Author's Toolkit, Vol. 28, No. 4

電子情報通信学会の季刊誌「情報・システムソサイエティ誌」に掲載
Authorʼs Toolkit by Ron Read, Osaka Branch Manager
Published in IEICE INFORMATION AND SYSTEMS SOCIETY JOURNAL
Vol. 28, No. 4 ©2023 IEICE

In technical writing, comparing different things, actions, or qualities is a common task. To compare clearly and correctly, you need to understand what you are actually comparing.

Comparing with Precision

Most comparisons made in a technical paper are fairly straightforward, easy to write and easy to understand:
〇 The synthesized material is several times more conductive than base metals.
〇 Automatic categorization performed only slightly faster than the conventional method.

However, sometimes we’re not directly comparing things but some attribute of those things, which requires special expressions:
X The temperature dependence of our method is 23% lower than the conventional method.
〇 The temperature dependence of our method is 23% lower than that of the conventional method.

It’s easy to see the problem with the first example: We are not comparing “temperature dependence” with a “method.” In the second example, we properly use the pronoun “that” to represent “temperature dependence” while also keeping the expression non-repetitive.

This handy writing tool for comparisons also has a plural version:
〇 User surveys conducted in summer have a higher response rate than those in winter.

In some cases, writers use “than that of” incorrectly, or at least unnecessarily:
X The problem of delayed activation, in some cases, is worse than that of failed activation.

This is a subtle distinction, but in this context, we know that “failed activation” is itself a problem, so saying “the problem of failed activation” is logically possible but in a real-world sense clearly redundant. We just mean:
〇 The problem of delayed activation, in some cases, is worse than failed activation.

This revision is clearly a “problem-to-problem” comparison, and the sentence has more impact. But if you want to avoid such subtleties, you can of course just recast the sentence:
X In some cases, a delayed activation is a more serious problem than a failed activation.

When you have a long series of attribute comparisons, such as reporting results, it’s usually acceptable to neglect the verbal references to the attributes after the precise comparison has been once established:
〇 Most noticeably, the frog cue had a 33.4% higher recognition accuracy than that of the bird cue; furthermore, snake was higher than mouse, dog was higher than kangaroo, and cat was the second-highest visual cue.

When making a comparison, always keep in mind what is being compared to what.

Don’t Use Ghost Numbers

In your paper’s Discussion or Conclusion, it’s often useful to focus on a limited number of findings, advantages, problems, etc.:
〇 Our experimental results identified three key advantages to our method. First…

A problem may arise when you get to that second advantage:
± Second, our method was able to…

This may be fine if it comes only a few sentences after the first advantage. But I’ve seen many cases where it’s written a large paragraph or more away. In this case, readers are likely to forget what “Second” refers to. Remind them:
〇 As its second advantage, our method was…