The Dim-Post

August 18, 2013

The grand inquisitor

Filed under: psuedopolitics — danylmc @ 8:49 am

One of the odd aspects of contemporary New Zealand politics is that opposition policies are subject to extraordinary scrutiny – what are the details? how much will it cost? can we see the figures? where will that money come from? – while the actual government can toss up schemes like Paula Bennett’s plan to subject about a tenth of the adult population to a paedophile test and instantly sack anyone who fails it.

Teachers, doctors and any other government employees who fail Social Development Minister Paula Bennett’s tough new child-abuse screening test will be instantly sacked.

The new child-protection laws will trump existing employment legislation, removing the need for bosses to go through a fair process of verbal and written warnings to dump anyone suspected of sexually preying on children.

Screening of all government employees working with children is one of the main planks of Bennett’s incoming child protection regime, unveiled last week.

All staff working with children in schools, hospitals, government agencies and organisations that get government funding must submit to security screening every three years. It is estimated to affect 376,000 people.

What test will the government use? ‘Yet to be revealed.’ What are the false negative and false positive rates? (ie, how many actual paedophiles will it fail to detect and how many non-paedophiles will it falsely identify? That’d be useful to know, right? If you’re screening 376,000 people using a test with a very high confidence interval – say, 99.9% – that means you’re incorrectly identifying 376 people. How long will the screening test take? A day per person? That works out at 376,000 days of police time. Sounds expensive. How invasive will it be? What happens if you refuse to take it? Will the government sack, say, a paediatrician who refuses to take the test because they consider it a breach of their privacy?

 

About these ads

36 Comments »

  1. These are all very interesting theoretical questions, Danyl. But let me ask you one in return: why do you hate our children so?

    Comment by Flashing Light — August 18, 2013 @ 9:17 am

  2. “One of the odd aspects of contemporary New Zealand politics is that opposition policies are subject to extraordinary scrutiny… while the actual government can toss up schemes.”

    +1.

    Although really, when you say ‘opposition’ you mean ‘Labour’. Green policies rarely get gone over in much detail. But broad strokes, yes.

    And you know, while in principle I’m in favour of scrutiny of any policy regardless of where it comes from, I do think that reversing this trend is the central task of Labour. (And, not to bang that drum, is a good example of why a leadership change isn’t going to solve their problems). It’s kind of hard to explain, but I think at least part of it is that National has, since they were in opposition, ‘trained’ the media and public to expect a very low level of policy detail from them. My worry is that Labour will, when they inevitably return to government, match this, leading to a development far more toxic to good governance in the long run than the continued tenure of Key et al.

    Comment by Hugh — August 18, 2013 @ 9:27 am

  3. “while the actual government can toss up schemes”
    Toff tossers?

    Comment by Clunking Fist — August 18, 2013 @ 10:07 am

  4. Politically, it’s quite a beautiful scheme because it appears to address the problem while still leaving you with a paedophilic bogeyman to exploit.

    You’d need to know the ratio of paedophiles:non-paedophiles in the population. Let’s say there’s 1,000 of them in all of New Zealand. Our unbelievably 99.9% accurate test would ID almost all but one of them. That’s our false negative. So, after all of that testing and expense, we’d still be allowed to fear that there was someone out there.

    But we’d only be identifying 375 people as paedophiles inaccurately. So there is an upside.

    Comment by Steve (yeah, that Steve) — August 18, 2013 @ 10:44 am

  5. Hugh, you should actually take a look at the Greens’ discussion documents – they’re the most detailed opposition policy documents you’ll find in NZ, and they put the government’s empty, half-thought out announcements to shame

    Comment by Dean — August 18, 2013 @ 11:31 am

  6. If someone has managed to hide being a sexual predator from their managers, co-workers, parents of children and various government agencies that already exist, how is an extra layer of screening from the police going to help? Child abuse already happens in secret by and large, so I would suspect that the number of false positives/negatives would be far higher than 1 in 1000 as was suggested in the post. Far, far too many innocent people will unfairly lose their jobs because of this hysteria, and the real paedos won’t be particularly bothered.

    Comment by alex — August 18, 2013 @ 11:41 am

  7. Here is what the Q&A sheet on the Government’s proposed vetting processes says will take place:

    - Identity verification – proof people are who they say, including former identities
    - Information requirement – thorough Police, records, history & behaviour checks
    - Risk assessment – judgement based process for interviewing staff
    - Periodic re-assessment – every three years.

    This looks to me to be (1) time consuming (what exactly will “behaviour checks” encompass?); and (2) fraught with the potential for human error (what does a “judgment based process for interviewing staff” entail?).

    Also, have a think about the incentives here. You are a manager of some description in the public sector, responsible for people who work with kids. One of the people who you manage is a bit of an odd-ball … you and others have a feeling about him that he’s “not quite right” in how he deals with kids. Your options are:
    (1) Bury those feelings in the absence of any concrete evidence of wrongdoing and keep him on in his role; or
    (2) On the basis of those feelings, conduct a risk assessment exercise that concludes the person is a potential pedophile and so sackable without any further notice or procedure.

    Then consider the potential downsides to getting this judgment wrong. If you keep him on, and he abuses a child, what happens to you as his manager? If you sack him, and he’s got no recourse under employment law, then what happens to you as his manager?

    Somehow I think the screening process will be a bit less than 99.9% accurate.

    Comment by Andrew Geddis — August 18, 2013 @ 12:56 pm

  8. So this will apply to “organisations that get government funding”.

    Political parties, for example? If they have to stop those yukky photo-ops in primary schools, I’m all for it.

    Comment by sammy 2.0 — August 18, 2013 @ 1:05 pm

  9. But this is what the left wants:

    Comment by OECD rank 22 kiwi — August 18, 2013 @ 1:21 pm

  10. If you sack him, and he’s got no recourse under employment law, then what happens to you as his manager?

    If your department is (coincidentally!) under-going restructure at the time? Well, you get a big bonus for minimising redundancy payments.

    Comment by RJL — August 18, 2013 @ 2:00 pm

  11. Yep. An easy way to fire under performing staff. You might taint their future prospects but would the average boss care. I’ve had a few bosses who wouldn’t give a flying fuck for their employees so… Not my current one though (in case she’s reading)

    Comment by nigelsagentinthefield — August 18, 2013 @ 2:32 pm

  12. +++1 to what Hugh and Steve said. When opposition parties and media hold National to account, the wheels quickly come off Key’s wagon. Labour are poor at this, the Greens better & Winston occassionally adds to the pressure. The media are very poor (with honourable exceptions like John Campbell). It is scary to think Labour expect to get back in power and receive the same lack of scrutiny as National currently get…

    “But we’d only be identifying 375 people as paedophiles inaccurately. So there is an upside.”
    Hehe – yeh, Paula gets a falsely accused paedo to lynch each day Steve, with 10 spare for Satan’s feast day or in case the mob get overkeen and lynch a few too early. ;(

    Dean – Hugh’s comment that “Green policies rarely get gone over in much detail” was critical of the lack of media scrutiny, not critical of the Greens.

    RJL – do you work for MSD policy development? ;)

    Comment by bob — August 18, 2013 @ 2:40 pm

  13. Say goodbye to male applications at teacher’s college.

    Comment by MeToo — August 18, 2013 @ 2:41 pm

  14. @Dean: Yeah, what bob said.

    Comment by Hugh — August 18, 2013 @ 5:53 pm

  15. If there were any even vaguely reliable test for pedophilia then we would already be using it.

    Comment by herr doktor bimler — August 18, 2013 @ 10:38 pm

  16. This is a tricky topic to deal with, because only a really tiny proportion of allegations of CSA (Child Sexual Abuse) are false, as far as anybody can tell.

    But at the same time, I can’t help thinking that in some professions (being a primary or early childhood education teacher leap to mind) being male is so rare, that any male professional is automatically going to be subject ot a much higher level of scrutiny during both vetting and regular assessments.

    The male primary school teachers I know are already pretty paranoid/aware of the risk that they might become subject to a CSA claim. At the same time, most of them are pretty confident that so long as they are careful (never be alone in a room with a child, if a child tries to hug you gently disengage etc, set up firm boundaries around children touching them “No, that’s not ok for teachers etc”) they will come out ok in an investigation. But imagine how frightening to know that you could lose your job and be barred from carrying out your profession for ten years, based on an unproven allegation from any unstable parent who took a particular dislike to you.

    That’s the worst case scenario and sure, it’s probably unlikely – but because there is so little detail in this announcement about what these background checks might mean, you can’t rule it out.

    Comment by Amy — August 18, 2013 @ 10:50 pm

  17. There’s also the question of whether non-offending paedophiles will be targeted. People who are sexually attracted to children but who do not act on their proclivities (a not insignificant minority of the population by some estimates) must not on any account be victimised where there is no evidence of wrongdoing. The fact that in our public discourse ‘paedophile’ is used synonymously with ‘child sex offender’ (and the fact that this is the one sexual minority is is still socially acceptable to loathe) makes me wonder what Paula Bennett has in mind. Clearly the poster known as Herr Doktor Bimler above is one of those who thinks “tests for paedophilia” are the relevant question, and not monitoring of individuals’ behaviour.

    Comment by Higgs Boatswain — August 19, 2013 @ 12:06 am

  18. …this is the one sexual minority is is still socially acceptable to loathe…

    Oh…is necrophilia socially acceptable now?

    Comment by RJL — August 19, 2013 @ 12:51 pm

  19. There is no way the PPTA/NZEI/PSA will stand for this. Bennet’s buying a fight with the unions (esp teacher’s unions) and it’s possible it might work to her advantage, but on the other hand it’s been a general rule that when this gov’t gets into a fight with the education unions it loses.

    Comment by Keir — August 19, 2013 @ 1:04 pm

  20. Will charter school teachers be government employees?

    Comment by Adrian — August 19, 2013 @ 1:34 pm

  21. I’ve got a friend that has joined the Defence establishment and is still waiting for security clearance a few months after being hired.

    Comment by Owen — August 19, 2013 @ 1:47 pm

  22. I’ve got a friend that has joined the Defence establishment and is still waiting for security clearance a few months after being hired.

    Nothing new there with such military systems. Robert Oppenheimer became the scientific director of the Manhatten Project in mid 1942, but only received his secruity clearance in mid 1943.

    Comment by RJL — August 19, 2013 @ 2:10 pm

  23. So I just tried out some numbers based on the criteria in Andrew’s comment at #7. The executive summary is this:

    1. If you’re screening 376,000 public servants, you can expect to falsely identify around 75,000 of them as paedophiles.
    2. I’m pulling numbers out of thin air for the rates of people with false identities, an inability to pass the information requirements, and presenting a risk to children (and for the accuracy of the testing).

    Given those, here’s my starting calculations for peer review

    Comment by Steve (yeah, that Steve) — August 19, 2013 @ 2:14 pm

  24. Oops. Posted too early.

    Here’s my starting calculations for peer review:

    False Identities
    - I assume the rate of false identities is 1:80,000 and the test for detecting them is 99.99% accurate
    - Out of our starting population of 376,000, we’ve got 5 people with false IDs and 375,995 who don’t
    - Screening picks up all the people who really have false IDs and inaccurately says 37 people also have false IDs

    Now we’re left with 375,958 public servants.

    Comment by Steve (yeah, that Steve) — August 19, 2013 @ 2:24 pm

  25. Information requirements
    - I assume the rate of being unable to pass the information requirements (criminal convictions?) is 1:20,000 and the test for detecting them is 99.99% accurate
    - Out of our starting population of 375,958, we’ve got 19 people who don’t pass the information requirements and 375,939 who do
    - Screening picks up all the people who really don’t pass the information requirements and inaccurately says further 38 people don’t. (I don’t know what that would look like in real life, but I expect it’d be a Kafkaesque bureaucratic fustercluck.)

    Now we’re left with 375,920 public servants.

    Judgment-based risk assessment
    - I assume the rate of presenting a risk to children is 1:15,000 (I’m sure there’s a source for more accurate figures on this).
    - Because this is judgment-based, I’m guesstimating the test for detecting people who present a risk to children is 80% accurate
    - Out of our starting population of 375,920, we’ve got 25 people who present a risk to children and 375,895 who don’t
    - Screening picks up 20 people who present a risk to children. The screening fails to pick up 5 people who do.
    - The screening also inaccurately identifies 75,184 people as presenting a risk to children. They’re all instantly fired.

    Comment by Steve (yeah, that Steve) — August 19, 2013 @ 2:25 pm

  26. “This is a tricky topic to deal with, because only a really tiny proportion of allegations of CSA (Child Sexual Abuse) are false, as far as anybody can tell.”

    In a study of 25 child sexual abuse cases in Canada, 36% of jurors said pre-trial that they were not impartial. Some jurors said they had been victims of abuse, others expressed fears for children, while some said they were unable to maintain the presumption of innocence. It is worth noting that the jurors’ comments were made prior to jurors hearing any evidence.

    In a study by Quas (2005), many participants expressed erroneous beliefs about child sexual abuse. More than half (56%) falsely believed that a child was unable to describe sexual abuse unless they had “actually experienced it.” More than half (52%) wrongly believed that “When a child’s description of sexual abuse is disclosed over time, with more details being reported each time the child is interviewed, this indicates that the child’s description is true.” One of the most widely endorsed myths, endorsed by 68% of participants, was that a psychologist could tell “whether a child’s description of an event has been influenced by another adult.” Even when most participants were correct, a large minority were wrong. More recently, a similar study in Australia led the authors to conclude:

    “the Australian public lacks a sound understanding of children’s reactions to sexual abuse, as well as children’s memory, reliability and suggestibility when disclosing and reporting sexual abuse …
    [l]ess than 10% of the participants correctly agreed that children may sometimes be led by an adult to falsely report sexual abuse. One third of the participants incorrectly thought that adults cannot mislead children to make a false report of sexual abuse and 61% were uncertain.

    Comment by Ross — August 19, 2013 @ 6:47 pm

  27. > only a really tiny proportion of allegations of CSA (Child Sexual Abuse) are false, as far as anybody can tell.

    Research by Gail Goodman has shown that a significant minority of young children can be error-prone when asked specific abuse-related questions. When questioned in a laboratory setting, between 20–35 per cent of three- to four-year-olds falsely assented to questions such as “Did he try to kiss you?”, “Did he keep his clothes on?” and “He took your clothes off, didn’t he?”. Studies have found that false and potentially troublesome claims can also be elicited from pre-school and school-age children, even when asked non-leading questions. Unintentionally false reports can be elicited from children even when the event in question is recent. See Garven et al, “More than suggestion: The effect of interviewing techniques from the McMartin Preschool case” (1998) 83 J of Appl Psych 347. The authors found that 58 per cent of four to six-year-olds accepted false or misleading information about a week-old event. Moreover, 44 per cent of children falsely assented to questions about touching after just five minutes of improper questioning.

    Comment by Ross — August 19, 2013 @ 6:58 pm

  28. Will charter school teachers be government employees?

    Good point.
    CS teachers will not be Government employees – hell, they wont even have to be teachers – and as the schools will be excluded from other oversight, I suspect this criteria probably wont apply either.

    Comment by Gregor W — August 20, 2013 @ 10:13 am

  29. There seems to be an assumption that only male government employees will screened and sacked. What about women government employees- are they under threat of immediate sacking?

    Comment by Leopold — August 20, 2013 @ 10:41 am

  30. Screening picks up all the people who really don’t pass the information requirements and inaccurately says further 38 people don’t. (I don’t know what that would look like in real life, but I expect it’d be a Kafkaesque bureaucratic fustercluck.)

    It’ll probably look something like this: http://www.stuff.co.nz/national/9061652/Woman-told-shes-not-who-she-says-she-is

    Comment by RJL — August 20, 2013 @ 1:17 pm

  31. I’m sure Patrick Gower is just reading up on the copious information provided and will be around to hold Bennett’s feet to the fire over this shortly

    Comment by Michael — August 20, 2013 @ 6:45 pm

  32. It’s my belief they would be referring to the kind of police checks that (I think?) some school staff get. Or if they don’t think they are, that’s where it will end up. Any actual psychometric screening tool will in itself only produce some sort of risk or list of risk factors rather than an answer. And you’d need someone qualified to administer it.

    (Banks has asserted whatever it is will apply to charter schools BTW http://www.scoop.co.nz/stories/PA1308/S00236/hipkins-wrong-pskh-employees-will-be-vetted.htm)

    Comment by lyndonhood — August 21, 2013 @ 10:56 am

  33. I’m not a racist, but I would have thought that replacing heavily accented foreigners in front-line positions in the Ministry of Social Development (including Work & Income officers and Work & Income telephone operators) with Kiwis that can speak the english language clearly would be more important than firing professional people that are suspected of being suspected of being suspected of being suspected of doing something criminal in the course of their job, when there is already significant processes in the public service sector to prevent these things from occurring.

    Comment by Daniel Lang — August 26, 2013 @ 10:15 am

  34. I’m not prejudiced against people called Daniel Lang, but I would have thought that people who write run-on sentences nearly 100 words long shouldn’t criticise others’ language use.

    Comment by helenalex — August 26, 2013 @ 2:50 pm

  35. Daniel, we are ALL heavily accented foreigners.

    Gist cus you aur frim knizilind.

    Comment by Clunking Fist — August 27, 2013 @ 1:07 pm

  36. Wasn’t there something from Gluckman recently about how politicians don’t understand science?

    http://www.sciencemediacentre.co.nz/2013/09/03/sir-peter-gluckman-on-the-role-of-evidence-in-policy-making/

    Yes, science and evidence and so on. So now we can add that politicians don’t understand Bayesian reasoning. Which I guess makes them the same as almost everyone, in that being right is hard, which is why we need to rely on evidence.

    But Ms. Bennett says this law could have possibly (but not certainly, of course) prevented one case of abuse. So there you go, if you only have to ruin the lives of thousands of teachers and doctors and nurses and councilors and coaches and police and firemen and … who knows who else, to save a few children some pain, huzzah!

    And won’t it be awesome once we’re living in a world where whispers can get you instantly sacked. Won’t have to put up with any of those homosexual men being teachers, that’s for sure. Evidence? Bah, you just can’t trust them, and now you won’t have to.

    Comment by tussock — September 7, 2013 @ 3:12 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

The Rubric Theme. Create a free website or blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.

Join 419 other followers

%d bloggers like this: