Monday, February 6, 2017

Did 10000 people show up at a Charlotte job fair for Siemens Energy?

(What follows is classic blog material, i.e. an argument about a very small point. But it's in the New York Times, so hey, it matters.)

From a tweet by Derek Lowe, a link to an article by author Jeffrey J. Selingo in The New York Times about vocational workforce training with a rather compelling anecdote: 
When the German engineering company Siemens Energy opened a gas turbine production plant in Charlotte, N.C., some 10,000 people showed up at a job fair for 800 positions. But fewer than 15 percent of the applicants were able to pass a reading, writing and math screening test geared toward a ninth-grade education.
That's a hell of an anecdote - it certainly suggests that there's a problem with North Carolina high school education or the North Carolina workforce. But is it accurate?

Let's start with the provable facts. There is indeed a Siemens Energy facility in Charlotte, North Carolina. This facility was originally there and expanded to include a gas turbine manufacturing line in November 2011. It employs somewhere around 1500 people as of 2012.

However, there is no evidence that I can find (not in the Charlotte Observer or any other North Carolina media outlet) of any job fair being held by Siemens for the opening of this plant. I think that's kind of remarkable, especially since it seems to me that job fairs tend to get a lot of media coverage. It seems reasonable that a job fair being held in 2010 or 2011 or 2012 by Siemens would receive a fair bit of publicity, especially one that received a visit by "some 10,000 people." Those are minor league baseball stadium numbers.

Rather, here's a relevant anecdote by Julie Cook Ramirez at Human Resource Executive Online (written in 2014, emphasis mine):
But when the company sought to expand its Charlotte, N.C., generator and steam-turbine plant to include a gas-turbine facility in 2011, it hit a hiring snafu. 
The expansion required Siemens to hire an additional 1,500 people, nearly tripling its existing Charlotte workforce of 800. Since many of the area's textile companies had shut down, the area was rich with unemployed people eager for a new opportunity. But while nearly 10,000 people submitted applications through the Siemens website, the vast majority didn't have the requisite science, technology, engineering and mathematics -- or STEM -- skills.
So, this story has a similar narrative (our applicants didn't have the skills we wanted), but things have changed - it's applicants to a website. Next, here's a relevant article from Lori Montgomery of The Washington Post describing the opening of the Siemens plant:
State and local officials kicked in millions of dollars in tax rebates to land Project Cardinal. The state also offered $3 million to provide customized training to workers at the new plant. 
More than 9,000 people applied for the jobs; 4,700 were evaluated at state expense, starting with a career readiness test that measured math and reading skills. Those who needed extra help were offered free classes at the community college.
This (Ms. Montgomery's version) makes a lot more sense to me, although her account is not a direct contradiction.

I suspect what actually happened is this: the Siemens announcement was made and the North Carolina state economic development folks (see page 19) set up a website where anyone could apply to work at the plant. Considering that unemployment offices tend to push people to put in applications (especially for mass hiring events like this one), the website got a lot of applicants from 1) unemployed folks and 2) folks with less education. It's not especially surprising that a significant portion of them did not have particularly good math or reading skills. I think Mr. Selingo's anecdote was factually incorrect, and when compared with competing narratives, it loses quite a bit of its force.

(Regarding Mr. Selingo's larger point that "apprenticeships are good" and "college shouldn't necessarily be for everyone", sure, that seems very reasonable. I don't think anyone (outside of higher education administrators, anyway) is arguing against that. That said, it seems to me that states (unemployment offices, economic development offices, community colleges, etc.) already bend over backwards to supply manufacturers with help assessing and training their potential workforces for free. Why should we spot these folks any more goodies? Also, is there any actual evidence that manufacturers are seeing long-term job growth that would support such an investment by the state? My suspicion is "no, not really.")

18 comments:

  1. In these times when the chemist gets no respect and dignity (a low salary), check out this post!
    http://www.foxnews.com/us/2017/02/07/cameras-reportedly-catches-bart-janitor-who-pulled-in-270000-in-year-spending-hours-in-closet.html

    ReplyDelete
  2. I know a lot of higher education administrators and I have never heard any of them say that they thought everyone should go to college.

    ReplyDelete
    Replies
    1. You probably know more than me - fair enough!

      Delete
    2. Tony Blair did (at least, he set a goal of 50% of young people going on to higher education), thereby devaluing degrees. Perhaps this helped to create what old people tell me is the modern problem of graduates not being able to find graduate-level jobs - or perhaps it isn't.

      Delete
  3. So was the New York Times article Fake News or merely Fake Framing?

    ReplyDelete
    Replies
    1. Which is worse: an incompetent journalist who doesn't bother to verify the facts in their article, or a journalist who knowingly writes false information in order to manipulate their readers?

      Delete
    2. Dunno.

      Which is worse, heart or kidney disease? One is certainly more common.

      Delete
    3. I think it was fake framing, i.e. an anecdote that was too good to check.

      Delete
    4. Knowing dishonesty is a higher-order sin than a failure. It's possible the failure was a mistake (though they usually require both failures of the person to check their own work and failures of their institutions to validate their work appropriately), and it might be curable (the person could learn how to appropriately check themselves). Even if someone makes enough mistakes to make it clear that they don't care about their work, it probably isn't as bad as knowingly lying (though it approaches asymptotically over repetitions, and at some point involves willful failure of their supporting institutions). If you knowingly lie, on the other hand, there is no mistake - you did what you intended, and what has failed is your character, which supporting institutions can't fix. If an institution keeps someone who has lied, then their flaws are almost as bad, because they are knowingly supporting dishonesty.

      I think the Ayn Rand quote about being (somewhat) willing to accept mistakes but not any evil acts committed with knowledge is relevant.

      Delete
    5. Many, many,...many years ago, I read an article that originally appeared in the NYT that was picked up by a local paper. It was about a pesticide or herbicide that was suspected of doing harm.

      The article mentioned that County A in Alabama where the chemical was used had a higher than average cancer rate. Neighboring County B had a higher than average number of birth defects. The implication was that both "may" have been caused by the chemical.

      The question that immediately entered my mind was "what was the rate of cancer in County B and of birth defects in County A?"

      The only things I could think of as to why the article didn't contain that information were 1)the journalist knew what the numbers were, but didn't think they were "material"; 2) the question never crossed his/her mind.

      Number 2 certainly qualifies as incompetence. Would Number 1 be a mistake or an evil act?

      Delete
    6. Number 1 would be closer to evil - there's likely an intent to mislead (the numbers don't fit the hypothesis so they get ignored) rather than simple incompetence. At some point, you'd like an editor to ask these sorts of questions before the readers do, but I'm not sure if that's what they do (or can do) - in theory, it's better to pop your own balloons before someone else does, but that assumes you care.

      Would normal rates of the other diseases (cancer/B and birth defects/A) strongly hint that the differences are likely random variation rather than causal? Could there be variation of the effect with something else? It wouldn't be likely, but possible. It might also depend how far out of normal bounds the rates are.

      Delete
    7. Both of those "counties with higher-than-average adverse events" are probably just small-sample-size effects, especially if they were rural counties. Given that scientists struggle to think about probabilities correctly, it's hardly surprising that journalists also get it wrong.

      Delete
    8. Or finding hypotheses and not having an independent data set to test them on ("false discovery").

      Epidemiology is a PITA.

      Delete
  4. I also suspect that Siemens promised lots of "local jobs" when they got their taxpayer-funded payoff, so wanted to pretend to be ready to hire. Ultimately, I am guessing these jobs will be filled, but not with locals.

    So pretense all around: Siemens pretends to offer local jobs, the local employment office pretends to be helping lots of local job seekers, dissolute job seekers get to pretend they're actually looking for a job, government gets to pretend their taxpayer-funded payoff was good for local employment rather than just a bribe (likely to be returned during next election fundraising). Win-win-win-win!

    ReplyDelete
    Replies
    1. Funny how the New York Times is oblivious to how this game is played. I guess they don't get local tax breaks for their regional printing plants...

      Delete
    2. If everyone got them, though, the problems with the use of tax breaks to recruit businesses would become readily apparent, so the tax breaks are unlikely to be universal (they would depend on how much mobility the company can afford or is willing to facilitate - how much the company is willing to bargain with local districts). They may either outsource local printings (in which case they don't worry about the tax incentives), may not get them, or may not ask for or accept them (though I would figure the last is highly unlikely).

      Delete
  5. The information contained in my article, which you reference, was obtained directly from Mike Panigel, senior vice president and chief human resource officer for Siemens Corporate Human Resources U.S./Americas, via phone interview. You neglected to include that attribution, which is made clear within the copy. While the trend is to cry "fake news" at anyone currently working as a journalist, my work has always been meticulous. I report facts, along with insights from the people I interview.

    ReplyDelete
    Replies
    1. Hello, Ms. Ramirez:

      I regret that you have me at a disadvantage, which is that perhaps I misunderstand your point? I found your article believable, and the New York Times article not believable.

      Want to talk more? You're welcome to e-mail me directly: chemjobber@gmail.com

      Best wishes, CJ

      Delete