Tuesday, April 16, 2013

Academia, half-truths and political bias

Should New Zealand Universities be trusted disseminators of balanced information?

The following is a Facts Sheet produced by the University of Waikato. I assume it is intended for use by students and other interested parties. It is available on-line



 The first fact:

Sole Parents
   At the 2006 Census, the median age of New Zealand’s Sole Parents was 42.9
years.


According to MSD, also using 2006 Census data:
Sole mothers tend to be younger than partnered mothers, with a median age of 37 years in 2006, compared to 39 years for mothers in two-parent families. Conversely, sole fathers tend to be a little older than partnered fathers, with a median age of 42 years in 2006, compared to 41 years for fathers in two-parent families.

Why the discrepancy? The MSD report relates to sole parents with dependent children.

It is unclear what definition of sole parent the University has used but later in the sheet we see this claim:

 At the 2006 Census, two thirds of New Zealand’s Sole Parents had been
previously married or in a civil union ....   13 per cent were widowed (median age 63 years)

So not sole parents with dependent children.

Next though is a blatant omission of data resulting in a highly misleading picture:

 The majority of ex-nuptial births are to women aged 30-34 years, followed by
women aged 25-29 years. Rates are lower than they were in 1972 for every
age group except 35-39 and 40-44 years, for whom ex-nuptial births have
increased substantially. 

This leaves the average person with the impression that fewer children are born outside marriage today than was the case in 1972.

Following are the statistics  required to provide a comprehensive understanding:

In 1966 11.56 percent of all births were ex-nuptial
In 2012 47.66 percent of all births were ex-nuptial

So why two seemingly contradictory 'facts'?

Because in 1966 there were far fewer unmarried women.

Take one of the age groups whose rate of ex-nuptial child-bearing has decreased, 25-29 year-olds.

If, in 1966, the ex-nuptial birth-rate was 100 (meaning 100 births per 1,000 unmarried females aged 25-29), and  there were 9,746 unmarried (or never married) females, that would result in 974 births.

If the nuptial birth-rate was also 100 for married females, who numbered 70,726, that would result in 7,072 births.

So the rate of ex-nuptial and nuptial births were the same but ex-nuptial births represented only 12% of all births.

No effort has been made to explain this to a student via this fact sheet.

It's an appalling ommission.

The facts as they are presented on this sheet paint a picture of most sole parents having been married, and ex-nuptial child-bearing on the decline. It is designed to elicit sympathy and political support for sole parents.

(It did occur to me to look for a further sheet but none exists - or not on-line. But this presentation provides a clue to the professor's political leanings where she describes the 2011 budget as being an Oliver Twist budget and calls for more investment in children. Unsurprisingly it's hosted at the Child Poverty Action Group website which brings me full circle. It was a paper by their supervised young scholar that led me to the fact sheet in the first place.)


2 comments:

Anonymous said...

I wouldn't trust ANY Western university in the world when it comes to information on either social science (including welfare), economics or climate science.
Those areas have been completely taken over by left-wing ideologues who want to push their agendas.

Pure sciences like chemistry, physics or mathematics are probably ok.

Psycho Milt said...

"Fact" sheets like this are an expected outcome of the social sciences. There are two reasons:

1. In the sciences, you have a hypothesis and set up experiments to test it. In the social sciences, you have a political agenda and look for statistical data that supports it.

2. A well-designed experiment does actually test the hypothesis. Well-designed statistical data can be manipulated to show whatever you want - hence "lies, damned lies and statistics."