Ranking vs. Rating Employees

Seems I’ve always been at odds with my friends in Human Resources…for one reason or the other.

(“Psssst…pssst…”)

It’s recently come to my attention that I’ve been, er, not always seeing eye-to-eye with a LOT of different people, especially those in authority.

I believe it began at Regina Pacis, the elementary school I attended (for grades 2-8) back in Brooklyn.

I think it was either in 2nd grade with Sister Mary Joseph Francis (the only 3-named nun I’ve ever come across) or with my 3rd grade teacher, Sister Mary Ruth.

🤔 Probably the latter…

We were studying English…various word types, proper pronunciation, sentence structure & other truly fascinating (🥱) stuff.

So there she is, teaching us the different sounds that letters make & providing examples to illustrate her points & help us better understand.

We all know the difference between “short vowel sounds” & “long vowel sounds”, right?

A “short a” (depicted as an a with a tiny u on top) as in “bag”, “cat” & “ham” versus the “long a” (ā) as in “bake”, “cave” & “mane”.

Everything’s going along smoothly…

…until she tries to convince us that the “long e” (ē) has the SAME EXACT SOUND as the “short i”!

😱  🙋🏻‍♂️ 🙋🏻‍♂️ 🙋🏻‍♂️

“That’s not right! They DON’T have the same sound!”

Note: For those of you out there who’ve never been subjected to at least 8 years of corporal punishment delivered by these “holy ladies married to Jesus” (their description, not mine), you have yet to really live!

Or die, for that matter!

Nothing like a 60-something year-old li’l costumed munchkin personifying the ol’ “spare the rod & spoil the child!” adage.

I know it ain’t Biblical…my description nor that saying…but it’s OK as long as I don’t get into the habit.

*pause*

*waits for the Catholic school boys & girls to react*

No, heh?

Nuns. Their garb is called a “habit”. Don’t get into the hab…

🚌 < < <  This is the non-stop bus to H🔥ll that I’ll be boarding shortly!

OK…

So, then, she starts pronouncing words in such a way so as to make the “long e” & the “short I” sound the same.

In my head, I’m screaming, “You crazy bastard, I don’t give a shit what your stupid teacher’s guide says or how many years you’ve spent teaching, YOU’RE DEAD WRONG!!!”

But since I knew a major beating would definitely be waiting for me when I returned home & handed my Mom the note Sister would write about me & then my Dad would hafta go to the convent for a meeting that night to avoid my expulsion from school (& God forbid it was his bowling night!), I decided on a less-dangerous approach…

…”Sister Ruth, are the words “FIT” & “FEET” pronounced the same way? “BIT” & “BEET”? A short i & a long ē are completely different sounds!”

She wasn’t one for student insurrection, so the next thing I know, we’re switching over to studying religion!

Then, in the 4th grade, with Mrs. Riccio (one of the few lay teachers in my Catholic school), she was teaching us “how to round numbers” to the nearest X.

The general rule is look at the “following digit”…the one that “immediately follows the digit in question to the right”.

If the following digit is 0, 1, 2, 3 or 4, you leave the digit in question alone. 42 rounded to the nearest 10 > > > 40. You “round DOWN”.

If the following digit is 5, 6, 7, 8 or 9, you add 1 to the digit in question. 573 rounded to the nearest 100 > > > 600. You “round UP”.

That’s the CURRENT RULE!

But years ago, the rule was a little different…that is, whenever the following digit was a 5!

“If the digit in question (that preceded the 5) was even, leave it alone.

if the digit in question (that preceded the 5) was odd, then round it up.”

Yes, that was the rule at the time. Only when the following digit was a 5.

It was a stupid rule…especially since it was WRONG!

We take a math test.

I get 98%…and totally flip out!!! (I LOVED taking tests & almost always got 100s.)

The question that she said I got wrong was, “Round to the nearest thousand: 36,587”

My answer was “37,000”.

But according to the “stupid, incorrect rule”, the (not-really)correct answer was “36,000”.

The “following number” was a 5 & the digit in question (6) was even so you left it alone.

(If the question was “Round 37,587 to the nearest thousand”, then the answer would be 38,000 as the 7 would increase by 1 to “become even”.)

THE OLD RULE WAS WRONG!!!

What should matter is the RIGHT ANSWER and NOT whether you followed a stupid rule. In reality, any “mathematical rule” is merely a process to follow to HOPEFULLY arrive at the correct answer!

However…

36,587 is CLOSER to 37,000 (my answer) than it is to 36,000 (“their” answer”)!!!

36,587 is only 413 away from 37,000…but 597 away from 36,000! That’s the TRUTH.

The “rule” is flawed (when the “following digit” is a 5)!

I argued with Mrs. Riccio for, like, forever. I even went up to the board to illustrate my point!

“You asked us for the correct answer. The correct answer when you round means the number that is CLOSER. That’s the definition of rounding! You didn’t order us to follow any specific rules…you asked for the correct answer!”

She wouldn’t give in.

But neither did I.

We continued to argue.

That evening, as I was eating supper with my family (BTW, it was only called “dinner” when we ate it in the afternoon on Sundays & special holidays or went out to eat), I brought up what happened on the math test.

(Note: In the 4th grade, we hadda get every completed test signed by a parent, then returned back to school the next day. Personally, I think it’s a GREAT practice to use & ensures that one’s parents actually know what’s the hell is going on.

And most forged parental signatures were easily spotted, unless, of course, you had a much older sibling who was a poor student. In that case, they already had a lot of practice faking Mom’s or Dad’s signature.)

I explained my rationale behind why I did what I did and while it initially was pretty tough sledding (“The teacher is always right!”), I finally convinced them that I was correct in this situation.

Not sure if I referenced the “long ē/short i” fiasco during my opening statement or if it was during my cross-examination of the expert witness, but I made sure to try to disprove the “teacher is always right” theory.

Besides, my parents were shocked with my embarrassing 98% mark, especially in math, so that helped my overall argument.

My Dad signed my test (usually, it was my Mom)…then wrote a note to my teacher. Of course, sharp Joe LoRusso put the note in an envelope that he sealed.

Yeah, I knew how to steam open an envelope (this was Brooklyn, remember?), but doing it without being noticed was impossible.

And I’m wondering if the school subject was actually called “arithmetic”, and not “math” or “mathematics”. I’m that old. 😭

Now, I’m NOT claiming that I was the driving force behind the rule eventually being simplified & thereby, corrected, but…

(Be thankful that I don’t discuss my theories on the Electoral College!

I won’t bother you with my arguments…supported by simple mathematics, no less…that prove that the EC is outdated & no longer applicable in today’s world.

I’ve argued vehemently with lots & lots of high school teachers & fellow students and to this very day, still argue with my very intelligent & learned friends who, in my opinion, use “flawed logic” to support their stance.

I guess I’m gonna hafta write another stor, er, essay on that topic.

Stay tuned.)

OK, now let’s return to the ORIGINAL topic of this whole thingie…

 

One of the biggest philosophical differences I’ve had with Human Resourced surrounded the “rating & ranking” of employees.

This difference in mindset was present when I was with Citi, Lehman Brothers, First American (then CoreLogic after the corporate spin-off) & finally, Nationwide.

In one way or another, all these companies used a “forced ranking” system whereby the top 10% of performers = “Exceptional”…the next 20% = “Outstanding”…the next 55% = “Average” (or “Meets Expectations”)…the next 10% = “Needs Improvement”…and finally, the last 5% = “Unsatisfactory”.

The “forced” part comes in where organizational levels…as low as a team &/or as high as a complete division…must ensure that their employees fit nicely into these buckets.

If you have 20 people, then it’s 2…4…11…2…1

100 people? 10…20…55…10…5

You can do the math.

To an extent, I’m fine with that.

Except if you have a unit, group, area, etc. that doesn’t fit nicely into that 10/20/55/10/5 curve.

What happens to a unit of mostly “superstars”? Only 10% in the top “Exceptional” bucket? Only 20% in the next “Outstanding” bucket?

That’s (only) 30% of your staff that are “above average”. But if you truly have mostly superstars, then it’s likely that 40%, 50% or even 60% of your team are “above average”!

A portion of them must now be ranked as “Average”/“Meets Expectations”…even though their performance is actually Outstanding.

(Yes, I fully understand that if you plead your case long enough to your boss, he may allow you a little room to exceed those top 2 buckets as he has other units/more employees that he can “adjust the other way” to offset your unit.)

Let’s look at the corollary where you have an average unit, with very few superstars or even, above average performers.

Yet 10% will get “Exceptional” & 20% “Outstanding”…even though some of that 20% are really “Average/Meets Expectations”.

But my real issue is that we use “adjectives” to describe their performance (that equates to a RATING, like you got in school), but we really put them in buckets on where they RANK with regard to other employees.

RATING is a firm assessment against set standards.

RANKING is how you compare vs. others.

When you went to school, if you got an A, you got an A, regardless of how many other students got an A.

The teacher wasn’t limited to giving out only a certain % of As.

(Note: Pls don’t get me started on that idiotic “grading on a curve” bullshit! That was actually RANKING the students, but fitting them into RATINGS!)

The issue is using words & phrases like “outstanding”, “usually/often exceeds expectations”, etc. to describe an employee’s performance…but we’re NOT rating them (which uses descriptive words & defined statistical performance), we’re RANKING them (putting them into order).

You could be using words to describe an Average/Meets Expectations employee, but his performance was truly Outstanding…but you had too many superstars & there wasn’t enough room for him at the Outstanding Inn.

Or vice versa, a truly Average performer may get RANKED as Outstanding because the unit didn’t have 30% (10% Except + 20% Outst) of employee’s who were above average.

I’ve always maintained that if you’re gonna use a RANKING system, then just say so. Be upfront & honest.

“You’re in the top 10% of performers” & stay away from specifically-descriptive words. Don’t call them Exceptional; just call them “Top 10%”.

Yes, I fully understand that “You’re in the 6th to 15th percentile (out of 100)” may sound too harsh or too plastic vs. “Needs Improvement”, but the truth is the truth.

And if you’re an employee in Team A (filled with superstars), you could very well be RANKED lower than someone in Team B (with only a couple of superstars) whose performance is actually not as good as yours.

Is that right?

You know what they say about the company you keep. But it may actually work in reverse with a RANKING system.

The worse your peers, the better you’ll look.

Not sure if things have changed much in the past 7-8 years in corporate America, but I doubt it.

And for those employees “on the edge”…

…between “Exceptional” & “Outstanding”…

…between “Outstanding” & “Meets Expectations”…

…between “Meets Expectations” & “Needs Improvement”…

…and finally, between “Needs Improvement” & “Unsatisfactory”…

…chances are about even that you’re getting screwed (or justly classified higher).

All during the year, you’re describing their performance ONE WAY, but when it comes to their year-end appraisal, they may find up in a higher or lower bucket than how you’ve been providing feedback all year.

Once again, we’re using words that may convey one message (and also have objective performance metrics matching the description) ALL THROUGHOUT THE YEAR, but may be “forced” to deliver a somewhat different message at year-end.

And, yes, I fully understand the rationale companies use behind “forced rankings” and while it will certainly work “better” the larger the organization you have, it certainly does have its flaws, especially when the group size is smaller.

That’s one of the basic precepts of statistics. As the base gets larger, so does the accuracy.

It’s that ol’ “plus or minus X points/percentage” that surveys use to describe their “accuracy” (or “confidence level”).

But that does not mean that it holds true in each & every part of the organization.

And when you have multiple teams performing the same jobs, while the forced ranking are “easier” at the overall organization level, there are way more anomalies at the individual team level.

Now, I can hear some people say, “Well, if you have “too many” or “not enough” top performers (top 30%), then you should adjust your performance standards according.

Raising them will lower the top buckets & lowering your standards will increase the top buckets.

But if you’re one of several teams that basically do “the same work”, you can’t just go around adjusting standards – – unless you do it across the board.

And even if you want to use “ACTUAL average performance levels” (derived from actual full-year metrics) instead of set standards established beforehand, that won’t work for the “multiple teams doing similar work” syndrome.

But it might just work if your team is unique in the type of work they do.

Wait until the end of whatever period (6 months of actual for semi-annual reviews, 12 months for year-end appraisals)…calculate the average performance, by individual performance metric…then you’ll have a “realistic standard” that’s applicable to your particular group & more meaningful (as it takes ranking into consideration in a much better fashion).

Just some food for thought…

 

As always, thank you so very much for listening!