iSixSigma

Six Sigma improvements mainly Hawthorne

Six Sigma – iSixSigma Forums Old Forums General Six Sigma improvements mainly Hawthorne

Viewing 46 posts - 1 through 46 (of 46 total)
  • Author
    Posts
  • #48321

    Dennis
    Participant

    How many believe that most Six Sigma improvements are mainly Hawthorne related?  If you go through the trouble of defining the problem and measuring the results, you will improve regardless of the subsequent analysis, improvement and control.  What percentage of 6 sigma improvements are related to the Hawthorne Effect?  I say 40%.  DMAIC is very clever.  It leaves SPC analysis and the hard work until the end, and maybe even half the time, you don’t have to bother because human nature fixes the problem for you…

    0
    #162564

    Brandon
    Participant

    I don’t know the percentage but your post has merit.
    Just like watching your kids – better behavior, usually.

    0
    #162565

    Hawthorne
    Participant

    Contrary to popular text books, a re-analysis of the Hawthorne data in the 1970s showed that many of the experiments that were reported to have a signficant effect did not have one, and those that showed statistical effect had a small effect size. Effect size calculations became only popular in the 1960s. So the original researchers are not fully responsible for the overblown conclusions drawn from the experiments. As Six Sigma practicioners, let’s not continue a myth that has long been falsified by data.

    0
    #162566

    qualitycolorado
    Participant

    Dennis,
    “An approximate answer to the right question is worth a great deal more than a precise answer to the wrong question.”
     
    I think people often miss the importance of the “Define” phase of Six Sigma — it is one of SS’s greatest strengths. I don’t believe it is merely Hawthorne effect; getting the Define step done right can often be a large part of the battle.
     
    Best regards,
    QualityColorado

    0
    #162568

    Mr IAM
    Participant

    The ol’ six sigma statement “measurement drives behavior” is very true.  But, the core concept of the Control Phase is to implement solutions that will be long lasting to sustian the improvement.  If sustaining the improvement simply is continuation of the measurement forever – your business will start to get pretty expensive to run once you are measuring everything forever!
    I think it’s bunk.  Gains made from simply initial measurement of the problem are not sustainable without other improvements. – M

    0
    #162579

    Confused
    Participant

    I agree with you, well said, the entire project is built from the foundation that is created in the define phase. But, i’m confused, is the Hawthorne effect a bad thing?

    0
    #162586

    qualitycolorado
    Participant

    Confused, Good evening!The Hawthorne effect is a temporary change in behavior (usually positive) that can occur when people are observed and paid attention to during a study or something similar, like a Project.The problem is “temporary”. If you have a Project that is pure Hawthorne effect, it means that what you have is only temporary — once you leave, things will “snap back” to where there were previously.So, yes, Hawthorne effect is not a good thing if you are wanting permanent change (which is the aim of Six Sigma).
    Best regards,
    QualityColorado

    0
    #162588

    I’m No Longer Confused
    Participant

    QualityColorado,
    This does not happen often but you did it – I’m No Longer Confused
    Well said and thank you

    0
    #162594

    Mike Carnell
    Participant

    QC,
    We used to call this the “cockroach effect” (we actually had a cockroach puppet from Rain Forest for an award).
    I know nobody has ever lived where there were cockroaches but if you speak to someone who has….. They will tell you when the lights are off the cockroaches come out to play. When you turn on the lights they all run and hide and your cockroach problem is solved – until you turn out the lights again. Thus cockroach effect.
    That may aleviate some of the concerns over Hawthorne’s data.
    Regards
    BTW all you people living in Florida (Darth) that don’t have cockroaches but you do have Palmetto bugs – those are cockroaches. Read that thing by Bill Shakespear where he is talking about the flower. Those are cockroaches no matter what you call them. Big ones.

    0
    #162603

    Beer
    Participant

    Wow, what an interesting build – thank you Mike
    Cheers

    0
    #162606

    Mike Carnell
    Participant

    You’re welcome

    0
    #162630

    Palmetto
    Participant

    Oh Mike, when was the last time you visited Florida? Here’s a link to the official “Palmetto Bugs”:
    http://www.dudeface.com/palmettobugs.html

    0
    #163112

    Cravens
    Participant

    I’m not sure I fully understand the Six Sigma semantic use of the term Hawthorne effect, but I have had practical experience with monitoring and reporting Key Performance Indicators outside the Six Sigma discipline.  When you “watch, record and report” conditions tend to “come into line” and when you stop things tend to drift “without purpose”.  When monitored conditions need further “tightening” you re-examine your KPIs revising them and continue your “watch”.  If that’s Hawthorne effect or not I’m not sure.  It seems to me that it is simply good management practice.
    Consider keeping a personal budget or balancing your checkbook routinely and you’ll understand my point.

    0
    #163113

    Fix the Real Problem
    Participant

    Hawthorne effect?  The Hawthorne effect is related to people-more interest from one party generates more interest in the uninterested parties.  I don’t like to rubber-neck accidents (can’t stand the sight of blood or any other bodily harm) and I will slow down based on safety.  However, everytime I’ve seen the details of an accident, it’s because the vehicles in front of me have slowed down to a crawl just to look.  Now, I’m sure not everyone has, but enough to make it so everyone else does.  However, once there’s nothing else to see, everyone is back to 65 mph.  You’re looking at an accident.  Now everyone is looking at the accident.  Once you quit looking, so will they and whatever perceived gains you had will be gone.  That’s the Hawthorne effect.  What you want to do is eliminate the Hawthorne effect by fixing the real problem.  If that means automating, better standards, more accountability, etc. then that’s what you need to do.  I’ve seen a lot of Six Sigma BBs that take the Hawthorne effect and run with it.  It makes the whole Six Sigma process look like a joke when everything goes back to the way it was. 

    0
    #163118

    The myth of Hawthorne
    Member

    Glad to see that there are still these dinosaurs out there who believe in the myth of the ‘Hawthorne effect’. “Once you quit looking, so will they”. That’s the very definition of infantile megalomania of someone who has overdosed on “pick-a-boo”. And of course, all of these, “I have seen posts” dressed in overzealous sincerity will keep on coming. Fix the real problem. The real problem appears to be in “Fix The Real Problem’s” little head. Go fix your real problem you silly thing.

    0
    #163121

    MW Richardson
    Participant

    I would like to bring us back to Dennis’ statement quoted below. The whole argument about the Hawthorne effect is a bit of a red herring. If you are running projects which rely on T distributions and AnOVa, you are kidding yourself. These are analogous to taking a snapshot of the process; what you see may be a transient effect. It may look like a significant difference but it may be due nonrandom variation in the process, not the effect of the action you took. That is why initiating a control chart it the beginning of the project is so important. We must understand the process variation over time in order to distinguish experimental effect from nonrandom variation.
    “DMAIC is very clever.  It leaves SPC analysis and the hard work until the end…”
    MW Richardson

    0
    #163127

    Brandon
    Participant

    Myth – certainly the best approach is to “fix the real problem”. However, you are too quickly dismissing the impact of “watching”. And, why do you presume the “watching” will stop? That’s like presuming the Control Plan will not be put in place or will not continue to be monitored.
    Is Hawthorne the cure? No. But you are patently wrong to dismiss its impact. Example – automated photo radar – we’re being watched, we respond.

    0
    #163129

    The myth of Hawthorne
    Member

    it looks like brandon had the urge to post under “Fix The Real Problem” :-)). You go and research the Hawthorne effect (including the original studies and the secondary analyses of the data). Then we can talk about “dismissing” the Hawthorne effect. Until then, your observations are your observations and yours only!

    0
    #163131

    Brandon
    Participant

    OK, Myth – you win. Isn’t worth my time to study your references. Apparently managing people has no impact. We can certainly save a bunch of money on middle mngt now. I’m going to go layoff about 30 people.

    0
    #163133

    The myth of Hawthorne
    Member

    You do that … btw, the examples that you gave is a cybernetic feedback. That’s very different from a “Hawthorne effect”. Next time, better know what you talk about before you spout off … and back it up with data. Isn’t that what you beloved Six Sigma is about?? 

    0
    #163135

    Deanb
    Participant

    Dennis,One problem with the original question (Are six sigma improvements mainly Hawthorne related?) is the presumption that the Hawthorne effects are single externally causes and effects, when actually they represent many complex external and internal causes and effects.For example, if you just consider changing the lights or other “external” sensing stimuli, then you probably are dealing with a very superficial and short term “cockroach effect.” However, if the participants feel “internal” stimuli such as feeling engaged, benefiting from information exchanges, and sensing they are a part of the synthesis process, then powerful culture positives can germinate which by themselves will become capable of improving many processes. With care these can become long term process capabilities in a culture. Without due care from management these capabilities can regress into appearing like “cockroach effects.” A relevant question might be “How much or your Hawthorne effects are natural external regressions, and how much are actually self inflicted internal regressions caused by managerial or cultural dysfunction?”If one is observing too many short lived effects, this can be symptomatic of a very common management/culture problem IMHO.

    0
    #163137

    BTDT
    Participant

    QC:We call like to call this the “senior management visibility effect”. Quite seriously, it’s very good for setting initial direction and priority for the team. A good control phase is how we keep the lights on.Please feel free to stretch the inappropriate business analogy to the extreme for the general amusement of the forum.;)Cheers, Alastair

    0
    #163138

    Fix the Real Problem
    Participant

    brandon, don’t waste your time trying to explain to the myth.  myth is either a naysayer by nature or someone looking to debate with no basis for it.  the hawthorne effect has no sustainability and from reading what myth wrote, he has no concept of what the hawthorne effect actually means.  the mumbo-jumbo about the hawthorne research being less than usable is something he grabbed from earlier blogs claiming the research was tainted.  taint is always relative.  we now know that the rubberband theory of the universe is not accurate and that in fact, the universe is expanding and continuing to increase in speed.  however, many things that we’ve learned are based on some of the rubberband theory being correct.  it’s the same with the hawthorne effect.  the research that myth references was done under entirely different research standards–however, this doesn’t dismiss some of the lessons learned.  only someone ignorant of how science works would simply dismiss data based on current standards.  perhaps we should dismiss newton’s discoveries based simply on his standard of testing was not the same as ours….

    0
    #163143

    The myth of Hawthorne
    Member

    There seems to be a recurring pattern: brandon talking to his alter ego. The following just blows that whole argument out of the water: “the research that myth references was done under entirely different research standards”. Well, so much so for an in-depth knowledge of that research study: The research that myth refers to is based on the exact same data … years later, to validate the initial claims of the researchers based on their own analytical procedures:-).
    Deanb’s observations are much more to the point and much more reflective and insightful of what the social psychological studies of Mayo and his team were all about. BTDT’s suggestion of how to name the phenomenon that dennis inquired about is also much more to the point. You are misusing the term “Hawthorne effect” because you obviously only have a very superficial and popular grasp of the content of the research and its history..  Just stick with what you know and can talk about intelligebly. Also, Newton has not been dismissed. His theory has become a special case of another theory, plain and simple.  

    0
    #163144

    Brandon
    Participant

    Good comments Fix. I’m with you – if that matters.

    0
    #163146

    Brandon
    Participant

    Myth – all this mumbo jumbo aside – are you really trying to say that “watching a process” (observing, tracking, reporting) have no impact on performance? Is that your message?
     
    Oh, never mind – my alter ego is answering for me (whatever the f that is).

    0
    #163150

    Stevo
    Member

    Stevo shares his view.  Draw your own conclusion.  (If you site me in a study, please give me credit)
     

    When I think somebody is watching me, I pretend to work.  Usually I just move this thingy on a string that is next to my computer.
    When I think nobody is watching me, I post to the forum on Isixsigma.
     
    Stevo
     
    Ps.  Please do not inform people to only write about topics that they are knowledgeable.  If that happens, I would have nothing to read.

    0
    #163151

    The myth of Hawthorne
    Member

    What I am trying to tell you is to think more about your responses and be more selective about what you post about. There are a few posters on this side who have done a lot of research, have a great amount of academic training and are accomplished practicioners. You don’t want to parade as a bafoon, or “ham actor” as Nero was called by Piso in 65 AD (Tacitus, Annals, 15.67).  

    0
    #163155

    Brandon
    Participant

    Stevo – you crack me up!!!
    Myth – go read some more classics so you have more quotes to cite and quit giving me garbage. I’ll put my accomplishments against yours any day.

    0
    #163156

    The myth of Hawthorne
    Member

    Brandon = a ham actor indeed, and what a bafoon :-)))))

    0
    #163158

    Deanb
    Participant

    Stevo,We also have to consider the world from the cockroach’s view. To them, bright light is just annoying interference and the real work can only get done in the dark. Perhaps some processes need more darkness.Deanb

    0
    #163160

    Brandon
    Participant

    DeanB – maybe that’s why Myth is writing so much – it’s too dark in here. Turn on some lights!!!

    0
    #163161

    The myth of Hawthorne
    Member

    brandon, thanks for confirming my suspicion: as you confessed yourself, your light bulb is truly not shining all that bright :-).

    0
    #163162

    Deanb
    Participant

    Dennis,If we remove the reference to Hawthorne, and re-characterize your observation to compare gains from “easy or soft” improvements vs “highly technical and hard” improvements, then your 40% from “soft” is not far from my observations.I have found that companies that strive only to do the “technical and hard” projects often improve at dramatically slower rates as a whole than those that get a big bang from the soft approaches.In a good six sigma program it is normal to see comparable value gains (and sometime more) coming from “easy-soft” projects. However, it is dangerous to turn this into an either-or argument, as each becomes highly dependent on the other very quickly. IMHO, every program needs a healthy mix of both, and finding this balance is the key to racking up major value gains over time.

    0
    #163163

    Dennis
    Participant

    Good post.  In my experience when the lights are off, management blames the poor manufacturing supervisor because the “black belt” is off on another project.  (“Why can’t you keep it going, the black belt did?)
    Then the system remakably perpetuates itself because the poor manufacturing supervisor has to take Green Belt training!
    Don’t get me wrong, I think the tools in the toolkit are great.  It is a great packaging job by Mikel Harry.  It is brilliant to leave SPC until the end because it is VERY hard to get manufacturing processes in statistical control (then there is the whole 1.5 shift bologna).  Leaving SPC until the end ensures that many projects are done or nearly done without having to show SPC.
    Lot’s of money for consultants.  But, six sigma is just a package job.  The effect of setting up the measurement system guarantees some impact and visibility.  It’s almost like job security for life!
    P.S.  My original 40% impact from Hawthorne is probably high, maybe 25% is more realistic.  What is the placebo effect in medicine?  Probably similar.
     

    0
    #163167

    Six Sigma Consultant
    Member

    No Hawthorne happens.  We can guarantee results. We can guarantee to make your process shift by 1.5 sigma.
    Our motto: “Six sigma makes money, not sense.”

    0
    #163175

    Mike Carnell
    Participant

    Stevo,
    I have to back you up on this one.
    I asked a guy once how long he had been working in a facility.
    His response “Just since they started watching me.”
    Regards

    0
    #163176

    Mike Carnell
    Participant

    DeanB,
    A cockroach convert. In the words of Steven Wright “You are skating on the other side of the ice.” Interesting observation.
    Regards

    0
    #163177

    Mike Carnell
    Participant

    Palmetto,
    I have to admit it has been about 4 months since I was in Florida. That is enough time for tremendous change in Florida. You guys can rip down wetlands and build condos faster than anyone I have ever seen. No wonder Hiaasen is p_ssed (for those in Australia that isn’t drunk either).
    From your link: “Odds on, if you live in South Florida, you have already encountered a Palmetto Bug or two. What are they? Simply stated, a Cockroach. But not your ordinary run of the mill cockroach. No, they are nothing like the cute little creepy crawlies you see in Orkin commercials.”
    Granted this perverse bunch is refering to cars but they speak the truth about your real Palmetto bugs as well. Good link. I loved those cars.
    Thanks.
    Regards

    0
    #163180

    Taylor
    Participant

    Dennis
    I have tried most of the day not to post on this subject as it someone a pointless endeavor and very ignorant question on your part. But, alas I cannot help myself.
    First of all, the DMAIC process does not save SPC for the end. ie MEASURE, how do you know what to improve, or even if it needs improved if you don’t have a baseline. Your understanding of each step in the process is obviously very little.
    Another statement is “poor manufacturing supervisor has to take Green Belt training!” Well I sure hope so, maybe then he will at least understand what the hell CPK is and why it is important to the business.
    As for the Hawthorne effect, In past projects it is always noted that once special attention is being paid to a project the process will see an increase in capability, as it should. The very core of the measure phase begins with process centering and gage studies. During this process everything is put back to “as good as possible” condition. If the Definition of the true problem was identified first then the real issue will continue on regardless of any minor improvements made by just watching the process closer.
    If merely watching the process closer acheives the results defined at the beginning of the project, then the improve phase will be simply to add process checks that insure this new found capability, something that is done all the time in Lean Projects. However this is very rarely the case and any Black Belt worth his salt will be able to sniff out the “cockroach” hiding behind the counter.
    Chad Vader

    0
    #163181

    Putnam
    Participant

    Placebo effect in medicine is normally estimated at 50%

    0
    #163197

    fake accrington alert
    Participant

    Please  explain  the placebo  effect  in  medicine and  how  it  is similar  to  the Hawthorne  effect.
    thanks  and  regards

    0
    #163198

    Six Sigma guy
    Member

    Both deals with psychology – in hawthorne people show change in behaviour when under observation in placebo people feel that by taking some medicine their illness will be cured – like taking some sugar pill and thinking its a powerful drug which will cure them( infact they dont have any illness at all they just feel they do!)
    Six Sigma success deals in fixing the process and not people.As deming said 85% of the problems are due to system failure so fix the system people will automatically get corrected. For people related issues go for change management. If you involve people into decision making and analysis they will have buy in and they will do it as part of daily life. Six sigma is not for punishing people if they dont do certain things it identifies reasons why people dont things and tries to put in system which enables people to appreciate and follow it.
    Control plans and control charts is for the system and not for the people.

    0
    #163199

    fake accrington alert
    Participant

    Thank You  for  the  excellent explanation,
    best  regards

    0
    #163210

    Ashman
    Member

    Very good point Fred. I think we should capitolize on the use of control charts in the Control Phase. This is where we make sure the “lights stay on” if they dim or go out, someone in the process owns the responsibility of finding out why and getting them back on.

    0
    #163214

    Ashman
    Member

    Let us not forget that an LSS deployment is also to change the entire culture and personality of the company. Sure the Hawthorne Effect is strong in the beginning. But as we go through the DMAIC we should expose the things in the process that are not performing to the levels necessary. Once again the key is to control those processes going forward with buy in from the people deep in the process. Then we have made the elements that contributed to the Hawthorne Effect part of the Control Plan.

    0
Viewing 46 posts - 1 through 46 (of 46 total)

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