Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Data on the uselessness of LinkedIn endorsements

blog.interviewing.io

Post reply on HN
Print view View original

101–110 of 156 posts

Re: Data on the uselessness of LinkedIn endorsements

#101

Post by manis404 »

colinbartlett wrote: post #64
ben1040 wrote: post #8

I've been considering just deleting my LI account. It essentially has just been a sink for terrible recruiter mail; in 12 years of having an account I don't think I've once received any sort of actionable message through it. I've also turned off the endorsements because I was just getting driveby endorsements for things like J2EE. Is there a job-search downside to deleting your account, provided you have a decent Goo…

I am personally astonished that LinkedIn is still a thing. For me, it was the most obviously useless of all the social networks and it was the first one I deleted, upwards of 7 years ago now. The other social networks eventually followed but LinkedIn just gave me zero value from day one. It was nothing but recruiter spam. Do people actually get work and jobs out of LinkedIn? Or is is just a way to keep up with what o…

It's still a thing in EU. They are a source for cheap talent IMO

Re: Data on the uselessness of LinkedIn endorsements

#103

Post by prodigal_erik »

amyjess wrote: post #54
nickbauman wrote: post #4

I've been endorsed for sarcasm on linked in. When friends want to tease, we endorse each other for things like ms-dos and j2ee it's beautiful.

When endorsements came out, some friends and I discovered that you could endorse people for violent crimes. Yes, you too can endorse your connections for murder, kidnapping, and terrorism.

I distinctly remember "Homicide", which is actually a specialty intended for attorneys working in criminal law.

Re: Data on the uselessness of LinkedIn endorsements

#104

Post by xapata »

leeny wrote: post #97
huac wrote: post #83

Earlier quoted context omitted.

And what's the distribution over the languages, in terms of 'preferred interview programming language'?

People overwhelmingly chose one language to interview in, so there wasn't much fo a distribution. Unless you mean which languages are most popular?

> which languages are most popular?

Is there any other (reasonable) way to interpret that question?

Re: Data on the uselessness of LinkedIn endorsements

#105

Post by nazgul17 »

mendicantB wrote: post #42

Your analysis is extremely flawed. I really appreciate your sharing the data with us and I like your service. But, this is a poorly done and a far from subtle plug of your business at the cost of LinkedIn. 1) Technical ability vs # of endorsements Jesus. Hiding stats that you don't like through aggregations? And please read up on Simspons Paradox, which is clearly the case here just by looking at your plot. Try a bas…

As someone interested in data science, I am curious to know more: could you elaborate on your points?

Re: Data on the uselessness of LinkedIn endorsements

#106

Post by mendicantB »

minimaxir wrote: post #53
mendicantB wrote: post #42

Your analysis is extremely flawed. I really appreciate your sharing the data with us and I like your service. But, this is a poorly done and a far from subtle plug of your business at the cost of LinkedIn. 1) Technical ability vs # of endorsements Jesus. Hiding stats that you don't like through aggregations? And please read up on Simspons Paradox, which is clearly the case here just by looking at your plot. Try a bas…

> And please read up on Simspons Paradox, which is clearly the case here just by looking at your plot. What is the instance of Simpson's paradox ( https://en.wikipedia.org/wiki/Simpson's_paradox ) in the scatterplot? There are skews on both X and Y axes, but I don't see disparate trends. Would faceting by preferred language/experience/LinkedIn age show different trends in this context?

If you breakup the plot by any number of categories of technical ability, then there are trends. But to your question, what I was suggesting was that the aggregations done, especially with categorical data that is averaged, are very susceptible to this. And those clusters are reminiscent of situations like this:

https://www.researchgate.net/figure/256074671_fig3_Visualiza...

I'd more so like to see this analyzed against who got to the next round (their binary signal), or yes against preferred language, which I suspect will be much more telling.

The takeaway from that plot is, there is more to the story.

Re: Data on the uselessness of LinkedIn endorsements

#107

Post by mendicantB »

nazgul17 wrote: post #105
mendicantB wrote: post #42

Your analysis is extremely flawed. I really appreciate your sharing the data with us and I like your service. But, this is a poorly done and a far from subtle plug of your business at the cost of LinkedIn. 1) Technical ability vs # of endorsements Jesus. Hiding stats that you don't like through aggregations? And please read up on Simspons Paradox, which is clearly the case here just by looking at your plot. Try a bas…

As someone interested in data science, I am curious to know more: could you elaborate on your points?

First 200 pages

https://www.amazon.com/All-Statistics-Statistical-Inference-...

Re: Data on the uselessness of LinkedIn endorsements

#108

Post by mendicantB »

leeny wrote: post #96
mendicantB wrote: post #42

Your analysis is extremely flawed. I really appreciate your sharing the data with us and I like your service. But, this is a poorly done and a far from subtle plug of your business at the cost of LinkedIn. 1) Technical ability vs # of endorsements Jesus. Hiding stats that you don't like through aggregations? And please read up on Simspons Paradox, which is clearly the case here just by looking at your plot. Try a bas…

Author here. I appreciate the notes and am happy to revisit and make corrections when needed. To respond to your points: 1. As a sanity check, I did do a t-test of technical ability vs. # of endorsements before publishing. There is no statistically significant relationship between the 2. (P 2. What do you mean by "language matters here" (re the histogram)?

Hope it was helpful.

It's unclear what you t-tested here. Ideally, you would test for difference between groups of "Is there a difference in number of endorsements between people who got a "yes" in advancing to the next round or not". As a followup, is there a difference between those who's preferred was most endorsed or not?

I'm a bit stunned that you didn't recognize Language as programming language...... :(

As an example, people probably get endorsed for SQL or CSS far more than their programming language of choice that is tested in an interview.

Re: Data on the uselessness of LinkedIn endorsements

#109

Post by deftnerd »

Up until the LinkedIn redesign a few weeks ago, I used to run a bit of javascript every day that would pick a 100 people randomly on my LinkedIn "friends list" and endorse them for one of their listed skills that I hadn't already endorsed them for.

Every time I did that, it caused about 20% of the people to view my profile, which gamed my LinkedIn statistics so my profile was ranked higher in various searches.

I haven't rewritten the script since their redesign, but I'll get to it soon.

Re: Data on the uselessness of LinkedIn endorsements

#110

Post by closed »

minimaxir wrote: post #99
closed wrote: post #98

Earlier quoted context omitted.

Do you mean that you fit a simple linear model, of the form below? ability = b0 + b1*endorsements + error And when you say t-test, are you saying you did a t-test for the parameter b1? Usually when people refer to a t-test, without more information, they are saying they tested the difference of means between two groups. (or one mean against a number). See, for example, the Wikipedia article on t-tests: https://en.m.w…

> Do you mean that you fit a simple linear model, of the form below? That would be the form of the best-fit line in the scatterplot. (and it would make sense to assume that the t-test refers to b1 != 0, as there is only one group)

Edit: on second thought, you're probably right. I think I was too off the cuff in responding. Left original response below.

If by best fit you mean minimizing sum squared error, that's fair.

But to be sure, if someone said t-test, and they only had one group, I would first guess they were doing a one-sample t-test.

Even with two dependent variables and one group, I would think over whether they did a dependent t-test.

I figured it was a simple linear model (in this case a correlation) because they mentioned that they tested the relationship, and it makes sense, but it seems important to sanity check the use of the term t-test, which can be highly ambiguous (and I have seen used in very surprising ways).

Post reply on HN