Friday , October 20 2017
Home / Brad Delong, Berkeley / Should-Read: Daniel Davies: From a logical point of view…

Should-Read: Daniel Davies: From a logical point of view…

Summary:
Should-Read: Daniel Davies: From a logical point of view…: “I have now read that ‘google manifesto’… out of a desire to forestall people saying ‘but have you ACTUALLY READ IT?’… http://crookedtimber.org/2017/08/11/from-a-logical-point-of-view/ … [and not] out of any expectation that it would contain new or unfamiliar information…. Indeed it was your fairly standard evo-psych “just asking questions”… mulberry bush…. I have changed in my old age; I used to be depressed at the generally very poor level of statistical education, now I’m depressed at the extent to which people with an excellent education in statistics still don’t really understand anything about the subject. I’m beginning to think that mathematical training in many cases is actually damaging; simple and robust metrics,

Topics:
Bradford DeLong considers the following as important:

This could be interesting, too:

Bradford DeLong writes Must- and Should-Reads

Equitablog – Equitable Growth writes Should-Read: Issi Romem: Paying For Dirt: Where Have Home Values Detached From Construction Costs?

Bradford DeLong writes Must-Read: Melissa S. Kearney: How Should Governments Address Inequality?

Bradford DeLong writes CPPC: Senator Amy Klobuchar: Drug Price Bills Will Likely Advance As Amendments to Larger Bills

Should-Read: Daniel Davies: From a logical point of view…: “I have now read that ‘google manifesto’… out of a desire to forestall people saying ‘but have you ACTUALLY READ IT?’… http://crookedtimber.org/2017/08/11/from-a-logical-point-of-view/

… [and not] out of any expectation that it would contain new or unfamiliar information…. Indeed it was your fairly standard evo-psych “just asking questions”… mulberry bush…. I have changed in my old age; I used to be depressed at the generally very poor level of statistical education, now I’m depressed at the extent to which people with an excellent education in statistics still don’t really understand anything about the subject. I’m beginning to think that mathematical training in many cases is actually damaging; simple and robust metrics, usually drawn from the early days of industrial quality control, are what people need to understand. Let’s talk about distributions of programming ability….

The male/female ratio at Google is… a variable under Google’s control. And when you think of the male/female ratio as an input rather than an output, you can start thinking about recruitment as a quality control process and everything becomes much simpler…. Recruitment… [is] a production process… to produce an output of employees of adequate quality…. Its failure mode is to recruit inadequate employees…. If the Google Manifesto was correct, then you would expect to see that Google was full of mediocre female employees, who had been hired by a process biased in their favour despite being inadequate to the task. Whatever the author of the manifesto thinks, Google does not believe this to be the case and as far as I can tell from industry blogs, it isn’t–female employees in tech are generally very good. This would, of course, be consistent with the hypothesis that the current selection process is biased against them.

I’d note that this argument could also be extended to one of the author’s other concerns about “ideological diversity” (mentioned in the context of Google, but most usually seen in discussions of university professors). If there were a genuine problem with a biased recruitment process, you would expect to see that the small minority of conservative professors were startlingly good and universally recognised as being so intelligent and productive of the best scholarship that they had got through the discriminatory process. One might call it the “Jackie Robinson Effect”. If, on the other hand, one had a situation where the writers of windy conservative manifestoes about not getting fair treatment were in fact mediocre whiners who inflated their CVs, then that would be evidence that there wasn’t a bias in the recruitment and retention system, and that in fact there was probably an inefficiency caused by the extent to which mediocrities were able to bump along because their face fitted in a homogeneous techbro culture…. The progress of gender equality in the workplace ought to be measured by the extent to which women can get into the ranks of time-serving dead-wood middle management roles….

marcel proust 08.11.17 at 4:59 pm: “On the veldt, all (surviving) programmers were mediocre because the star programmers couldn’t outrun the sabre-toothed tigers. Welcome back Daniel. If I may speak for my fellow bent timbers, we have missed you…

Bradford DeLong
J. Bradford DeLong is Professor of Economics at the University of California at Berkeley and a research associate at the National Bureau of Economic Research. He was Deputy Assistant US Treasury Secretary during the Clinton Administration, where he was heavily involved in budget and trade negotiations. His role in designing the bailout of Mexico during the 1994 peso crisis placed him at the forefront of Latin America’s transformation into a region of open economies, and cemented his stature as a leading voice in economic-policy debates.

Leave a Reply

Your email address will not be published. Required fields are marked *