iSixSigma

Statistics knowledge

Six Sigma – iSixSigma Forums Old Forums General Statistics knowledge

Viewing 39 posts - 1 through 39 (of 39 total)
  • Author
    Posts
  • #50908

    Savage
    Participant

    Hi, some advise needed -what level of knowledge of statistics is required to be successful as a master black belt .Degree / Masters in stats ? Cheers

    0
    #175606

    Darth
    Participant

    Based on some of the posts on this Forum and my own experiences….very little.  Not that it isn’t important, it’s just many folks aren’t willing to do what it takes to learn the stuff.  With that said, a successful MBB needs to have a much broader set of skills than just the stat stuff.  There are many soft skills that are required for success in addition to the technical skills.  If you happen to have a Masters in stat then you should be able to acquire the needed technical skills rather quickly and easily and should then focus on your people, facilitation, project management, teaching and change management skills.  I totally reject the arguments that claim a good MBB doesn’t need depth in the technical aspects of their job, they can get that from others.  I always wonder where???? a BB or GB or another MBB.  If MBB is supposed to be the top of the food chain then they need to have the ability to personally acquire the technical knowledge but not to the exclusion of the other skills as well.  You often see those who make the argument that they don’t need the stat skills coaching projects devoid of any rigor with big emphasis on basic tools and lean.  Heck, you can get a BB to do that so why do you need to pay the MBB the big bucks?

    0
    #175669

    Roger Noble
    Member

    I agree on the basic level with Darth’s comments. However, an MBB doesn’t need to be the top of the food chain for Six Sigma statistical application. MBB’s guide the deployement and program in general, ensuring alignment to corporate strategic initiatives and that resources are allocated efficiently for improvement and training efforts. Therefore, the MBB is typically more of a facilitator, leader, mentor for the Lean Six Sigma effort, and less of a statistical whiz. Sure, the more you know, understand, and can apply the statistical toolsets, the more effective you can be in the MBB role. However, stat knowledge and application are critical to success in the role. The ability to guide the program and efforts of the rest of the participants is what will make the program win or lose in the end. Besides, there are plenty of software packages out there that do all the number crunching for you and some even tell you which ones are the most appropriate to do. So in the end, how much statistical expertise do MBB’s or BB’s for that matter really need?
    Something to ponder: I have found that those who are too focused on rigid text book statistical application tend to overlook factors or potential solutions that may in the reality of a variable world, be the most appropriate and probable option leading to a positive outcome. Afterall, all stat text books start with the phrase “given everything else is constant, use this…” When have we ever encountered a situation where the world stops while we evaluate something?

    0
    #175680

    Mikel
    Member

    This may well be the dumbest post I’ve seen in weeks (caveat – I
    have not seen all posts).Wrong. The number one job of a MBB is a mentor.

    0
    #175687

    Darth
    Participant

    Stan,
    That’s a cop out answer. What do you mean by a Mentor? Yes, guiding the project and coaching the Belt is certainly an important component of the MBB job and is critical to the eventual completion and success of the project. Then again, properly using the tools and methods is also important to the outcome of a project. If not, we do away with all the tools and rely on intuition and gut feeling as we did in the good ole days and hold hands and sing Kumbiya. Granted, the MBB doesn’t have to be the source of all technical knowledge but they better possess a whole bunch. If not them, then who? Relying on pushing the buttons in Minitab is not the solution. Minitab will do an ANOVA all day long regardless of the data. If a DOE is needed, do we just wing it since most BB curriculum doesn’t cover it in much depth? Or do we hire a consultant…wait, that’s not a bad solution. I don’t believe the two skills are mutually exclusive. A stat nerd without the ability to “manage” a project will be a disaster. On the other hand, a cheerleader MBB without a clue will also result in a disaster. There needs to be a balance which is why there are so few good MBBs out there. It’s a tough job and the hacks are more than willing to try and diminish the value of the skill sets because they are too lazy to put in the work to develop both. The nature of some of the posts on this Forum should make this obvious. So bite me!!!

    0
    #175690

    Mikel
    Member

    Agreed. Roger’s answer is still dumb. Thanks for elaborating why.

    0
    #175691

    Darth
    Participant

    I’m glad we cleared that up :-).
     

    0
    #175692

    Bob Rome
    Participant

    I’ve been an active MBB/Lean leader for 10 years and a BB before that doing projects – lead two deployments and trained/mentored  100’s of BB’s & GB’s.  I only share that because I’ve seen a lot of folks & projects over the years and learned a lot.  I’m a deployment leader presently as well.  THE MBB NEEDS TO BE AN EXPERT IN STATS, how they are applied and interpreted.  If you can’t help your BB’s draw solid conclusions to solve their project problems, they will fail.  As Darth eluded – someone has to have the answers – and if it ain’t the MBB who?  Where else are they going to go for answers?  You better be preparered or you will lose cred.  Software won’t do your thinking for you – it doesn’t solve a thing – people do.  And you do need all the people skills as well, as well as business knowledge so you can relate to mgt and keep the program aligned and contributing to the strategic objectives.  After 10 years I’m as much a student as my first day of BB training.  You can’t know too much – and it ain’t getting any easier.  I have met plenty of lightweight MBB’s & BB’s and they don’t last long typically – unless they’re in an environment where ther is low expectations on results.  Delivering results consistently is the true test.

    0
    #175693

    Taylor
    Participant

    Sorry ROME, but I disagree, reason being, I have never met an MBB that was an “Expert” in Stats. They don’t exist, even if you think you are one of those “Experts” I can assure you, that you are not. Statistions are experts, MBB’s , Not. Unless they happen to be one in the same.
    What I have seen are MBB’s that are well versed at interpreting results produced from tools used in Six Sigma analysis. That is a far cry from being an expert in stats.

    0
    #175694

    Severino
    Participant

    I would hope that a MBB could at least to be said to possess a higher level of proficiency with statistics than the average BB.  To be an expert, I guess you would have to have a level of proficiency that is 6 standard deviations above the average BB since it has been shown that you will undergo a 1.5 sigma shift as you get closer to retirement…

    0
    #175695

    Taylor
    Participant

    Jsev
    Thats is the best explanation of the 1.5 sigma shift I have ever seen………..Even Stevo would be proud

    0
    #175696

    Severino
    Participant

    You should see what happens when I really get rolling.  At one point I’ve considered marketing the official beverage of Six Sigma:  OC Cola.  The product slogan would be “Sip OC”…
    … then I realized that Deming was against slogans so I chose not to market the product.  Needless to say sales have been less than stellar. 
    … I think its time for my medication.

    0
    #175698

    Mikel
    Member

    Amen brother.

    0
    #175699

    Darth
    Participant

    I never said that the MBB needed to be an expert in stats.  They do need to know a bit more than the BB.  Not from a textbook approach but to be practical and have a pretty good understanding of the whys not just the hows.  A Green Belt can push enough Minitab buttons to get reams of output.  But, what does it mean and what do you do with the information is what separates the tool jockeys from the true MBB. If you want an expert then you have to spring for the big bucks and hire Stan.  He will probably outsource it to Robert Butler and Vinnie.

    0
    #175702

    The Badger
    Member

    This is one of the best posts I’ve seen on the MBB role. I’ve worked for a range of MBBs. Some more technical than others. As an experienced BB myself, I look for the following qualities in an MBB: –
    – The ability to think at a strategic and a detailed level, knowing when and where to apply their skills, influencing people as required
    – To lead people, be it BBs, GBs, project managers, Process Owners, whatever
    – To coach and mentor BBs/GBs with their knowledge of the methodologies and tools that will deliver the best results for the job in hand being where being ‘pragmatic’ is no excuse for being wrong
    – To manage a program of activities that deliver the companies process management goals
    – To really understand and live the principles of BPM, Lean, 6-sigma etc etc challenging the status quo in a positive way
    I think the trick to being a good MBB is not to try too hard to do all of these things (granted, some can do all but are rare!) but instead, recognise their own strengths and weraknesses and focus on developing a team with all the component parts so as a group they add real value
    Even then, my personal belief is that a prerequisite for any MBB should be their firm understanding of the benefits of BPM and it’s strategic fit, their change management skills and, if they are going to coach/mentor thay’d better be good at the technical aspects.
    To illustrate the last point, I’ve seen some pretty poor attempts at mentoring (a little knowledge is a dangerous thing) causing confusion and the wrong result which drove a process owner to the wrong conclusion.
    MBBs, if you’re not sure of the answer to a technical question, use you network to find it. It will earn you respect!
    The Badger
     

    0
    #175711

    Stevo
    Member

    Matt,
     
    As with life, one size rarely fits all.  Granted, since six sigma has aspects of stats, an MBB has to have enough stats expertise to do the job.
     
    Some MBB positions are statisticians with a title, some are more coaching and consulting and others are strategic in nature.  If your BB’s are not running complex DOE’s, then you do not need the “Robert Butler” level on knowledge.
     
    People will usually say what every they are good at is important.  So naturally I think it takes a sophomoric sense of humor, that is our duty (I said duty).
     
    Stevo

    0
    #175712

    Mikel
    Member

    I agree on the “expert” comment. The MBB needs to be good at stats
    but I’ve always looked to have a Robert Butler or Ernesto Garcia within
    arms length of me. And I pay them as well or better than any other
    with a MBB title.

    0
    #175716

    Robert Butler
    Participant

      Since it seems my efforts on the fourm are being used as a yardstick for knowledge of statistics in this thread I’ll offer the following:
      It has been noted that the role of a BB or an MBB is not that of a statistician.  As I’ve said before what a BB or an MBB needs to remember is that, no matter how good their formal instruction, it is of necessity boiler plate.  It consists of overly restrictive but “correct” rules which, if followed, will prevent the BB or MBB from going wrong with great assurance.  However, because they are overly restrictive they can put a BB or an MBB in the position of believing in barriers where none exist.
      From the standpoint of statistical knowledge I think the hallmark of a good BB or MBB is that of an individual who has mastered the boilerplate, never forgets that it is boilerplate, and who has resoruces which will allow him/her to compare boilerplate with the real world of statistical methods.
      Since most people don’t have an applied statistician on call what any good MBB should have is a well thumbed reference library and the courage to use it in the presence of witnesses. 
      If you look up some of my past posts and do a simple pareto chart of the books I’ve referenced you can’t help but notice how the same books keep coming up again and again. I reference them because they are the most comprehensive and readable single book references on their respective subjects of which I’m aware. Those books, or their equivalent, should be the core of any MBB’s statistical library.
      Rather than have you rummage through past postings they are as follows:
    General Statistical Methods (Central Limit Theorem, ANOVA, t-test, etc.)
      Statistical Methods – 7th Edition – Snedecor and Cochran
    Regression
      Applied Regression Analysis 2nd Edition – Draper and Smith
      and
      Regression Analysis by Example – Chatterjee and Price
    Process Capability
      Measuring Process Capability – Bothe
    Quality/Process Control
       Quality Control and Industrial Statistics 4th Edition – Duncan
       and/or
       Understanding Statistical Process Control – Wheeler and Chambers
    Experimental Design
       Statistics for Experimenters – Box, Hunter, Hunter
       and/or
       Understanding Industrial Designed Experiments – Schmidt and Launsby
    Distributions
      Statistical Models in Engineering – Hahn and Shapiro
    Categorical
       Categorical Data Analysis – Agresti
    Proportions
       Statistical Methods for Rates and Proportions – Fleiss, Levin, Paik
    Graphical Methods
      The Visual Display of Quantitative Information – Tufte
       What I find so curious is the reluctance of individuals to spend the money needed to put together a basic library such as this.  The odds are very good that the money/time saved by checking the reference before doing, in a single instance, will be far greater than the cost of all of the books combined.
     
     

    0
    #175721

    Darth
    Participant

    Stan, do you have another arm available and how much do you pay??? Do you pay better than Giganto Consulting LLC or Carnell?

    0
    #175724

    Mike Carnell
    Participant

    Matt,
    I read the posts and couldn’t really figure out if my answer was to you are the side comments so you get it. My opinion is the overused  “it depends” answer. I think it is obvious that the MBB needs more knowedge than the BB otherwise it isn’t going to make sense having an MBB train BB’s. That pretty much determines that there needs to be additional training in stats for the MBB’s. To what level? I like Stan’s answer. They do not need to be statiticians primarily because the problems you run into aren’t that complex so paying for the extra knowledge is a waste of money. I do believe you need a statistician as a resource when you have issues that require real depth in statistical theory.
    I think almost every response is missing to a point. MBB’s are not going to be little clones. There are people who are good at mentoring, some at training, some at problem solving, etc. If you have a large enough organization you need to understand the functions the MBB’s fill and look to get the best at those functions and build a strong team. If you are small organization you can pretty much be assured you will be suboptimized in some tasks. If you remember the book “First Break All the Rules” you need to think about the message. You aren’t going to change these people much.
    I would also look at the stuff around PI and some of the psychometric testing. Let’s all get some statistical genious and then get a profile with a low A, Low B, High C and High D or doesn’t have the ability to think strategically.
    Take a look at Bloom’s Taxonomy and figure out what level a MBB needs to operate on. We make the assumption they need to be at the top and BB’s at Application. Those can be two different groups of people and proficiency in stats doesn’t have a thing to do with it.
    We always seem to get all wrapped around the axel when it comes to the stats stuff. People can get good at it because it is predictable. When you are adding, subtracting, multiplying or dividing (even long division) it only is correct one way. It is predictable but when you read a response from Robert Butler you can see he has transcended to pure academic understanding of the various tools. The people side can go an infinate number of ways at an infinite number of different times. Two distictly different skill sets and neither one of them is more important than the other (unless you can only afford one MBB). If this whole thing was about just statistics then you wouldn’t need a BB or MBB, you would just hire statisticians. If it was just about people we could get some of those guys from the camps around Austin that sit around campfires naked beating drums and chanting or better yet employ a bunch of the old retired hippies and go back to the days of “free love.”
    This question smacks of the silver bullet type question and answer when someone wants to say “see how cool I am” but it is also the same person that loves to run around saying “there isn’t any such thing as a silver bullet” when they discuss SS or Lean. Basically I think this is a pretty stupid question. I can’t believe it created so much controversy.
    Just my opinion.
     

    0
    #175725

    Mike Carnell
    Participant

    Darth,
    We place more emphasis on the intrinsic rewards than the extrinsic rewards. A chance to see the world, make a difference and on weekends (since you typically aren’t home) you get to prepare meals for orphans or paint schools. Not for everyone.
    It’s good for your karma.
    Regards

    0
    #175727

    Darth
    Participant

    Mike,
    Two things…first, my karma is fine, it’s the stock portfolio that sucks. Second, how did you have time to feed orphans and paint houses unless you were cooking and painting with one hand while guzzling booze with the other? Painting bars and cooking for hookers sounds more your style. Hope all is well and that you are getting out of Texas as we speak.

    0
    #175731

    Severino
    Participant

    Just to contribute something to the discussion that may be worth looking into Matt.  RIT offers an online course which results in a Masters in Applied Statistics.  I’ve been considering it myself quite heavily and it seems like just the right level of knowledge that the average MBB could use.  By no means is it a prerequisite for being a MBB but it might be worth looking into.

    0
    #175733

    Darth
    Participant

    Know some folks who took that approach.  Pretty decent program.

    0
    #175737

    Craig
    Participant

    Mike,
    I liked your post but do you  really think it was a stupid question ? Back in the Moto days, it was the statistical wizards who were the “elite” group. (Actually it was the statistical wizards who used the tools successfully who were the “elite”!) Today, you just have to know someone who knows someone who knows about statistics.
    To the newcomer I can see where this is a confusing topic. If one hires a Master Mechanic, the expectation is that the job will be performed with the highest level of expertise. When you throw the word “Master” in front of “Black Belt”, shouldn’t we expect some level of expertise in the toolbox drawer labeled “Statistics”? Or should we label that drawer “Statistical tools, do not open without a degreed statistician present”?
    If you are in IKE’s path….stay safe.
    HACL

    0
    #175739

    Severino
    Participant

    Thanks for the feedback Darth.  I find the concept of holding an advanced degree rather than a certification very attractive and it is a reputable school which makes it doubly so. 

    0
    #175740

    Darth
    Participant

    Yes, an advanced degree from a reputable institution will work out better than an industry type of certification which unfortunately is all over the place and diminishing in value.  Good luck.

    0
    #175783

    Mike Carnell
    Participant

    Darth,
    Your karma is fine and your stock portfolio sucks? those may not be independent events.
    The great part about working in Africa is the visible difference you can make with a little bit of effort. In the US where we have billionaires giving serious sums of money smaller efforts disappear in the noise. When we are there you can affect a lot of peoples lives by giving up a small amount of your time.
    Now wipe away the tears and yes after we finished painting and cooking there was time for a couple beers.
    Regards

    0
    #175786

    Darth
    Participant

    Mike,You convinced me to be more altruistic so I have transfered ownership of all my stock holdings in Lehman Brothers, Merrill Lynch, Freddie Mac and Fannie Mae to your favorite RSA charity. This should put Gates to shame. I feel good now, thanks for the encouragement. This week I will be providing financial opportunities for the local working woman.

    0
    #175787

    Mike Carnell
    Participant

    hacl,
    I moved from Houston to New Braunfels so I was out of the path. All the friends from Houston are doing well. It was amazing the job that people did to get the city prepared. There was an entire network of EMS guys spread across the southern part of the US that just kept colapsing in on the expected path until they knew exactly where it would hit. Pretty amazing people that gave up weeks away from home to help people survive. The worst part of all that is that after all the bad press around Katrina I’m not sure anyone will ever mention what these people did to anyone.
    Yes I think it was a stupid question. When we started doing Allied and GE (not as much there) there were lots of people telling why it was a crap idea and what was wrong with what we were doing. The same people that sit in meetings and can’t contribute more than looking at a capability study and questioning if it is normal or not. It took off without them and now they want to claim devine rights to being all omnipotent over the methodology. There will be a wide spectrum of personality and talents that make it work and those are a function of the environment that they function in. To believe that a person needs a masters in stats to be an MBB is ludicrous.
    Let me ask you this. We have watched group after group go back through the various tools and tell us that for transactional, helthcare, etc. these tools are not applicable so they won’t train those tools for their BB’s. So when we want a MBB we want them trained extensively in tools that their BB’s have never seen and never heard of so they can help them do what?
    As far as the elite at Moto? Mario Perez-Wilson, John Hathaway, Gary Cone (did have training but not a Masters), etc. There were more actually doing the work that were not formally educated than those that were.
    I can see it being confusing right up to the point someone actually figures out what the MBB is actually supposed to do. An MBB that spends all their time teaching isn’t creating benefits for a company. Outsource it. The MBB’s need to be involved in the business and integrating the methodology into the business which means it is more important to have someone with less stats knowledge and more business acumen.
    As far as that piece of paper goes you need to take it back to Bloom’s Taxonomy. You can get that diploma and still only operate on the bottom two levels of the Taxonomy. When you read Robert Butlers posts or work with people like Shar Stocker (Intel) or Jeff Heslop (GEAE) they are operating on a different level that is beyond the classroom. If you have a serious problem and need a DOE expert I have no issue with calling Bob Launsby and asking him for help. What he costs versus what he will save you is a good investment in anyones book. Why would we believe that every MBB needs to be one stop shopping?
    My apologies for such a disjoint response. Way to many random thoughts around this topic. I spent way to many years listening to the pointless pontifications of CQE’s that couldn’t do s__t but they had passed a test and thought they had a better pot to piss in. When I passed the test it didn’t change what I did so I never recertified. In any Value stream Map it would have been NVA.
    Now I am in deep s__t with ASQ and I have to submit an article for their magazine.
    Just my opinion.

    0
    #175791

    Mike Carnell
    Participant

    Darth,
    I will be watching the finacial markets for a dip after you move your funds.
    I don’t even want to know what that means about the financial opportunities for the local working woman.
    Regards

    0
    #175800

    Ryan
    Member

    Hi Robert,
    What book(s) would you recommend to learn the relationship between the distributions Chi-Square, t, F, and z?
    Thanks,
    Ryan

    0
    #175802

    Robert Butler
    Participant

      I suspect any good book on basic statistics would have this and it would probably be found under a heading such as “Relationships Between Distributions.” 
      The best short summary of this issue I know of can be found on pp.288-290 of Statistical Theory and Methodology 2nd Edition – Brownlee.  This book is old and I’m sure there are newer versions.  If it is still in print the section title is “The Interrelations of Various Distributions.”

    0
    #175804

    Deka
    Participant

    After reviewing this entire messeage thread, I have noticed that Matt has only posted one time..the orginal question.  Maybe Matt has learned that any question posted here experiences a 1.5 sigma drift. Or, maybe the process is so out of control, than Matt can not rely on the psoted data to allow him to draw any conclusion toward true north.

    0
    #175843

    Craig
    Participant

    It’s only fitting to answer Matt’s question as if I were a statistician, and preface it with “it depends”! (only poking fun here)
    If you look at Pyzdek’s Six Sigma Handbook, there is some good information on roles and responsibilities. Regarding the MBB, one of the items states” Technical expert beyond black belt level on one or more aspects of process improvement (e.g. advanced statistical analysis…………”)
    I suppose it depends on the types of problems one is faced with. Is your environment data-rich, how complex are the problems, do you need to run highly fractionated DOEs?
    My advice is to research the source I mentioned above, as well as ASQ and other reputable sources for overall MBB roles. If you are a PhD statistician, you are not necessarily a Master BB. If you are a Master BB, you are not necessarily a technical expert in advanced statistical analysis. Confused yet?
    More humor:
    A) No stats knowledge
    B) Some stats knowledge
    C) Advanced stats knowledge
    D) PhD
    E) Any of the above
    F) Any combination of the above
    G) Statistically indeterminate
     

    0
    #175864

    [email protected]
    Participant

    I just vomited on my keyboard…WHISTLE…I’m throwing the BS flag on Carnell post.http://www.bsflag.com/order.html

    0
    #175868

    BC
    Participant

    Regardless of how altruistic Mr Carnell is in actuality, he has to be doing what he’s doing for the love of it.  No amount of monetary reward can in any way compensate for all the days and weeks away from home, all the time in airports, all the long hours, b.s., etc.

    0
    #175871

    anon
    Participant

    Yeah, that or he has ADD.

    0
    #175882

    Mikel
    Member

    And what is wrong with Mike’s post?I know him, I think that was an honest statement.

    0
Viewing 39 posts - 1 through 39 (of 39 total)

The forum ‘General’ is closed to new topics and replies.