More crap research on smacking attitudes

An interesting article from the Social Policy Journal was placed on the MSD website recently. It was called Just who do we Think Children are? New Zealanders' Attitudes about Children, Childhood and Parenting: An Analysis of Submissions on the Bill To Repeal Section 59 of the Crimes Act 1961.

The project - funded by a well-known anti smacking organisation that gets 40% of its income from the taxpayer - looked at the submissions to the select committee on the Section 59 legislation. Researchers examined two particular contrasting social viewpoints of children - children as "human beings" and as "human becomings" - and whether these two viewpoints were implicated in people's views on the use of physical punishment.

It boded the hypothesis that people who advocate the use of physical punishment are likely to conceptualise childhood as a phase of development, where the child is on his/her way to becoming an adult; in other words, children are "human becomings". These people are also likely to support the use of physical punishment. The researchers also hypothesised that people who see childhood as a complete state in its own right - "human beings" - are more likely to reject physical punishment. The hypotheses were supported. That doesn't mean it was proved, just supported.

There was no mention of the fact that children are becoming adults - as opposed to "human becomings", whatever that is - and are at varying stages of development.

But they ditched nearly half the sample because they did not fit in with the framework to address the hypotheses.

How can you, with validity, support, prove or disprove a hypothesis based on a sample from a population of fixed views on the wider topic where you disregard nearly half the sample in making a decision on that hypothesis AND make a reliable research-based conclusion attributable to the wider population that had nothing do to with the hypotheses. Because that's exactly what these researchers did.

Namely, if children continue to be viewed as in a state of becoming, they may be more vulnerable to abuse.

Given that the process oversampled some groups and undersampled others, if we then were to extrapolate from these populations using this sample to make claims about the pattern of support for the legislation nationwide, it would be problematical.

Anyway, I asked one of them how the research finding directly relate to the hypothesis raised.

"It doesn't", she said. Before I was able to ask why the hypotheses were raised and the findings were reported the way they were she said, "I have to go now".

And she did. Just who do researchers think they are? It looked like the "conclusions" to the report were written before the hypotheses. Perhaps that is why the hypotheses weren't even tested in a way that provides reliability and validity throughout the wider population.