Performance Appraisal Methods
6
Methods
BARS, 360, MBO, Forced, Graphic, Critical
6
Bias Types
Halo, Horns, Central, Recency, Leniency, Similar
5
BARS Anchors
Behavioral specificity per rating
deal six method cards · pick the right one
The Method Deck
Each card has its own validity, bias risk, and development cost. Higher validity isn't always the right answer — pick the card that fits the deck the manager has time to play.
⚓
methodBARS
◯
method360°
◎
methodMBO
✦
methodCritical Incident
▦
methodGraphic Rating
⚠
methodForced Distribution
tilted on purpose — pick a card, any card
how to read this
- › Green bar is validity — does it predict performance well?
- › Red bar is bias risk — how likely is the rater to drift?
- › Yellow bar is build cost — how much HR time it takes to set up.
- › BARS has the best validity-to-bias ratio. Forced distribution is the worst-of-both. Use the deck — don't default to one card.
six raters, six ways to drift
The Six Rater Biases
Picking the right method only goes so far. The rater's mind is the second variable. Calibration sessions exist because of these six failure modes.
bias 1
Halo
one good trait inflates all
bias 2
Horns
one bad trait deflates all
bias 3
Central Tendency
everyone rated middle
bias 4
Recency
last few months dominate
bias 5
Leniency / Strictness
whole team high or low
bias 6
Similar-to-Me
like-me employees rated higher
Exam Traps
360 is for DEVELOPMENT, not compensation
Using 360 for raises creates gaming. Peers stop being honest. The exam tests appropriate use cases.
Forced ranking has known legal risk
Often treated as a wrong-answer trap in best-practice scenarios. Engagement and disparate impact concerns.
Critical Incident is a documentation method
Not a standalone rating method. It supports other methods by capturing specific events as evidence.
BARS reduces bias by anchoring
Higher reliability than Graphic Rating, but more time-consuming to develop. The exam tests the trade-off.
Forced curve forces drift
Forcing a bell curve onto a strong team artificially deflates top performers. Engagement falls.
BARS pays back over years
Time-intensive to build. Anchoring ratings to specific behaviors reduces rater bias for the long haul.
Customer fills out a 1–5 card on food, service, ambiance. Easy but vague.
Each rating point describes specific server actions. 5 means greeted within 60 seconds and offered drink suggestion.
Customer agreed at the start that the meal would deliver three things. Server rated on those three.
Customer, kitchen staff, and other servers all evaluate.
Manager rates servers by ranking and only the top 2 get bonuses. Engagement risk.
Manager keeps a notebook of specific server moments throughout the month.
Ready to test your PHR knowledge?
1,700+ practice questions written by certified professionals.
Start Practicing PHR