To ensure surveys across the State of Iowa are clear, consistent, and easy to interpret, this page provides a reference guide to commonly used rating scale labels. These recommended scales are designed to improve the quality of survey data and make results easier to communicateâwhether you're measuring satisfaction, agreement, likelihood, or frequency.
The table below outlines examples of fully labeled 5-point scales, along with guidance on how to apply Top Two Box scoring, which highlights the percentage of respondents selecting the most favorable responses (e.g., âExcellentâ or âVery Goodâ). Youâll also find best practices for using text-based labels instead of numbers, and when itâs appropriate to include options like âDonât Knowâ or âN/A.â
These practices are not mandatory but are strongly recommended to help you design surveys that are accessible, meaningful, and aligned with Iowaâs commitment to quality public service.
Understanding Top Two Box Scoring in Surveys
Scores can range from 0 to 100 and measure the % of customers who evaluated their experience as Excellent or Very Good. Other scale points are also available and can be displayed online or in reports. A score of about 70% is considered positive and above 80%, best-in-class. Scores less than 60% are considered weak.
Why textual labels instead of numeric?
- Preference: Respondents prefer to give word ratings than numeric ratings
- Communication: Textual labels allow for easier communication of results again. It enables us to say â80% of customers âStrongly Agreeâ or âSomewhat Agreeâ that Company A is trustworthyâ rather than â80% of customers said â5â or â4â on âCompany A is a company I can trustââ
Should I include a "Donât Know" or "N/A" option in a rating scale?
- Itâs not recommended for any question except when asking about very specific experiences (see below). Better to let them skip the question.
- Relational surveys: There is rarely if ever a logical, rational, emotional reason to have N/A included on the NPS behavioral question. If a customer has interacted in any way, no matter the length of time, they have developed or have begun to develop an attitude toward the organization. They can reasonably answer the question. Better to have them skip the question.
- Touchpoint surveys: However, it may be more acceptable to ask in touchpoint.
- Example: If they are vegans and never visit the meat department, N/A is appropriate for the touchpoint.
TOP TWO BOX |
| BOTTOM TWO BOX | |||
---|---|---|---|---|---|
5 | 4 | 3 | 2 | 1 | |
QUALITY/ EXPERIENCE | Excellent | Very Good | Good | Fair | Poor |
LIKELIHOOD SCALE | Extremely likely | Very likely | Somewhat likely | Not Very likely | Not at all likely |
CUSTOM TO CONCEPT | Extremely <XXX> | Very <XXX> | Somewhat <XXX> | Not very <XXX> | Not at all <XXX> |
XXX insert is based off of question text, ,i.e. easy, concerned, competent, effective, well, trustworthy
| |||||
AGREEMENT - BIPOLAR | Strongly agree | Agree | Neither agree nor disagree | Disagree | Strongly disagree |
SATISFACTION BIPOLAR | Very satisfied | Somewhat satisfied | Neither satisfied nor dissatisfied | Somewhat dissatisfied | Very dissatisfied |
SATISFACTION UNIPOLOAR | Extremely satisfied | Very satisfied | Somewhat satisfied | Not very satisfied | Not at all satisfied |
COMPETITIVE RELATIVE | Much better | Somewhat better | About the same | Somewhat worse | Much worse |
FREQUENCY | Always | Most of the time | About half of the time | Sometimes | Never |