A friend recently had a medical condition that required EMT (Emergency Medical Technician) response. The lead person asked my friend, who was flat on his back, “How much pain are you in where one means no pain and 10 is the worst pain you’ve ever experienced.” That oral presentation was based upon the commonly used Visual Analog Scale (VAS) for pain assessment.
We’ll look at the evolution of the Visual Analog Pain Scale in the context of how measurement scales are presented to the respondent.
How to Present the Survey Scale
In the surveying process, we typically use lots of scalar questions. You know, those 1-to-10 questions… We hope that they are interval rating scale questions, but that’s a separate discussion. At worst, they’re ordinal scale questions.
The survey designer must decide how to present the scale to the respondent – a critical survey design decision.
The surveyor has four different options:
Verbally:with words, whether spoken or written that describe some ascending order, such as Not Satisfied to Very Satisfied. (In the vernacular, “verbal” means “spoken” but properly it means use of words.)
Numerically:with a numbered scale, such as a 1-to-5 or 1-to-10 scale.
Ideographically:with pictures, such as icons or emojis that display increasing emotions.
Visually:with a continuous line, marked with some low-to-high descriptors.
The respondent is asked to identify the place on the scale that best matches their views. The choice is coded with a number to measure whatever we’re trying to understand – the phenomenon of interest – such as degree of satisfaction. The response could be coded with the verbal descriptors, but text is a hassle to analyze.
All four of these presentation methods can be administered on paper or video screen, but only the Verbal and Numerical methods can be presented orally — in a telephone or personal interview.
Combination Survey Scales
Commonly, we use a combination of scale presentations. We’ve all seen or heard the verbal plus numeric scale presentation. (“On a scale from 1 to 10 where 1 represents Not Satisfied and 10 represents Very Satisfied…”) It’s probably the most common scale presentation approach.
In fact, numeric and visual scales must be combined with words or pictures that label or anchor the scale for the respondent. Numbers and lines have no inherent high-to-low designation.
You might think that numbers do run low-to-high, but whether a 1 is high or low is driven by context and cultural factors. Ever hear fans at a sporting event cheer: “We’re number 10!”? 10 can be good or bad depending upon the context. Thus, a verbal label is needed.
The Visual Analog Scale (VAS) is one such combination scale presentation, developed by Hayes and Patterson in 1921. It was created for use in hospitals where pain levels need to be assessed quickly with patients who might not understand the local language.
Here’s the concept. The patient is shown a sheet of paper with a 100 mm horizontal line where one end of the line represents no pain and the other end extreme pain. Ask the patient to indicate how much pain they’re feeling by marking a place along the line. The distance to the mark is measured, and voila, we have a measurement on a 0 to 100 scale.
Why is this “visual”? Visual is an odd term here. “Linear” might be better. Arguably, ideographs are “visual.” The line as the “visual” representation of the scale makes this, well, visual. On a webform, it would be a slider.
What makes this “analog”? Analog means “continuous” – as opposed to “discrete”. Verbal, numerical, and ideographic scale representations have distinct, individual points to the scale. For example: 1, 2, 3, 4 and 5. That makes these scale types discrete. (And I don’t mean “discreet”!) The longest discrete scale commonly used is a 0-to-10 scale – all of 11 points! — for the Net Promoter Score measurement.
A line (or slider) technically has an infinite number of points on it. That’s continuous or analog.
Why use an analog or visual scale? The argument for a visual scale: it’s more precise. Think about final exams in school. You were graded on a 100-point scale. An 89 (B+) was different from an 91 (A-). So, the 100-point scale was a more precise reflection of your grasp of the material that a 10-point scale would be.
While an analog scale has an infinite number of points on it, whatever point the respondent chooses is usually coded as a 0-to-100 scale. While we could be more precise and measured to the 10th of a millimeter, at some point we simply have “false precision.
The Evolution of the Visual Analog Survey Scale
Now, look at the commonly-used example of the VAS shown nearby. You might be thinking, “I see a line divided into 10 segments with hashmarks, coupled with smiley faces and words. How does that jibe with your description?”
The VAS has been “improved” over the years – or bastardized if you prefer – to the point that “Visual Analog” is a truly misleading descriptor.
Is it Analog? The hash or tick marks along the horizontal line make the scale discrete rather than analog. Even though someone could mark a point anywhere along the line, research has shown that the presence of the marks leads respondents to select near those marks even with no instructions to do so.
Is it an Interval Scale? For a scale to be interval, there must be a consistent unit of measurement, that is, the difference between each pair of points on the scale must be the same. (Think of a ruler.) For the 1st VAS example above, notice the descriptive terms in used for the high-pain end of the scale: Dreadful, Horrible, and Agonizing. Do those described equal intervals? Not in my world. They’re synonyms to me.
Is it Visual? Yes, it has the visual component of the horizontal line, albeit with hash marks, but the line has been “augmented” with verbal, numeric, and ideographic labeling.
So, it’s proper name is the Visual Verbal Numeric Ideographic Discrete Scale (VVNIDS).
Now that’s a mouthful! But it is a better description of this measurement instrument. The current name distorts the meaning of Visual and Analog.
Survey Sample Size Calculator
Get our Excel-based calculator. It can also be used to gauge statistical accuracy after the survey has been completed.