Data on the uselessness of LinkedIn endorsements
blog.interviewing.io
blog.interviewing.io
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 basic t-test, or rather some statistical rigor, the next time you try to make conclusions from data.
2) Most endorsed vs Language of Choice
As pointed out, this is not the way to frame your problem. By obfuscating what's happening in your histogram (which isn't technically constructued right either) you are again hiding what you dont like through aggregation. By the way, language matters greatly here, and you'd have benefitted by standardization.
3) Your conclusion
"After running some significance testing, though" and not posting your results or methodology, which is at best questionable after reading your analysis.
Again, I enjoy your service, but blog posts on technical ability that are ironically lacking in technical ability don't really make me want to come back.
PS: A little birdie told me that endorsements are quite strong in predictive power for jobs :)
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?
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.
> It turns out that people’s interview language of choice matched their most endorsed language on LinkedIn just under 50% of the time, so, you know, just slightly worse than flipping a coin.
A coin has 2 sides. How many programming languages are there, again?
Unless you mean which languages are most popular?
Is there any other (reasonable) way to interpret that question?
Right now, the most heavily used languages on interviewing.io are Java, Python, JavaScript, and C++ (in that order).
Ruby, C#, and Go are in the middle.
At the bottom are Perl and PHP.
Yet if there was no relationship between the two entities and there were N languages, one would expect the random probability of TRUE to be N/N^2 = 1/N.
Although, the writer doesn't seem to allege that he's comparing to random or anything like that.
I certainly think that LinkedIn endorsements aren't exactly a meaningful metric. But they're not meant to be. They're just a simple tool for finding people who can even nominally program.
If you're interviewing a mid-career programmer, and her endorsements are all for accounting and financial related things, you may have a fit issue to explore. Likewise, if you have a candidate with 200 connections and 3 endorsements, and that's atypical for the industry, you may want to focus more heavily on real reference checking.
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 < 0.335)
2. What do you mean by "language matters here" (re the histogram)?
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.wikipedia.org/wiki/Student's_t-test
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)
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).
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.
As a result, nearly all the endorsements on my LI have come from non-technical folks who just know buzzwords (e.g. python was my most-endorsed skill even even though this isn't my most used language by far, and any engineer taking a quick glance at my projects -- almost all native C/C++/ObjC -- would realize it. Now I'm getting lots of pitches for python gigs...)
If there was a way to (optionally) approve an endorsement beforehand, that would help.
I also agree though, people randomly endorse stuff. I see D.o.D. all the time or Security Clearance. What does that even mean? That's not a skill as much as, hey I worked there, or hey I have one.
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 Google cross section otherwise via conference talks, blog, github, etc?
I can't tell if LinkedIn has marketed itself into an actual perception that not having an account means you don't care about career or the like.
Worthless. Craigs List is cheaper and far more effective, with a well-written help wanted ad.
What kind of recruiter waits for responses to their help ad?
Recruiter spam dropped off. The couple of recruiters who were actually any good still send me an email and/or a call once a year to find out if I'm looking for anything. So I don't feel like I lost anything.
I don't get pelted with a ton of messages - maybe a dozen a month at most, and maybe 1/2 of them are actual ops I could be interested in/are applicable to my skillset. I've had several interviews which originated from a LI message - easily over a dozen, and I may have landed a job in the past this way (can't remember unfortunately). This is sharply contrasted with my experience during the .com days upon graduating college - I would get 2-3 daily calls from recruiters for all sorts of things.
Unsure if it's because I'm older, more skilled, have a note that I'm busy on my LI intro, or what... but I have a relatively in demand skillset (Rails/React dev) and am getting messages for quality leads, even quite a few for remote gigs (I have a note that I prefer remote work).
As I alluded to above - my life prior to LI had far time sunk due to being badgered by recruiters.
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 old coworkers are doing? I frankly do not get it.
Personally, it's a great business card replacement. If I'm at networking events, conferences, etc. I also prefer to have my personal (Facebook) and business (LinkedIn) networks separate since I use them for very different purposes. On occasion, I've also leveraged groups to get specific advice I wasn't easily able to find elsewhere.
In running my company, LinkedIn also creates a quick talent pipeline and allows candidates to get a rough understanding of who will be interviewing them so we don't need to spend precious interview time answering their questions about our backgrounds.
I don't know of any replacements that do all of these things under one roof reasonable well. I do know of a few individual tools that offer incremental improvements over what LinkedIn offers, particularly in the area of finding qualified candidates, but in other areas (ex. digital business card) LinkedIn is the only real game in town.
Yes. Multiple times.
I ignore recruiters if I even get a hint that they're not employed directly by the company they're purporting to be with. It's become less of an issue within the last year or so.
Yes because yes. When I last looked for a job, I poked around LinkedIn to see who was working at companies that I found interesting, got a former co-worker to refer me for his new place, and got a job there.
I would compulsively check my account however, and felt the need to respond to all kinds of messages I got on there in some form or fashion. When I took an honest look at the time I put into using it and what I actually got out of it, it was clear for me it was a time sink with no reward other than 'feeling like I'm doing something to help my career' when I could have spent all that time actually doing something that mattered to me.
Edit: If a company won't hire me due to not having a LinkedIn account, I guess that's a downside, however I don't see how a profile there is much better information than a resume I've prepared.
That being said many (including myself) recruiters now use other sites to find developers. But it's definitely less than the amount who just search linkedin. It doesn't seem like you were getting very good emails from recruiters to begin with so it may not be an issue for you to get rid of your account.
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.
I've read your comment thrice and have not been able to understand it! It's starts out mildly reasonable, then drifts out into bot-territory
To rephrase:
Have you ever encountered a flash of insight that seemed to approach a challenge from a completely different angle? It could be something like someone solving a deeply technical problem in a non-technical way, or the first time you were introduced to affordance [1].
This is similar. Instead of the obvious solution of injecting buzzwords into his LinkedIn profile, deftnerd did a social hack that nudged his own connections towards viewing his profile, equivalently boosting his visibility.
What I meant by my questions is that I'm interested in learning both the flash of insight and the thought process that led to that. Sometimes the reason is as bland as accumulated knowledge (but even then, what knowledge?). But if not, maybe I can learn a completely new perspective from which to approach a whole class of problems.
[1] https://www.quora.com/What-are-examples-of-affordances/answe...
I am, by nature, a hacker. I often search for unconventional solutions to problems, even when conventional solutions are already available. It's a form of thought exercise.
I take great joy in brainstorming problems and not stopping even when the solution presents itself.
For example, housing for refugees might result in coming up with plans for flatpack structures, designing automated machinery to manufacture bricks, to research on genetically modifying gourds to grow large enough to be able to be harvested and turned into structures.
As a child, I hoped to work for a think tank one day but the opportunity never presented itself. Instead I attempt to build small automated micro-businesses whenever a new idea captures my interest.
This is my old (now unworking) code: https://gist.github.com/DeftNerd/4abf63c7d0e680d62a2cd26d534...
It was able to pull from an API of "Suggested endorsements" so I could run it from one page. That "Suggested Endorsements" API endpoint doesn't seem to be around anymore.
I'm going to rebuild using Selenium. It's heavier, but it'll be easier for me to script actions across page clicks and I can run it headless as a cron job.
Endorsements tend to be just extra noise with all those pictures, and on many profiles there is rather important information (education, projects, etc.) below the skills section. You don't want your reader getting lost in the endorsements before getting to some useful info.
I'd hope a resume coach would help work towards selling yourself to all parties involved in considering you for a position you'd actually want.
Long version: If they have significant and current experience with hiring processes (especially in your industry), they'll be able to help you with everything from how to word and structure it to maximize your odds of getting through the automated filters nearly every big company uses these days, to figuring out (based on job postings) what the hiring company's HR department is going to care about. Which is often different from what the person doing the actual hiring will care about, but you need to satisfy both.
They also have visibility into other aspects (salary, titles, mobility) that recruiters often can be helpful with, too, but I've found recruiters sometimes have their incentives somewhat misaligned with those of their clients (e.g., like a real estate agent who makes more money if they focus on turnover over deal size, so the homeowner isn't happy to get $15K less for their house than they could, recruiters make their cut when someone gets hired for anything, at any salary, and a small percentage of a salary that's 10% lower is less of a concern to them than being 10% underpaid is to you), whereas a résumé consultant's job is to do their part to ensure you get the interview.
I run a couple businesses and therefore don't really have a résumé myself, but I've probably referred a total of a half people (friends, relatives, and people moving on from my employ to better things) to a close friend who is a résumé consultant, and they in turn have referred many multiples of that, given how happy they were with the results. Virtually all got interviews with the companies they wanted to. (Of course, a good résumé consultant will also tell you when you're completely on-paper-unqualified to enter that job via traditional means.)
I consider myself a fairly capable person, but seeing the work of a highly talented résumé consultant made me realize that that's just not an area I (or the vast majority of most people) are adept at navigating.
It's sort of like hiring an attorney to handle a legal matter one could technically handle without being an attorney. Sure, if you already have a talent for that kind of thing, and are willing to spend at least hundreds of hours working in that field, you'd probably be able to take care of that one task fairly easily, but anything short of that and it's far better to pay an expert.
A few hundred bucks to put one's résumé into the top few-percent of applications is money very well spent.
The thing I like about resume writing and coaching is that it's 100% focused on helping an individual job seeker, and there are no misaligned incentives.
I'm not a "resume coach", and I'm also not sure what one would do if they exist. I'm a resume writer, and I also coach.
Of course LinkedIn doesn't care. The very idea of endorsements is broken beyond repair, because it allows people to say that some guy knows something they've never had an opportunity to see him working with.
Fly-bagging (10) Squirrels (10) Walking (8) Fierce Conversations (9) Rhinoplasty (2) Puns (3)
After that, I just cared only about getting recommendations from CEOs and CTOs etc. on LinkedIn. These actually are useful and are a true marker of your reputation. Endorsements need to be completely overhauled or abandoned. They have the potential to be cool if somehow there were rep points that could be distributed for the given skill set that people were actually qualified in to give but that would be way too complicated for most people to even understand or care about.
It's just an engagement growth hack by LI. I understand why they did it, but I think it's short term gain for long term risk of irrelevance (of LI itself; endorsements are already irrelevant). Which, for a social network, is the biggest existential threat.
Yes, you too can endorse your connections for murder, kidnapping, and terrorism.
"If you were any other man, I would kill you where you stand!"
And I allegedly endorsed someone for "TeamCity" before I knew what it was!
Either they're faking the data or their UI design is negligent-to-malicious levels of bad.
I am not sure about anything on LI page - it so full of dark patterns that spamming random people with invites is always at most 2 clicks away. It might be similar with endorsements.
In other words, no kidding endorsements are useless. That isn't the point.
My potato salad is pretty good. My twilight fanfic, however, only deserves endorsement because I have not yet written any.
"It turns out that people’s interview language of choice matched their most endorsed language on LinkedIn just under 50% of the time, so, you know, just slightly worse than flipping a coin."
That would only be true if there were only two programming languages.
Screenshot: http://m.imgur.com/uyXXEPl
Source: https://www.linkedin.com/in/andrewjohnmcgrath/ [my recommendations]
I want to say that is how it started, but honestly I think I was just annoyed at people who had never worked closely with me endorsing me for "The Cloud", so I intentionally added some stupid things in the hope they'd prove to me once and for all they actually didn't know anything about me.
My profile has been like this for years now...one day when I actually want to do something different, I look forward to it acting as an automatic filter against the people I don't want to work with :)
"Most jobs are never available publicly, just like most worthwhile candidates are not available publicly (see here). Information about the position travels at approximately the speed of beer, sometimes lubricated by email."
How does interviewing.io plan to attack this problem? I recently have had a lot of trouble getting placed - through one of interviewing.io's competitors (still not employed), but am now somewhat cynical about the process.
There's this guy which I used to work with that was not very good in the job and used to distract others at work very often, not being really productive most of the time - but he was generally nice to everyone, so nobody really disliked him.
He ended up leaving the company, and was hired by another company which we had business with. After some time, his new employer called our boss to tell he left, taking with him all the code of the project he was and starting a company by himself with this code.
This may seem bad already, but it was not the end: a few years later, the guy comes back asking me for recommendations on LinkedIn - when I checked his profile, I found he got a lot of code from the first company as well - where we used to work together - and put on his Github, with copyright headers stating he was the author. In fact, he did that with code I personally wrote.
Of course, I didn't endorse him - but if I go to his LinkedIn page today, there are tons of recommendations by ex-colleagues, including recommendations for things that he barely worked with.
And this is just one example. Most co-workers I have or had that were really good in what they do have zero to a few recommendations, and usually in very specific stuff that don't every start to cover their knowledge.
So, my advice is: if you use LinkedIn recommendations to anything serious, you are doing something wrong.
Instead of deleting my LinkedIn account to get rid of recruiter spam, I deleted my entire history. My profile is just my name and my photo. That way I can expand my network of colleagues without getting bombarded with spam.
I highly recommend doing this!
I was semi-worried that my computer will explode when it is loaded. It takes so long and gives no visual feedback what it is loading. I would either put that near graph or better just use a static image
I am pretty sure I already knew this.
I've received several endorsements for each.
"You have 1 new notification!"
tap
... "Do you know these people?"
Turns out when they released them and every time people logged in they got prompted to "endorse", they just pressed all of the buttons (and expected reciprocation) whether or not they knew the first thing about the skill they were endorsing that I had...
Definitely wouldn't endorse the author for machine learning :)
There should be a law punishing arbitrary linear fits and ridiculous claims which are concluded from their correlation.
At that point, why not use static images for data visualizations instead?
Or
Some people think everyone should consume charts a specific way and thus disable everything that doesn't fit that precise consumption model.
Even if the tag is wrong, that someone vouched for you in some minor way might be statistically significant.
Of course, it could be that the UI encourages mindless tagging, therefore it's all cruft.
If you have done work for him, I'd argue the tag at least means he liked the work. If not, then yeah, useless data.
They encourage people to endorse their friends, and they give us some tags. Lots of people will, on the theory they're somehow helping their friends, do this thing.
But the tags they offer have a tenuous relationship with reality.
This is very useful and a collective statistics while being completely useless for individuals.
Didn't take long for people to endorse each other's BASE jumping and pole dancing ability.
You would need some kind of subjective survey on if any prospective employers/recruiters even look at the recommendations and how much weight they tend to give them.
If you look at the world through rational self-interest recommendations and linkedin profiles should be viewed sceptically.