What's the ideal scale be for a STAR ballot? 0-3? 0-5? 0-9? 0-100?

As it turns out, humans have a limit to how many distinct options we can typically remember and compare at once. We like having choices, but if we have too many, people get bogged down and overwhelmed. How many is too many? It turns out the ideal number is between 5-7. That's why phone numbers were traditionally 7 digits and why in school, there are 5 letter grades: A-F. It's a concept call "Cognitive Load" that has been extensively studied.

The 0-5 (six point) scale is right in the sweet spot, perfectly matching the amount of nuance naturally found in an average human opinion. That's why when we need to accurately measure public opinion on important questions, the 5-star rating is the gold standard -- and it's the method people prefer. 

Another reason to choose the 0-5 star rating is to prevent distortion in the ballot data. It's been found that with larger scales people tend to use certain points on the scale the most by far. The max (5), the min (0), one in from each side (4 and 1) and the middle (2 and 3) are the most used. The 0-5 scale maximizes the likelihood that people will make good use of the full ballot, which improves results. 

The Equal Vote Coalition also looked into how changing the scale for STAR Voting might impact results and the system's accuracy. In 2017, Equal Vote asked Dr Jameson Quinn, then Vice Chair of the Center for Election Science, to simulate STAR Voting with different ballot scales using his "Voter Satisfaction Efficiency" model. The results showed that accuracy did improve compared to smaller scales but that beyond 0-5, the accuracy plateaued. 

Learn more about how to vote in STAR Voting here. 

Sources: 

• George A. Miller (1956) — “The Magical Number Seven, Plus or Minus Two.”
  Origin of the widely cited 7±2 limit on items in immediate memory; commonly invoked when people reason about how many choices or chunks humans can hold at once. UT Psychology Labs

• E. I. G. Sparling (2010) — “Cognitive Load of Rating Scales” (honors thesis / paper)
  Directly compares cognitive load across scale types (unary, binary, five-star, 100-point slider). DigitalCommons

• Papers on rating-system format and user effort — e.g., “Five-star or thumbs up? The influence of rating system types on users’ perceptions, cognitive effort …” (studies comparing rating systems and their cognitive/behavioral effects). These look at how 5-star interfaces affect perceived effort and decisions. ResearchGate

• Qualtrics (2024) — “Handbook of Question Design.” Comprehensive practical guide to writing survey questions and choosing scales Qualtircs