Defining Accuracy in Communication – Interpreting Beyond the Chart

She came in the room with great authority. She, a qualified interpreter, introduced herself while I watched, still a student and there only for support. Her signs were clear, crisp, and technically accurate. She conveyed what the medical provider said, navigating terms like tumor, malignancy, and benign. As quickly as she came, she left.

You can be technically right and still miss. My friend looked at me, lost and confused, not from the news, but from a lack of understanding. The words floated past, technical and without meaning.

What did she say?

And so, I was the one to break the news: they were looking for cancer. And, realization landed with a sharp jab.

Our dashboards are rarely this dramatic or life-altering. Our targets are farther away, rarely faces that we see. But, we still hunt for accuracy. Yet accuracy is more than a chart-level decision.

No, bear with me…

As interpreters, we learned numerous signs for concepts, finding 20+ ways to sign love, for example, or playing with idioms. Language is a proxy for our thoughts, which can encompasses all of our senses. The words we select may come pretty close, but they’re not always precise. And in the height of emotions, we may find no words meet our needs.

As interpreters, we debated about the use of words – was this sign better than that for this concept? We kept coming to the same answer no one likes – it depends. It depends on the context, the mood, the other words surrounding it, the place, and speakers themselves. Yes, we had to factor all that in to be accurate.

Accuracy beyond the chart

So, this is where we get to dashboards. We know our chart choices depend on various factors, such as the goal of the dashboard, the type (infographic, dashboard?), the story, the business environment, and the audience. This isn’t much different than interpreting. Except, we rarely consider our charts in relation to the dashboard.

Consider these:

OR

This difference is small, a matter of a minor chart change and some space. In the first, this dual axis is not synchronized, allowing profit to fill the full space. Some people argue this is the right way to show this chart. It emphasizes the how profit and sales dovetail – is the pattern the same? It has an unintended consequence, though, within the dashboard – it forces an extra bit of space at the end. The axes’ numbers are colored to give an idea which belongs where (some people would encourage going so far as to shade the entire area of the axis) and a legend is also needed. Some people might shade the numbers at the side in lieu of a legend. Within the context of the entire dashboard, the more we do, the more this one chart stands apart. Is this chart really worth all that attention? Is it that profound?

The second iteration looks at profit as part of sales. You can still see the dips, but they’re more understated as the axes are synchronized. We can use placement as a legend, telling people bigger things are sales and smaller things are profit. This allows us to recoup some space and explore this chart in relation to other charts.

Does one flow better to you?

Our audience looks at this differently than we do. Charts are highly abstract forms that rely on standard literacy (look ma, I can read!) and moderate numeracy skills (what do these numbers mean to me?). If that’s not enough, they ALSO require people to sift through the visual chaos. Depending on their exposure, this may all be too much. We spend all our time on this, kids. We obsess over nuances and love the complexity, just as writers love playing with words and artists push the medium.

People using this have a whole different approach to reading this.

1. First pass view
2. Digging deeper into parts
3. Using to get info

I’m only going to look at the first part of this. Remind me and I’ll do more.

End User Perspective

Our users make a first pass judgement view. They try to answer the most important question within about 2 seconds: Is this thing worth my time? Just like with meeting people, first impressions matter.

Our unsynchronized view introduces a lot of complexity from the get-go. The jagged lines, parts where profit pops above sales, and dual acts a bit like that breathless person who rushes through hello and proceeds right into a tirade about the party. We have multiple legends, which end users love hate. In short, we’re not even past the top and we already need that martini.

Compare that to this:

When we communicate visually, that first look matters. Our end users immediately start translating clutter and complexity into calculations about how much effort and energy they have to put into this. It better be worth it.

Within the dashboard, I may make choices around accuracy. The chart may be less textbook, because it alone is not the message. Nope, I need the entire dashboard to come together as a unified message. It’s not a collection of charts. It’s a composition using charts as a medium with various other shapes and pieces interspersed to help support its message. Certain pieces are the pauses, the flicks, and the tones that guide us. It’s the charts, the way we link them, and everything that ties it together that comprise a visualization or dashboard. Without this, you’ve just made a pile of charts. People will read a dictionary and they will read a pile of charts. Narrative, tone, and creativity are what make conversations like dashboards worth engaging.

We do this with our spoken communication. We choose softer words, because we realize the harshness or the perception a word may have. We turn sentences around, making them passive if we don’t want to point the finger (“the glass was broken”). Many times, we choose to be less accurate technically to be more clear or heard overall. Or, in the interpreting example, we choose words our audience understands (cancer and not cancer vs malignant and benign) to meet them where they are. You see, accuracy isn’t defined by the interpreter – no, it’s the people getting the message. Even the Americans with Disabilities Act sees it that way. It’s incumbent on us to understand our audience.

Is accuracy the measure of the precision of our charts or the ability for the end user to understand what the chart is saying?