.
Compare your experimental data with your hypothesis: Was your hypothesis right or wrong? In the weight loss example, if you lost 15 pounds within four months, then your hypothesis was right. If not, then you were off the mark. But even if you were wrong, you've still learned something valuable that can form the basis for your next effort.
If you did succeed with your weight-loss goal (just to extend that example), you can develop a few possible theories for why:
- The weight was lost because eating less fried food decreased total calories consumed.
- More sugar leads to raised insulin levels, which stores more fat in your fat cells. By decreasing sugar intake, you decreased the amount of fat stored in your body.
- In this experiment, it took 15 days to lose the first 5 pounds, and then another 30 days to lose the next 10. As your body fat percentage decreased, your weight loss slowed down accordingly, even as your diet continued.
When interpreting your data, beware of the natural tendency to find meaning where there is none. Remember: correlation is not causation.
Scientists often publish their findings in journals so other scientists can make use of their conclusions. So tell your friends how your experiment went. Let them know what you learned. In science, every theory leads to even more questions—which means experiments beget more experiments. One reason for using the scientific method to improve your skills, accomplish goals, or answer questions is because it's designed to build on existing knowledge and keep you wanting more.
After losing those 15 pounds, your next question might be, "How much do I need to exercise to keep my current weight steady?" You may not know the answer to that yet, but now you know how to find out.
Robert Chen is an executive coach who uses his science, business, and cross-cultural background to help technical leaders communicate with more impact and build better working relationships. He works at Exec|Comm, a global communication skills consultancy in New York City.