The fact it's only "nearly half" is a testament to the amazing level of, er, "marketing" that has been done on this issue, of which this article is another example.<p>First, it's not a "pay gap", but an <i>earnings differential</i>. "Pay gap" implies two people being paid different amounts of money for exactly the same work. That has been illegal in the US since 1963, and there is very little credible evidence of it happening.<p>The earnings differential does not compare individual pay, it takes all the money earned by all working women and compares it to all the money earned by all working men, and calls any difference a "pay gap".<p>This earnings differential, as the article admits, is largely due to things that rightly <i>should</i> affect earnings, such as hours worked, career choices, experience etc. All else being equal, the person who works more should earn more. All else being equal, the person with more experience typically earns more than the person with less experience etc. All else being equal, there are some jobs where you can earn more than others. Men tend to flock to these jobs, at least in part because society values men by what they earn (and certainly women do!). So motivation is a huge factor.<p>I am somewhat dubious of the claim that 1/3 of the differential is unexplained, the figures I have seen are significantly lower than that. But even if the 1/3 were true, it only means it is unexplained, it does not in any way show that this small remainder is due to discrimination.<p>For example, men tend to take greater risks, which on average leads to greater rewards. In a German study, this factor alone led to an earnings difference between men and women inside the experiment of around 20%.<p>What is really telling is that even an obvious advocate can freely admit that at least 2/3 of the earnings differential is not a "pay gap" (i.e. "made up") yet unironically lambast those with a more accurate grasp of the facts than herself as being "incorrect".