In discussing the Evidence2Success survey of pupils, it is worth pointing out that children often understand a lot more than we give them credit for.

In general, young people pick up on a lot more about the world around them than many adults realise.

That insight may be one reason why the Scottish Government has backed the work done in partnership with Perth and Kinross Council by the Dartington Social Research Unit and why the survey is likely to be rolled out in other council areas. If we want to know what is going on in the lives of schoolchildren, the best way to find out is to ask.

Nevertheless, it is reasonable to question whether it is appropriate to inquire of primary pupils whether "sometimes I feel that life is not worth it"; likewise, whether they or their friends have tried to steal a car within the last 12 months or "attacked someone with the idea of seriously hurting them".

The response from the researchers is that such information is extremely useful. They say it helps councils, health services and charities make better use of public money. They can understand children's needs and respond in a more targeted way, intervene earlier and plan better.

Perth and Kinross says the survey, which it piloted last year, was voluntary and anonymous, and an attempt to engage with the community to plan services. There is nothing wrong with councils seeking the best information possible to help officials do their jobs better. But it is slightly misleading to suggest that the controversial questions were only a small part of last year's 24 page survey. In fact, it covered legal and illegal drug and alcohol use, domestic disturbances, parental approaches to discipline, weight problems, theft and weapon-carrying.

If such research in schools can be used to improve health interventions, identify bullying or flag up domestic violence, it has the potential to be useful. But there are caveats.

The information must be genuinely representative and reliable. School pupils have a tendency to exaggerate in response to such questionnaires and controls need to be in place. Already some of the findings have wildly contradicted other information available about a given school.

Opting-out must also be a realistic choice. With the initial survey, both parents and pupils had the chance to decide not to participate, but information provided was limited.

As other councils begin to use the survey, changes are planned. Pupils will still have to opt-out rather than opt in. This makes sense. If the survey participants are self-selecting, it is likely to become worthless as a source of information.

But researchers now acknowledge too much priority was given to keeping communications with parents brief, rather than informing them of the detail of the exercise.

Better information for parents will make opt-out meaningful. Meanwhile the removal of the most intrusive questions, about sexual behaviour and risky practices, until alternatives can be considered is also the right decision and a welcome change.