View Full Version : The "J" curve
10-10-1999, 09:40 PM
No real info, but my instincts tell me that the "J" curve concept sounds pretty accurate. In my experience, and observation he bulk of the people do their job as required followed by a additional amount of effort to look good, or make overtime money. I think there are far more slackers who do very little productive than there are people who are maniacs and create huge profits for the company. This concept is sound IMHO.
The supposed graph is either different than the bell curve, or impossible, and therein lies the skeptisism. The curve that would represent the "j" curve concept as explained would still look like a bell curve, but have a slope on the left representing slackers closer to zero, and a curve on the right with a slope closer to infinity. The image would look a bit like a ramp, or a "J" flopped over onto its front. Maybe thats what they are trying to convey as opposed to a exponential graph line.
10-10-1999, 10:14 PM
I'm not sure about a J curve, but I have seen several reports/studies/etc. that indicated that a bell curve was usually a statistical fluke. Most graphed populations come out on a skewed curve--usually favoring several very good individuals.
For example, looking at a class with 30 kids (or a department with 30 workers) a bell curve would say something along the lines that there will be 3 utter dolts, 5 poor performers, 14 standard (or mediocre) members, 5 better-than-average performers, and 3 all-stars. Any class or department that I have been in would generally show 1 dolt, 2 poor performers, 13 standard (or mediocre) members, 7 better-than-average performers, 3 really-really good performers, 2 all-stars, 1 outstanding member, and 1 super-geek. (And often in business, there will be no dolt, because it is tough enough to survive just the poor performers without putting up with a dolt in your midst.)
The bell curve is a hypothetical construct that simply does not occur very often in nature.
As to the decisions of management: I have seen outside consultants and internal task groups. They all have one thing in common: the group that presents management's preconceived notions back to management is the one that will be deemed successful/useful by management.
10-10-1999, 10:49 PM
Tom: Look at a graph showing your example. It's a bell. It rings true, if you follow my drift.
10-10-1999, 11:05 PM
Of course: my examples always ring true.
It is unfortunate, however, that so many people believe in the reality of the bilaterally symmetrical bell curve, when the skewed curve is much more frequently found in nature.
10-10-1999, 11:19 PM
The skewed curve is more often found in nature, or in the study of behaviour?
The bell curve is found all through nature wherever random behaviour is found.
10-10-1999, 11:47 PM
What is random behavior?
Earl Snake-Hips Tucker
10-11-1999, 12:02 AM
I work in a company that every 18 months or so hires some consultants to come in and examine some department to tell them why they're not doing their job correctly.
Then, the consultants file a report, and management does whatever they were planning to do before the consultants were brought in. I assume that same phenomenon occurs elsewhere in Corporate America.
In one roundtable discussion with some other employees and a consultant, the subject was our annual review and evaluations. One of us made the "bell curve" comment--that on either end you have a relatively small number of outperformers and underperformers, while the bulk in the middle are just doing their jobs.
She said "Oh, that's what the social engineers want you to believe." BTW, she was one. The last place I worked, we graded things on the "J" curve, apparently so-named because of its resemblance to the letter. "These people (at the bottom of the 'J' were the low performers, and we dealt with them accordingly. The others were the high performers."
Could someone help me out on this? It sounds like some social engineering mumbo-jumbo.
Earl Snake-Hips Tucker
10-11-1999, 12:03 AM
Sorry--meant to post this in GQ.
10-11-1999, 12:19 AM
Tom, I think you might be selling the bell curve a little short. When you're talking about statistics, a population of 30 is nowhere near enough to get an accurate reading.
Your example (the one with the "super-geek") could form a bell, depending on what levels of productivity you assigned to various classes. The lack of a "dolt" only means that the bell gets shifted to the left, with the graph starting with the "poor performers", in mid-curve.
Besides, the bell curve (and the "J curve" as well, I assume) are meant to represent approximations. The actual numbers will vary, particularly with small populations. They will also vary based on other factors, like the availability of workers in the business's location, the pay and benefits of the employees, and the current value of the goods or services the business is providing.
Mjollnir, I did a quick web search and the only things I could come up with there about "J curves" had to do with macroeconomics and a jazz record label. I was thinking about the way you described the curve, and it sounds to me like the chart is saying that a company will have an infinite number of employees being the most productive. (Something tells me I've got it wrong.) This is going to annoy the hell out of me until I can find out more about it.
10-11-1999, 12:26 AM
I was part of corporate America for several years (before they started downsizing). They had consultants contracted, and they were around for so long they had permanent offices. I've long maintained (even before I was released) that a better name for consultants would be "professional scapegoats". They're main purpose seems to be to keep the executives from looking like the "bad guys" when things start changing. Problem is, it doesn't work very well.
I had never heard of a "J" curve, but it sounds like the sort of thing a consulting company would invent to make it easier to sell their services to executives. If I'm understanding their description, it makes it seem like the CEO is doing infinite work.
Earl Snake-Hips Tucker
10-11-1999, 12:35 AM
Read before "Submit." OK
Part of my post was ambiguous. Here goes again:
She said "Oh, that's what the social engineers want you to believe." (BTW, she was one.) "The last place I worked, we graded things on the 'J' curve." (Apparently so-named because of its resemblance to the letter.) "These people (at the bottom of the 'J' were the low performers, and we dealt with them accordingly. The others were the high performers."
J curves are common in medicine - take weight, blood pressure, or any of a number of possible measurements:
There's a fairly flat, low portion of the curve where risk is low (diastolic BP 70-90, for instance). Above 90, there's a well-known association between increase risk of heart disease, kidney failure & stroke.
Increasingly, though, physicians are becoming aware that lowering diastolic BP below 70 not only fails to further reduce the risk of the undesirable complications, but also increses other risks - hip fractures, or auto accidents, for example.
This is just a surmise on my part, but if you drew a straight line between ability & productivity, & then added a "J" underneath it, you would find that the marginally qualified individuals were working near their potential, and that the best-qualified individuals were working near their potential, but the the worst performers were actually those with middling qualifications, who were not living up to potential, & who might tend to either be disruptive due to jealousy when others got better assignments, perks, or promotions, or who might might direct some of that unused potential towards sabotage or espionage. At least that is the mental image of a "J" curve in corporate settings that I see.
Sue from El Paso
Earl Snake-Hips Tucker
10-11-1999, 12:38 AM
And, Mr KIA, you're right on the way the "curve" looked--it looked sorta like the graph of an exponential function.
10-11-1999, 12:41 AM
You still left out a parenthesis, but I get the picture. Still, was "she" a consultant as well as a SE, or just a SE?
Anyway, I'm curious about this theory. I don't buy it, but I would like to know more about it. (Sounds like something good to hone my "debunking" skills.) If I come up with anything, I'll pass it on.
Earl Snake-Hips Tucker
10-11-1999, 08:15 AM
Just a follow-up on these social engineers:
In our round-table, we asked this consultant "Just what is supposed to come out of these meetings?"
She said, "Well, we can't tell you now--but about six months from now, you should start seeing some changes." All of the consultants were similarly close-lipped about it.
Six months later, and no word on their report.
Some of our "trouble makers" started asking "management" about the report. They said, "Uh, we'll get back to you on it."
They stalled for a year and a half before they released "selected excerpts" from the report.
Bottom line (off the record from people who actually read the whole report)--their recommendations were so nutso and wacky, that even "management" realized it, and didn't want the "common folk" to realize, "You paid all this money for these yo-yos and they told you THIS?????"
That's one group I don't think they'll hire again.
10-11-1999, 04:05 PM
As I understand it, a "J" curve represents the top half of the standard bell curve. The consultants operate on the assumption that the bottom half of the performers in a given field wouldn't have gotten hired in the first place, so that the employees in a given company already represent the top half of the curve. Thus, there assumption is that there are just a few star performers in the company while there are many mediocre performers the company could get rid of with no real loss of production.
10-11-1999, 04:23 PM
Low blood pressure causes AUTO ACCIDENTS?!
In an older person with stiff carotid arteries, sometimes lowering the pressure measured in the arm to nice normal values means too little is getting to the brain, leaving the person light-headed, prone to falls, slow-to-react, etc.
Sue from El Paso
10-12-1999, 05:08 PM
There is also a J curve effect in economics that is used to explain why the changes in the flow of real goods and services between countries lag behind very fast changes in currency exchange rates. As far as the Bell curve, in large enough population studies, the curve often does approximate the symetrical bell shape (statisticians can make this bell shape appear on small samples by a process called "eliminating the outliers" - basically if the data doesn't fit the theory, discard the data).
10-13-1999, 09:38 AM
Sorry to be late. Since you meant to post it in General Questions, I'll move it there now.
David B, SDMB Great Debates Moderator
vBulletin® v3.7.3, Copyright ©2000-2013, Jelsoft Enterprises Ltd.