Table column heads...?

dtw's picture

Here’s a matter of style. And while I welcome opinions, I’d also like to know if there’s an authoritative, “definitive” ruling anywhere!

Say you have a table where some columns have a simple column head, and other columns are grouped and have span-head with subheads underneath. Like, for example, this:

Now, should that single-column header be aligned with the top row (as per this example) or with the bottom row? My manager (and, it seems, our typesetters) think the former. Many of our authors seem to think the latter. I can see both sides of the argument: aligned at the top seems to make structural/semantic sense, but aligned at the bottom keeps the heading near to the data it relates to.

Example tables in Chicago and Oxford style guides seem to side with our authors, but they don’t explicitly talk about this issue.

riccard0's picture

In such a case, I would center the values (and their headers) too.

dtw's picture

Vertically? So in this case having "Knowledge resource" on the same line as "Uniform", you mean?

riccard0's picture

No, horizontally: Adaptor, Innovator, etc. centered with High, Low, etc.

Vertically, Knowledge resource would maybe better aligned to the bottom.

dtw's picture

Okay; today I’m not concerning myself with horizontal alignments - that’s a whole separate kettle of fish! So, vertically, you’re agreeing with our authors and the style-guide examples.

riccard0's picture

Oh, sorry, I misread your question.
But, yes, I'd say bottom, because it relates directly with the column's content.
(in a different design, maybe vertically centered could work too)

Joshua Langman's picture

I would say bottom row, for the reason you suggest. Think of that heading not as a major heading with no subheads, but as a subhead with no overarching categories it belongs to. Imagine your levels of headings building out from bottom to top, not vice versa.

dtw's picture

That's two votes for bottom row.
Anyone else?
(Plus, anyone have any nice, authoritative reference books that rule on the subject?) :^)

JamesM's picture

I would say top aligned, to reflect that they are equal in hierarchy. But I don't think there's any right or wrong answer, it's mostly personal preference.

Theunis de Jong's picture

I find myself in agreement with James. It's about hierarchy -- the right hand side may have three subdivisions or it may have a dozen, but it should not take the other major heading at its left down to its own level.

Nick Shinn's picture

I’m uncomfortable with the italics.
Italics are a secondary style, so they should not be first in a layout.
Bold, caps, or small caps would be better.

It doesn’t hurt to have vertical rules and/or tonal areas for different cells: asymmetric minimalism is a bit vague for tabular layouts, which seems to be the source of your problem.

Joshua Langman's picture

So, your votes so far are two vs. two.

JamesM's picture

Nick's suggestion of vertical rules is a good idea. They aren't always necessary but in this case I think they may help.

dtw's picture

Hee hee, thanks all, and I do appreciate the inputs on other aspects of the design (not that I’m in a position to implement them); Nick I quite agree that if it were down to me the column heads would be bold, not italic, and italic would then be for headings in the left column that “group” rows of related data, in such tables as warrant them.

Yep, so it’s looking like a fairly even spread of voices, and therefore down to personal preference. :^/

charles ellertson's picture

In this case, align on the bottom row. See Glossary of Typesetting Terms (Chicago), the appendix on tables.

Each "row" (level of head) signifies what the head ranges over. In the bottom "row" (level of head), the head ranges over the data (or stub) of individual columns. If you had a squib line, that would introduce a bit different complication.

Some people view tables as being an exercise in graphic design, some of us as a way to present data. They can do both, but one is primary. Guess which one I pick.

Nick Shinn's picture

Some people view tables as being an exercise in graphic design, some of us as a way to present data.

Uh, what’s the diff?

charles ellertson's picture

Uh, what’s the diff?

Good one. Ideally, there isn't any.

When I go to the book show at the AAUP national meeting, I always spend more time with the rejects than with the selected books. You can learn a lot about the jurors that way. What you learn isn't always favorable. In a way, a good show is one where the "not selected" books do have some flaws, or at least, in some way aren't quite as good as those selected. There is always a judgement aspect, but I'm not talking about that.

Be interesting to hear Kent Lew's reaction to being a juror at AAUP.

A long time ago, I was a juror -- a smaller show, the Rocky Mountain Book Show. Some of the hardest work I ever did. The other (interior) juror was Rich Hendel. I'm proud that we almost completely saw eye to eye.

dtw's picture

Thank Charles for being the first to find some citable source!

I had hoped that either Mitchell & Wightman or Bringhurst would have something to say, but they both treat tables very briefly in a couple of pages without going into any detail.

Syndicate content Syndicate content