Slashdot Mirror
Are IT Certifications Meaningless?
superflippy writes "In his article Hiding Behind Certification, MIT's Michael Schrage argues that CIOs who rely too heavily on certifications as a measure of an employee or sub-contractor's abilities are wasting their companies' money."
here here! by the time you have gone through the hoops and mastered their little quizzes, much has become irrelevant and you are out of touch with the issues in your particular workplace. what ever happened to being able to give a decent discussion to determine what is important in an employee? have management become so out of touch that they no longer know what questions to ask?
As I am sure many will agree (and say here), MSCE consists mainly of buying the books and decent memorization skills.
I am a high school student and took a course through a vocational center for Network+ and iNet+ certification. I received 900 on both exams (perfect score), yet don't feel as if I know much about networking at all beyond the basic "this is a Cat5 cable" and "this is how to configure a network interface in Windows." The fact that anyone can get a perfect score, let alone a teenager like myself who does computer stuff as a hobby, shows how meaningless these certifications really are.
In one aspect, we can all thank Microsoft for this one with their MCSE mills which turned out a bunch of talentless mouse jockeys. Mind you, not ALL are talentless...but a lot I knew from the boom were. This had the unfortunate effect of taining a bunch of people who really didn't care about much more than dollar signs.
Now, I know a lot of people who have, as a result of articles such as this, let their certifications expire...which I think is a bad idea in some cases.
The problem with certifications is that in many cases they have been overvalued by the people who get burned by hiring the talentless paper monkeys. Unfortunately, certifications are still required in many cases to get through the HR vortex.
However, if certification is used as a minimum baseline of knowledge, it can at least determine a minimum amount of knowledge required. It should be part of a set of tools used to gauge the quality of a candidate, and leveraged by the employer as part of a further interview process.
I'm standing in defense of certifications, partly because I renewed my CCDP and am working on my Solaris 9 certs. Exciting? Not really, but there is still a minimum amount of knowledge required, at least conceptually. To me, it's a validation of my experience that I can at least still learn something. At a minimum, I'm trainable...and familiar with concepts that the application/hardware vendor wants me to know.
Now, for the other tools...it depends on who really controls the interviews. Awhile ago in the network analysis team where I used to work, there was one particularly brilliant hardass. His only interview question was to hand the candidate a dry-erase marker and draw out their home network and explain how it worked, was addressed, and protected. As far as he was concerned, the group needed a net geek, and someone who didn't have their own network at home wouldn't be interested in the job enough to excel. Anyways, I digress...
The hardest test I've taken to date was the CWNA, which really threw me for a loop...and I dread the CWSP which I want to take by the end of the summer.
Take three candidates with roughly the same experience: one has nothing more than a high school diploma, another a college degree, and the other has a 4-year degree and some certifications...HR is likely going to pick the third candidate. Sorry folks...that's just how it is in the business world.
(CCNP - CCDP - CWNA - A+/Net+)
// Agent Green (Ian / IU7 / KB1JQO)
// IEEE 802.3: All 10base Are Belong To Us
I see a bunch of folks in this thread talking about how certifications and education are worthless, because they're quickly obsoleted in the fast paced, quickly changing world of IT. I call bullshit. Most certifications are worthless because the cert's exam questions become compromised rendering the test invalid. The people running the GMAT manage to put out a new test every thirty days, I don't understand why MS, Novell, and Cisco can't do the same thing.
For that matter, I've never understood why people are happy to post their "braindumps" of memorized exam questions on the Internet. The people you're feeding answers to are the same people you're going to be competing with for jobs. You're flooding the same market you want to compete in!
I've been in the fast, quickly changing world of IT since 1993, and for all that's changed, many "tried and true" tricks still work. They might need to be updated, but the concepts are similar. For example, suppose back in 1994 I had a bunch of identical machines I wanted to configure quickly. I'd pull out the old laplink cables, pull out my special floppy that would copy the disk from my working configured "master" to the "clones". In 2003, I use a network and Ghost software, but it's pretty much the same. In 1996, I made a firewall with a floppy disk and an old 386. I needed a router in a pinch a few weeks ago, and I made one with a bootable linux CD.
In IT, understanding a few basic concepts will get you a long way. Until earlier this year, I'd never touched Windows XP - we hadn't used it at work, and I have Macs at home. But when a few Windows XP computers showed up in the office and on customer's desktops during support sessions, did I throw my hands up and whine, "Omigod! The fast pace of the quickly changing field of IT has obsoleted my skills and left me behind!" No, I didn't - I applied what I'd learned from previous Microsoft operating systems and *I* *figured* *it* *out*.
If you took someone off the street and taught him Windows NT 4.0 inside and out, then gave him a computer with XP or Server 2003 on it, it's not like he's going to be completely lost because the tech blew right past him. He can take the skills he's already picked up, and apply them as he learns a new system. Same thing with certifications. If I've been using and am certified on Netware 4 (and I mean CNE-level, not a CNA), then I'll probably be able to get the hang of Netware 5 pretty quickly, even if my certificate doesn't say so.
A certification, or any sort of technical training is valuable if you learn its main lesson - how to think when looking at a particular manufacturer's products. If you think the goal of the certification process is the piece of paper, you've missed the whole point. The problem with most technical cetification testing programs is how easily they can be "gamed". Someone who's learned what's really supposed to be taught by the certification process is invaluable. Someone who's memorized the answers off a few dozen braindump sites will be near useless.
Blogging Weight Loss, Distance Education, and more at verlin.com
Those that can learn on their own
And those that must be trained and tested.
You, unfortunately, fall into the later, and THAT is whats wrong with certifications. As you yourself articulated, the hardest interview you ever had was what those in the former category would consider the EASIEST. If you truly are an IT person, you don't need some silly piece of paper to prove your skill - you can simply convey it by talking about yourself, and showing that you learn on your OWN. If you are another of these papermill creations, that has to be sent through training to learn your IT skills you are of NO USE. The market and technology changes too fast to accomodate people with certs that need to be trained, and that is what the author of the article was really dancing around.
We live in a time when its not enough just be someone that can be taught how to run a Cisco box, how to configure a sun or install patches on a Windows box - you have to be someone that learns extremely fast and enjoys the process of change. Ergo, the interview that you dredded, show me your home network, is possibly the best way to know if someone is truly qualified for any IT position. No certification on earth can prove that someone has genuine raw talent and enthusiasm for their work. At its best, it just shows you can take a test. Whoopie. Show me what you can DO and how well you do it.
Certifications are a joke. As the old saying goes, what do you call a doctor that graduated from the worst medical school on earth at the bottom of his class?
Doctor.
Would you want that highly certified doctor working on you if you had a choice?
I was amazed that candidates with networking-related certs couldn't adequately answer basic questions like the difference between shared and switched Ethernet, or the purpose of a subnet mask. Eventually it got to the point that I was less likely to consider a resume that had certs listed prominently compared to a resume that had no certs at all.
There are two attributes that I found were most likely to result in a successful, productive hire:
- Good interpersonal skills. Sounds trite, I know, by in my view IT is a customer service position. You should enjoy helping people, not get riled easily, and be able to talk to them on their technical level without being condescending. Candidates with successful experience in front-line retail sales (department stores, automotive shops, etc) often downplayed this element of their work history, but I found it to be a positive indicator of a "customer service" mindset.
- Self-motivated technical experience. Many people find it hard to break into the IT industry; that's fine - what did you do in the meantime? Build a home network? Put together a PC from components? Try other operating systems? Do volunteer IT work for schools, libraries, churches, friends & family? Great. Write some software of your own, esp. OSS? Even better! I found that the candidates who explored and learned new technologies just because they thought it was cool made the most capable employees when it came to integrating diverse systems and solving odd problems.
So are certs a waste of time? Not necessarily. My perspective is specific to a startup environment, where everyone needs to be able to do a lot of different things. Large companies often use certs as a filter, though, and if you don't have them you might not even get in the door.Plastering your certification logos across the top of your resume is unlikely to impress anyone who is competent technically. You can still mention them, but make sure to have plenty of evidence of actual doing in addition, even if it's not formal job experience.
Fortune 500 companies are run like you say, full of HR staff who can't tell a valid candidate from their ass, so they latch onto some kind of meaningless benchmark like a piece of paper which, in all reality, means that individual spent a crapload of money to get that piece of paper. In other words, they're hiring out of the good ol' boy network. New money, old money, you're hiring the upper crust.
This isn't the complete picture. I have a friend who works in HR at a very large corporation. I commented on their "scoring" system that weeds out a lot of people simply based on experience-based questions for each position (ie. "do you have a bachelors in ____, do you have experience with SAP"). I told her a lot of very good people probably won't score in the top 10% that they actually look at.
She said that of course, N*ke wants the very best person for the job. But each position may have a between 100 and 1000 applicants. Even if they simply cut the bottom 90% based on their score, they feel reasonably certain that they'll still get someone who be able to do the job very well... even if the best person was in that 90% they didn't consider.
It's kind of the like the decision-making problem of "value of perfect information". When making a decision, you try to evaluate "what would the outcome be if we had 'perfect information' that would give us the absolute best outcome". You then figure that you'll have a certain probability of a "good outcome" and determine the cost for that. The difference in return between your reasonably assured "good outcome" and the "very best" outcome is the most you should be willing to pay for better information.
In the case of hiring, you could get near-perfect information by individually interviewing all 1000 applicants. But that would cost quite a bit more than interviewing only the top 10%. If you interview the entire field, what are the chances that someone not in that top 10% will bring enough value to the company to compensate for the much higher costs from interviewing more people?
It's a gamble, and a successful company finds the right balance.