Four Box Behavior: The Problem with Labels
August 29, 2006

by Ed Muzio

Four-Quadrant behavioral assessments are hardly new.  If you’ve ever been through a process in which you were shown a square or box, divided in quarters, and told it represented four personality styles, you know this is true.  In one model, there are four colors.  In another, there are four letters.  A third uses animals, and still a fourth uses very polished, businesslike descriptive words.  But they all use the same basic square, which was originally published in the 1920’s by a student of Carl Jung named Dr. William Moulton Marston. 

Although these models (and still others like them) appear at first glance to be very different, they all work in the same way.  And they all share the same flaws!  

 Here’s what happens… through assessment or self-selection, a person is assigned either
 one or two of the four possible elements.  Participants are told how to “talk to” each
 element, and how each element will “act.”  Ultimately each person chooses one of the four
 labels as his or her own.  Often times we are handed ribbons or coded name tags, so we can
 wear our “label.”  It’s entertaining and even mildly educational, and we leave the activity
 thinking about who else we know that fits into each box.  (“Mom is definitely chartreuse.” 
 “My boss is a piranha.”)  Some versions give us a simple graph; others allow us to pin a second label behind the first one with a dash. ("Uncle Bob is an R-L.")

It’s better than a parlor trick, but it’s not quite a science.  And in every group, there are some troublemakers.  These people aren’t sure where they fit.  They see elements of themselves everywhere.  They object to being placed in a box.  They claim to be different at work and at home, and ask which one is for real.  Instructors and facilitators prepare for these “difficult” students, and have tricks and strategies to keep them engaged and not let them distract the other participants. 

The problem is, the troublemakers have a valid point.  More than one, actually.

First, research has shown that only about half of us actually align to the framework that these simple models use.  Everybody else’s pattern is more complex than the sixteen (or fewer) possible choices offered.  So, the troublemakers who grow confused, and claim they don’t feel like the model fits them, are probably right.

Second, research has determined that how we behave in the workplace, or with our colleagues, or on the golf course, can be quite different than our natural inclinations would dictate.  But the simple models don’t allow for a difference between our “natural” state and our “adaptive” behavior.  So, the troublemakers who cry foul, and claim to be different at home than they are at work, are probably right.

Finally, research has clarified that there’s a lot more to that simple graph than we first realized.  Simple models ask a simple question: of four scores, which are your highest?   But that method misses half of the story.  It turns out that our lowest scores can influence our behavior as strongly as our highest ones.  If I measure your behavioral style and don’t account for your minimum scores, I’ve missed critical information.  So, the troublemakers who find fault, and claim that something seems to be missing, are also probably right.

That's why when Group Harmonics decided to use behavioral assessment as part of our practice, we were very selective.  We wanted something that kept the benefits of the four-quadrant model, but didn't keep the flaws.  Something based upon substantial, current research.  After all, our executive staff is full of engineers!

We found it.  Based upon the same four-quadrant research, the tool we use today is to the simpler models what the PC is to the abacus.  Rather than blindly choosing from high scores, this approach conducts a full analysis of all results, high, low, and moderate.  Rather than reporting one behavioral style, this tool separates the displayed or adapted style from the natural or inherent style.  And rather than force-fitting the pattern to one of 16 possibilities, this tool considers over 19,000 mathematical results, reducing them to over 300 descriptions and 60 possible patterns.

And, perhaps most importantly: Rather than producing a few sentences or bullets about the person, it produces a twenty five page custom report.  The report comes with descriptive text, communication parameters, and job and environmental information, to list just a few sections.  It also includes a graphical representation tool that allows for the summary of four dimensions of behavioral data in a two-dimensional plot:  A visual representation system which is most decidedly not a box.

So if you need a behavioral assessment to use as an icebreaker or to get a laugh, any of the simple models will do.  But if you want to truly coach someone, to help two people in conflict come to resolution, to deeply understand the tendencies of an individual or a team, or to include in your pre-hire screening process a behavioral assessment so well-validated as to be proven nondiscriminatory across EEOC-protected categories, consider something more sophisticated.

.