There was a wonderful moment in the BBC drama Last Tango in Halifax when the grandmother, Celia, attempted to explain to a policeman how she was related to a mixed-race baby.

Confronting a predicament many grandmothers might face in future, she explained her daughter was the wife of the baby's deceased mother and that the father was a donor.

This was rich human material. Viewers like me were amused and delighted by the philosophical acceptance of grandparents who, with difficulty, took all of these in their stride: gay marriage, infidelity, promiscuity, divorce, tragedy and unconventional pregnancies. It made the vagaries of our own families seem tame by comparison. And it demonstrated the pointlessness of being judgmental.

I found it salutary to bear in mind that, in my lifetime, any one of the misdemeanours could have cost the perpetrator their place in society.

We should be grateful to live in so liberal an age - at a time when almost anything goes so long as it is between consenting adults. It has brought sanity. It offers people respect, equality.

But Last Tango championed something else. It was a paeon to the family. Its characters survived, licked their wounds and regrouped precisely because, despite everything, the institution held. There is a lesson in that too.

This week the long-term survival of marriages is in the news. A report has been published that will make uncomfortable reading for some. It says that couples are much more likely to break up if they have a child before they marry than if they wait.

It's not a survey from the 1950s. It's not the teaching of any church. It's the latest research.

It'll cast a small shadow on the tens of thousands who have pressed their eldest child into service as a flower girl or page at their own wedding. But all is not lost. Their chances of a lasting relationship are still a good deal higher than for those who remain cohabiting and never wed at all.

The figures are stark. Three quarters of couples who marry first and then have a child will still be together 15 years later. That figure drops to under a half for couples who have a baby first. Only one third of cohabitees will still be together by the time their first born reaches their mid-teens.

The research was conducted by Harry Benson of the Marriage Foundation with analysis from Professor Stephen McKay of the University of Lincoln using data from Understanding Society (a study that surveys 40,000 British households).It looked at 1,800 mothers with a child aged 14 or 15. The results revealed that the timing of the first birth had much greater statistical significance on the survival of a marriage than age, education or social class.

I find it both enlightening and mildly dispiriting because I like the patchwork quilt of partnership arrangements that we now regard as normal. What the statistics can't reveal is why it should be this way. But facts are facts.

In my youth a child born out of wedlock was a social disaster. Brides walked up the aisle with large bouquets if there was a bump to camouflage. Rather than weather the shame of an unmarried daughter, many families insisted on adoption or, after the 1967 Act, abortion.

There were homes for unmarried mothers, adoption agencies and the now infamous Magdalene laundries. It was a system of moral dictatorship that spawned an industry designed for the humiliation and heartbreak of young women who had sex before marriage. Thousands never fully recovered from the loss of their baby. It was wicked.

Fortunately we live in a more compassionate age. Come what may, the newly released statistics mustn't be used to turn back that aspect of the clock.

But they are worth looking at with an open mind precisely because they are factual and not the product of a moral crusade of some kind or a belief system like a religion.

It could be argued that we have recently been living through a social experiment. Children are being born and brought up in all manner of arrangements. We are now in a position to measure what the comparative effect has been.

Instead of looking at what is acceptable or what answers to this teaching or that, we can simply measure what works best.

And whether we like it or not - and however many exceptions there are to the rule - we can see that children (and men and women) fare better in stable families.

There is a plethora of studies that show married men are happier and healthier than bachelors. This holds true even in a mediocre marriage. Married men earn more, work more and spend less time with friends. They then enjoy the physical benefits of a balanced lifestyle.

It's harder to find evidence that marriage benefits women physically. However, women flourish in good emotional relationships and are helped by the ability to share parenting.

For children, the evidence is clear cut. Fathers are important and their constant and unquestioned presence is a boon. As one study put it: "Fathers don't mother." They offer protection, a firmer discipline, vigorous play and the encouragement to take risks. There is evidence that an adolescent from an intact family, who has a good relationship with their father, is half as likely to be delinquent as a child from a broken home. Girls are one third as likely to get pregnant and both sexes are a third as likely to get depressed.

Figures like these demonstrate the importance of maintaining a family structure. It's figures like these which -in our secular age- may prove more compelling an argument for marriage than the pulpit thumping and moral dictates of former generations.

The breakdown of marriage and the social ills that fall from it are reckoned to cost the state £47 billion a year. But worse is the emotional cost. If we can avoid it, we should.

I'm not suggesting that we force ourselves back to lives of quiet desperation, such as some of our parents may have endured. I'm not suggesting we draw punitive boundaries around human frailty. I am however suggesting that we can inform ourselves of the odds and place our bets accordingly.

If cold hard statistics demonstrate that not getting pregnant until after we get married will offer the child a higher chance of being raised with two parents in a stable family, wouldn't waiting make sense?

After all, we have had enough time to weigh and measure different approaches to the rearing of children and it seems clear that marriage is the best and marriage before children best of all.

I'm not being sentimental about it. Marriage was, after all, a societal arrangement before it was a romantic bond. Since the Victorian era the pendulum swung the other way. Now the evidence is clear that, for the good of all, we need to use our heads as well as our hearts.