Statistics and choices

By now the news has made the rounds: According to a recent study, over a third of secondary schoolchildren live with just one biological parent. That places Malta at the top of a league table of 41 countries, which includes the usual suspects (in Maltese public perception anyway) when it comes to "weak family values". Even after adding stepfamilies (children living with a remarried parent) to the one-biological-parent category, Malta still hovers right at the top.

The discussion, online and offline, generated by a report that appeared on The Times on June 18, has not generated a lot of light. Statistics are a way of expressing relationships. It is difficult to judge the reliability of an isolated statistic, let alone its wider significance. A lot of what has actually been said is nonsense.

Nonetheless, even the nonsense uttered has been useful. The news story has shown various things. First, even one statistic can change perceptions. Second, perceptions are changed because our arguments are actually based on presumed (however imaginary) statistics or relationships.

Third, the reason why one statistic has such an impact has to do with the dearth of accessible family-related statistics to do with contemporary Maltese family. Much of the local debate about family-related issues is fuelled by statistics to do with other countries. The US and UK feature heavily, largely because, as far as I can gather, the commentators are more comfortable with English-language studies than for intrinsically sociological reasons.

Fourth, the single dangling statistic exposes how little we know about family relations in Malta. This recent study has generated a discussion about its reliability because there is little else to correlate it to. It has also generated a rumour mill about unofficial knowledge: which unpublished statistics kept by other institutions it may actually tally with. In contrast, most other countries are in a position to weigh their statistics against a finer-grained background of other studies.

Fifth, the online discussion has shown up the startling statistical illiteracy of some regular commentators on family matters. It is not news; statistical correlation and causality (sounds heavy but it is an elementary distinction in the area) are routinely confused by one commentator. But it always helps when confusion is exposed.

Some of the commentary about the report has immediately drawn conclusions about the need to introduce divorce legislation. Such legislation would, of course, do nothing to reduce the rate of children living with one biological parent, although it might increase it. But there is a connecting logic: Many of the arguments against the introduction of divorce focus on the negative impact on children; some of the arguments assume that divorce would shift Maltese family arrangements from a nearly "pristine" state to one that resembles "European" instability; discovering that we actually top the European league table (assuming the statistic is reliable) removes that central plank of the argument against divorce.

That is a heavy burden to put on one statistic but it does show what difference information makes to how one approaches the question of divorce. This space has, for years, been arguing that we need more data to make an informed decision about divorce. However, it has sometimes been mistaken for an argument that is actually trying to stall the introduction of divorce legislation.

I suspect there are three reasons for pro-divorce campaigners' suspicion of the argument. The call for more information can be confused with a call for systematic academic knowledge, which notoriously requires many years of collaborative research. Second, saying "Let us first... have some information" can be confused with another argument: "Let us first... do all we can to strengthen the family" - a task that is never finished, if only because we will never fully agree on what it means. Finally, calling for more information sounds like one is changing a political question - to do with public policy choices - into a technocratic question to be resolved by objective experts.

Hopefully, one good thing to come out of the news story in question will be that such confusion is greatly reduced. The fuss about a single statistic shows how little we know and what a great difference a little more knowledge would make.

It should not take much to find it out. There is much that can be extrapolated from surveys that have already been conducted.

Such knowledge would be politically neutral - in the sense that it would inform both pro and anti divorce sentiments. At the same time, such information would be essential for the political decisions that need taking; by political I mean not "partisan" but questions to do with equality and protection of the weak - in other words, what provisions a divorce law, if it is introduced, would make.

Let us remember, no divorce law ever simply states: You may divorce. In the 1986 Irish referendum, the no vote won on the strength of pro-divorce feminist organisations urging a no vote to the particular package proposed.

ranierfsadni@europe.com