Traditional statistics have enabled researchers to understand what things — such as education, occupation, and gender — predict how much a person will earn. Now, in the first study of its kind, researchers have used machine learning to estimate the importance of these factors, finding that a person's ability to delay immediate gratification ranks among the best predictors of wealth.
For the first time, researchers have used machine learning to rank key determinants of future wealth. Education and occupation were the best predictors, but surprisingly, a person's ability to delay immediate gratification was also one of the most important factors for higher income, beating age, race, ethnicity and height.
Many factors are related to how much money a person makes, such as age, occupation, education, gender, ethnicity, and even height. Behavioral variables are also involved, such as one related to the famous "marshmallow test." This study of Delay discounting, or the extent to which a person discountes the value of future rewards compared to immediate rewards, showed that children with greater self-control were more likely to earn higher wages later in life. But the study's lead author says more traditional ways of analyzing data have failed to identify which of these factors are more important than others.
This study collected a large amount of data – from more than 2,500 different participants – and divided it into a training set and a test set. The test set was set aside while the training set produced model results. The researchers then returned to the test set to test the accuracy of their findings.
Unsurprisingly, the models indicated that occupation and education were the best predictors of high income, followed by location (as determined by zip code) and gender – men earned more than women. Delay discounting was the next most important factor, as it was more predictive than age, race, ethnicity or height.
