Monday, August 05, 2013

More inconvenient truths CPAG overlook

Last week the Child Poverty Action Group released a report into the link between substantiated child abuse and socio-economic deprivation, ethnicity, benefit income and the youth population.

I've blogged about the shortcoming in their methodology to conclude there is "no correlation between benefit receipt and child maltreatment."

So let's move onto their next finding, p.2:

"...there is an association between ethnicity and child maltreatment, however given the strong association between ethnicity and socio-economic disadvantage in New Zealand this finding needs to be treated with caution."

On p.13 the following "discussion" appears:

Disproportionately high rates of child abuse among Māori need to be treated with caution: the ethnicity given is that of the child; New Zealand data shows a strong link between socio economic deprivation and ethnicity (Perry, 2009, 2012). Data from Perry (2012, p. 76) shows that between 2009 and 2001 the median income of Māori families fell 1.1% ($26,300 to $26,000), the median income of Pacific households fell an astonishing 5.3% ($28,300 to $26,800) while that of Europeans – starting from a much higher base – fell 0.8% ($35,500 to $35,200).

Here's the relevant table (note their typo - 2001 should be 2011):

The updated 2013 report shows that between 2011 and 2012 Maori income increased to $30,000 and Pacific income increased to $29,800 (Perry, 2013, p.80).

That in itself is inconvenient. Perry notes:

From a longer-term perspective, all groups showed a strong rise from the low point in the mid 1990s through to 2010.  In real terms, overall median household income rose 47% from 1994 to 2010: for Maori, the rise was even stronger at 68%, and for Pacific, 77%. 

But the issue I really want to focus on is this.

CPAG got ethnicity breakdowns on the substantiated findings of abuse across CYF site offices.

They provide a comparison in their appendix:

Note the bottom line which shows that, on average, substantiated abuse findings are exactly the same for Pacific when compared to the NZ European baseline.

CPAG describes this table thus (p.12):

Raw data for all substantiations (not distinct cases) shows nationally Maori children are more than twice as likely to suffer abuse as Europeans (abuse rates for each ethnic group by site office and country overall are listed in Table 7).

So if "a strong link between socio economic deprivation and ethnicity" is the reason Maori children are over-represented in abuse statistics, why isn't it operating amongst Pacific and other ethnicities?

This is a fly in the ointment for CPAG. For them poverty has to be associated with  child abuse because it provides support for their main campaign which is to increase benefits.

The report ends with this:

Rates of child abuse in a society are not pre-determined, nor do they remain static. New Zealanders’ rates of child abuse have increased over time: they can change for the better if we so choose. Reducing the risks associated with poverty would be a good place to start.
So "rates of child abuse have increased over time" yet Maori households, where a disproportionate amount of the abuse occurs, have incomes rising faster than the median.

Their position is implausible.

1 comment:

Brendan said...


The notes on the 'abuse rates' table says that each office does not take account of the ethnic composition within the site offices. Neither does it take into account (I suspect) the total population being serviced by that office, which adds another level of distortion. For example, is the Papanui Office serving twice the number of people as Otara? If so, then giving each office's results the same 'weighing' when deriving an average creates a distortion.

Regardless, the table implies that for every one European child abused, there are 2.6 Maori children abused. If Maori represent (say) 15% of the population, then are Maori children 17 times more likely to be abused that European children?

Or am I missing something?