A most interesting proposal

Fiction editing is great for exploring subtle shades of meaning, of controlling how ideas are placed in a reader’s mind, and perhaps opening up whole new worlds. Although there is a definite structure, deviations from the main trip are delightful, if skilfully executed. You want the person to enjoy the journey as much as the destination, to smell the flowers along the way. Corporate, academic or science editing, on the other hand, is about communicating a message efficiently, about making complex ideas accessible (or, some might say, with tongue firmly in cheek, as inaccessible as possible), and letting the reader take the information and do something with it. No flower smelling here, thank you, I’m trying to save the flowers from extinction. Let’s go.

This makes corporate editing a lot of fun in terms of finding out stuff, stuff that I can well envisage one day turning into fiction. For example, some years ago I was writing a brief section of a report that was a demographic study of Australia. As part of the research for that I had to trawl through excel tables from the Australian Bureau of Statistics (which I secretly enjoyed very much), and I noticed that the number of women who said they were married differed from the number of men who said they were married. The statistic I was looking at said that more women than men thought they were married. Who were these women married to? Why were some of the men not declaring their marital status?

I’ve long lost the actual example, so I thought I’d look it up again to see if it was really true, or if I’d made it up. It’s true. And by looking at the historical data from here, I saw a pattern. From 1976 through to 1993 (for the years 1976, then every year from 1981 to 1993) fewer men than women said they were married. Then, between 1993 and 1994, something strange happens. In 1994, and for every year up to 2001, when these historical statistics finish, fewer women than men declare that they are married. (The categories people can choose are ‘married’, ‘never married’, ‘widowed’, ‘divorced’.)

The actual percentage of people who are apparently married to nobody fluctuates from almost -0.4% to +0.4%.

On the left hand side of the graph, the negatives show the number of men apparently not married, although there are women who think they are married to these men, as a percentage of the total number of married men.

On the right hand side of the graph, the positives show the number of women apparently not married, although there are men who think they are married to these women, as a percentage of the total number of married women.

I’m sure Bernard Salt, or in fact any statistician worth their salt (sorry, Bernard) would have a field day with my amateurish number necromancy. But you have to admit, this is fun. It makes me want to find out who these singly married people are and what happened around 1993/94 that made this statistic suddenly switch. It’s probably something quite boring, such as people whose divorces are in train, but not yet final, disagreeing as to whether they are married or divorced at the time of the census. But I would expect if it were that simple then the numbers would just cancel out, or at least the graph would swing between the positive and negative from year to year. Perhaps there was a change in the numbers of women filling in their own census forms, and therefore answering questions differently from how men might have answered them. I don’t know.

But I do know that the science of looking at large numbers reveals truths that we otherwise don’t have access to. In the hands of a skilled communicator, they really come alive. It was seeing Hans Rosling’s TED talk this week that reminded me about the ABS marriage statistics. What this man does with data and cardboard boxes is amazing.