The Feedback Fallacy

Post by Ididntdothis » Thu, Dec 12, 2019, 8:20 PM UTC

smacktoward wrote: post #7

When I hear people talking up their policy of > “encouraging harsh feedback” and subjecting workers to “intense and awkward” real-time 360s ... I can't help but think how much a review there sounds like a capitalist version of a Maoist "struggle session" (see https://en.wikipedia.org/wiki/Struggle_session ). The point of a struggle session wasn't to build people up; it was to tear them down. A culture where the way y…

Post by whatshisface » Thu, Dec 12, 2019, 8:28 PM UTC

borramakot wrote: post #9

> Adding up all the inaccurate redness ratings—“gray,” “pretty gray,” “whitish gray,” “muddy brown,” and so on—and averaging them leads us further away both from learning anything reliable about the individuals’ personal experiences of the rose and from the actual truth of how red our rose really is. I don't understand this comment. How does averaging noisy signal, even systematically noisy signal, result in somethin…

Post by elwell » Thu, Dec 12, 2019, 8:36 PM UTC

whatshisface wrote: post #13

borramakot wrote: post #9

> Adding up all the inaccurate redness ratings—“gray,” “pretty gray,” “whitish gray,” “muddy brown,” and so on—and averaging them leads us further away both from learning anything reliable about the individuals’ personal experiences of the rose and from the actual truth of how red our rose really is. I don't understand this comment. How does averaging noisy signal, even systematically noisy signal, result in somethin…

The author is arguing that the real signal is zero and the systematic error is large, so you will always end up converging on a repeatable but useless value. Technically, taking only one sample could have gotten you closer because there is a 50% chance that the random error would have gone in the opposite direction to the systemic error, although the author is wrong to phrase that like it's some kind of advantage, be…

Post by borramakot » Thu, Dec 12, 2019, 8:43 PM UTC

whatshisface wrote: post #13

borramakot wrote: post #9

> Adding up all the inaccurate redness ratings—“gray,” “pretty gray,” “whitish gray,” “muddy brown,” and so on—and averaging them leads us further away both from learning anything reliable about the individuals’ personal experiences of the rose and from the actual truth of how red our rose really is. I don't understand this comment. How does averaging noisy signal, even systematically noisy signal, result in somethin…

The author is arguing that the real signal is zero and the systematic error is large, so you will always end up converging on a repeatable but useless value. Technically, taking only one sample could have gotten you closer because there is a 50% chance that the random error would have gone in the opposite direction to the systemic error, although the author is wrong to phrase that like it's some kind of advantage, be…

Post by beckingz » Thu, Dec 12, 2019, 8:53 PM UTC

whatshisface wrote: post #14

Ididntdothis wrote: post #10

Earlier quoted context omitted.

GE and Jack Welch come to mind. For a long time they were held up as an example of good management but somehow this didn’t result in a long term healthy company.

If you're hearing about good anything, you're hearing about good marketing. ;)