Joseph Lemien's Shortform

post by Joseph Lemien (jlemien) · 2022-04-30T14:03:24.618Z · EA · GW · 6 comments


Comments sorted by top scores.

comment by Joseph Lemien (jlemien) · 2022-04-30T14:03:24.713Z · EA(p) · GW(p)

Would anyone find it interesting/useful for me to share a forum post about hiring, recruiting, and general personnel selection? I have some experience running hiring for small companies, and I have been recently reading a lot of academic papers from the Journal of Personnel Psychology regarding the research of most effective hiring practices. I'm thinking of creating a sequence about hiring, or maybe about HR and managing people more broadly.

Replies from: ryan_duplock, Yitz
comment by ryan_duplock · 2022-05-02T18:00:49.715Z · EA(p) · GW(p)

Yes! I think people on both sides of the table (employers & prospective employees) would find your insight useful! Recruitment is a big deal and is an area of much uncertainty for people. After all, most people are not regularly going through the process. It might help to quell some apprehension.

comment by Yitz · 2022-05-01T21:15:45.173Z · EA(p) · GW(p)

Please do! I'd absolutely love to read that :)

comment by Joseph Lemien (jlemien) · 2022-07-18T02:12:02.725Z · EA(p) · GW(p)

Evidence-Based Management

What? Isn't it all evidence-based? Who would take actions without evidence? Well, often people make decisions based on an idea they got from a pop-business book (I am guilty of this), off of gut feelings (I am guilty of this), or off of what worked in a different context (I am definitely guilty of this).

Rank-and-yank (I've also heard it called forced distribution and forced ranking, and Wikipedia describes it as vitality curve) is an easy example to pick on, but we could easily look at some other management practice in hiring, marketing, people management, etc. 

I like one-on-ones. I think that one-on-ones are a great way to build a relationship with the people on your team, and they also provide a venue for people to bring you issues. But where is the evidence? I've never seen any research or data to suggest that one-on-ones lead to particular outcomes. I've heard other people describe how they are good, and I've read blog posts about why they are a best practice, but I've never seen anything stronger than anecdote and people recommending them from their own experience.

It was an HBR article from 2006 (which I found as a result of a paper titled Evidence-Based I–O Psychology: Not There Yet) that I recently read that got me thinking about this more, but I'm considering reading into the area more and writing a more in-depth post about it. It lines up nicely with two different areas of interest of mine: how we often make poor decisions even when we have plenty of opportunities to make better decisions, and learning how to run organizations well.

Replies from: Linch
comment by Linch · 2022-07-18T04:51:03.077Z · EA(p) · GW(p)

I'm curious if you have evidence-based answers to Ben West's question here [EA · GW].

Replies from: jlemien
comment by Joseph Lemien (jlemien) · 2022-07-18T06:20:27.136Z · EA(p) · GW(p)

I haven't read any research or evidence demonstrating one leadership style is better than another.  My intuitions and other people's anecdotes that I've heard tell me that certain behaviors are more likely or less likely to lead to success, but I haven't got anything more solid to go on that that at the moment.

Similarly, I haven't read any research showing (in a fairly statistically rigorous way) that lean, or agile, or the Toyota Production System, or other similar concepts are effective. Anecdote tells me that they are, and the reasoning for why they work makes sense to me, but I haven't seen anything more rigorous.

Nicholas Bloom's research is great, and I am glad to see his study of consulting in India referenced on the EA forum. I would love to see more research measuring impacts of particular management practices, and if I was filthy rich that is probably one of the things that I would fund.

I'm assuming that there are studies about smaller-level actions/behaviors, but it is a lot easier to A-B test what color a button on a homepage should be than to A-B test having a cooperative work culture or a competitive work culture.

I think of the the tricky things is how context matters to much. Just because practice A is more effective than practice B in a particular culture/industry/function, doesn't mean it will apply to all situations. As a very simplistic example, rapid iteration is great for a website's design, but imagine how horrible it would be for payroll policy.