December 7, 2018
For over two years, this blog has covered data topics from the serious (voting) to the somewhat silly (Valentine’s Day), and from the thematic (housing, student populations) to the technical (margins of error, maps). Despite writing about data for 28 blog posts, we have not yet done a blog post on just that: the skill of writing about data.
This isn’t going to be a textbook. We’re just going to go over a few pitfalls and highlights of clearly communicating data in writing, with a focus on writing about comparisons between data points.
Using Data to Make Comparisons
A lot of data analysis is just about making comparisons. For example: The median household income in Area 1 is greater than the median household income in Area 2; the median household income in Area 1 was greater in 2016 than in 2015.
Below are a few common data comparison mistakes to recognize and avoid.
Fewer vs. Less
Both “fewer” and “less” obviously indicate that there is not as much of what you’re talking about as there is of whatever it’s being compared to, but the two words are appropriate for different contexts and are not interchangeable.
Use “fewer” when the statistic you’re referring to can be split into component units.
- Example: There were fewer boxes of stuff in storage at the end of the year than at the beginning of the year.
- Example: This container holds fewer gallons of water than that container.
Use “less” when the statistic you’re referring to cannot be split into component units.
- Example: There was less stuff in storage at the end of the year than at the beginning of the year.
- Example: There is less water in this container than that container.
You can count individual boxes and gallons (e.g., 10 boxes, 20 boxes, 30 boxes; one gallon, two gallons, three gallons; etc.), but you cannot do the same with “stuff” or “water” without assigning units (such as boxes or gallons). So if you’re unsure about which one you need, try to count your subject by units. If you can do that without adding the units yourself, use “fewer.” If you can’t, use “less.”
There is no equivalent rule for “more”; “more” is correct for both contexts.
Percentage Points vs. Percent
The concepts of a difference of percent and a difference of percentage points are also not interchangeable, and misuse of these can lead to some misleading analysis.
When you’re comparing two statistics expressed in percentages, the concept of “percentage points” is used to draw a simple comparison between one statistic and the other.
- Example 1: In 2015, two percent of households in Sample A had a household income of $100,000 or more. In 2016, eight percent of households in Sample A had a household income of $100,000 or more, an increase of six percentage points.
The concept of “percent” is used to describe the percent by which one statistic is different from the other.
- Example 2: In 2015, two percent of households in Sample A had a household income of $100,000 or more. In 2016, eight percent of households in Sample A had a household income of $100,000 or more, an increase of 400 percent.
It’s important to be familiar enough with both concepts to use them correctly and not interchangeably. There’s no rule on where to use one and where to use the other, as long as they’re used correctly. You couldn’t state that, in Example 1, it was an increase of 400 percentage points. You also couldn’t state that, in Example 2, it was an increase of six percent. Both statements would be an incorrect representation of the data. However, as long as you use the phrasing of your choice correctly, either can be appropriate in any context.
This is a fine distinction, but an important one: “increasingly more” is often redundant, and often used incorrectly. If you find yourself about to use this phrase, consider whether it reflects what you’re actually trying to say.
- Example: These examples are getting increasingly complicated.
- Example: These examples are getting more complicated.
These sentences mean exactly what they sound like: that each example is more complicated than previous ones. Many uses of “increasingly more” really just need one word or the other, not both.
- Example: These examples are getting increasingly more complicated.
This sentence not only means that examples are getting more complicated, but also that they’re doing so at a faster and faster rate. It suggests that the difference in complexity between this example and the last one is larger than the difference in complexity between the last example and the one before it.
Let’s put some numbers to this.
Table: Snowfall, Year 1 - Year 4 (Inches)Download table data for Snowfall, Year 1 - Year 4 (Inches).
Here, it would be correct to say that, in the Year 1 – Year 4 timeframe, winters became increasingly snowy (or just that they became snowier). There was more snow in Year 2 than Year 1, more snow in Year 3 than Year 2, and more snow in Year 4 than Year 3. But the difference in inches of snow from year to year varied across the time period (1.00 inch, 0.25 inches, 0.75 inches).
Table: Snowfall, Year 5 - Year 8 (Inches)Download table data for Snowfall, Year 5 - Year 8 (Inches).
In this table, winters became increasingly snowy, but they also became increasingly snowier. There was more snow each year than the previous year, and the difference between each year and the previous year was also greater each time (an increase of 0.25 inches between Year 5 and Year 6, a difference of 0.75 inches between Year 6 and Year 7, a difference of 1.50 inches between Year 7 and Year 8). Not only did the amount of snow increase over the time period, but it increased by larger and larger amounts.
Both phrases have contexts in which they’re correct; it’s important to make sure you’re using the one that’s appropriate to the data being discussed. This will keep your analysis clear and effective.
Writing about data doesn’t need to be intimidating: prioritize clarity and avoid some common errors, and analytical writing will seem like an easier and less daunting task.