Peter Bassett: Headline promises a lot but delivers little.

If you survey people who are still standing, don’t be surprised when most say they haven’t fallen.

That’s the problem with the New Zealand Herald’s headline this week that a “Majority of NZ drug and alcohol users experience no negative effects, survey shows”. 

That’s a headline that is technically defensible, politically useful — and profoundly overreaching.

The article, by Derek Cheng, leans heavily on a survey run by Massey University’s Shore and Whāriki Research Centre, led by Professor Chris Wilkins. 

Cheng frames it as “groundbreaking”. In reality, it is neither new nor neutral, and it barely supports the policy conclusions quietly hung on it.

A survey of survivors

The core finding sounds reassuring: 63 percent of respondents to an online survey of nearly 5,000 people reported no negative effects from alcohol or drug use in the previous six months.

But read the methodology carefully. The sample is self-selected. It is anonymous. It is drawn exclusively from people who admit to using drugs in the last six months and who are sufficiently functional, motivated, and socially intact to complete an online questionnaire.

Those who are dead, incarcerated, hospitalised, psychotic, estranged, or otherwise detached from “polite survey populations” are absent by design. So are people whose harms are delayed, cumulative, or primarily inflicted on others.

In polling terms, this would not pass muster with any professional firm. 

As David Farrar or any experienced pollster would point out, it is an uncontrolled convenience sample with no baseline, no control group, and no capacity to infer population-level risk.

Calling it “nationwide” does not fix that. Large numbers do not rescue biased sampling. And self-reporting — particularly on stigmatised behaviours — systematically undercounts socially disapproved harms such as violence, impaired driving, or parenting failure.

Acute harm, not real-world harm

Wilkins is careful, to his credit, to acknowledge some limits. The index measures acute harm over a six-month window. It does not capture chronic illness, long-term cognitive damage, cumulative addiction, or downstream effects that emerge years later.

Yet this crucial caveat is buried beneath confident rankings and a tone that invites policy inference. LSD and cocaine are described as “lower risk”. MDMA is said to score near zero for driving harm. Methamphetamine violence is downplayed as rare or underreported.

This is not how harm works in the real world.

Drug harm is unevenly distributed. It is episodic, cumulative, and often delayed. Many of its most serious consequences first affect other people — partners, children, victims, emergency staff — or remove users from survey reach altogether.

Emergency department admissions, coronial data, crime statistics, and child protection files exist precisely because self-report surveys cannot see these harms clearly. 

Excluding the inconvenient 

One methodological choice deserves particular scrutiny. Respondents who reported harm involving more than one substance were excluded from the harm index “to avoid confusion”.

In plain English: the messiest, most realistic cases were removed.

Poly-drug use is not an edge case. It is the norm. Alcohol combined with cannabis, meth with alcohol, MDMA with stimulants — these combinations are where risk multiplies. Excluding them sanitises the picture and biases the results toward lower apparent harm.

This is not a neutral decision. It materially shapes the outcome.

Advocacy disguised as analysis?

The article then pivots smoothly from preliminary data to policy critique. 

The Misuse of Drugs Act is said to be misaligned with harm. Decriminalisation models are praised. Portugal is invoked.

None of this flows logically from the survey.

Wilkins’ framing aligns closely with long-standing harm-reduction advocacy. So does Cheng’s selection and presentation of quotes. This matters, because readers are not told that the same arguments have been made — repeatedly — by political actors who want the law changed.

Such as … Greens co-leader and Auckland Central MP Chlöe Swarbrick.

The Herald does not name her. But the framing is unmistakable.

For years, Swarbrick has argued that public concern about drugs is irrational, that criminalisation multiplies harm, and that policy should be guided by user experience rather than social alarm.

Her stock lines are well worn:

• “Drugs are winning the war on drugs.”
• “Forty years of punishment hasn’t worked.”
• “Criminalisation causes more harm than drugs themselves.”
• “Policy must be evidence-based and health-focused.”

The Cheng article packages Wilkins’ survey in precisely this rhetorical mould.

Fear is “out of step with reality”. Harm is overstated. User self-assessment is elevated to evidential primacy.

The political argument is subtly validated before the reader reaches the data.

Portugal: the unexamined talisman

Portugal’s 2001 decriminalisation is cited again as proof of success (Something Swarbrick unfailingly peddles). It is now more than two decades old. Yes, some early indicators improved: HIV transmission fell, opioid deaths declined, and incarceration dropped. 

But serious researchers have long noted that many trends predated legal change, outcomes depended heavily on sustained treatment investment, and recent years have seen rising use and service strain.

Portugal is not a magic template. It is a complex, context-specific case — not a blank cheque for policy transfer. The Herald does not explore this nuance.

A ‘Pleasant’ assertion 

Cheng also quotes Sarah Helm, executive director of the NZ Drug Foundation, describing psilocybin and MDMA as having “pleasant effects with low harms”.

This is presented without challenge or sourcing.

Helm is an experienced public health advocate and communications professional. 

But in a piece already reliant on self-reported data, adding unqualified assertions about “low harm” substances further blurs the line between evidence and persuasion.

What this survey can — and cannot — say

None of this is to argue that every drug user is doomed, or that punitive policy is cost-free. It is to insist on intellectual honesty.

A survey of functioning users over six months can tell us something modest: that many people who are currently using drugs and are able to complete surveys do not perceive immediate harm.

It cannot tell us how harm accumulates, how risk concentrates, how many fall out of view, or how many others pay the price.

It certainly cannot justify sweeping claims that public concern is misplaced or that existing law is fundamentally irrational.

Cheng is an experienced reporter. He would know how framing works; that headlines shape interpretation, and that caveats buried at paragraph 27 do not neutralise narrative thrust.

This article selects a particular kind of evidence, from a particular kind of source, and aligns it with a familiar political soapbox. 

If you only ask the standing whether they have fallen, the answer will always comfort those already convinced.

Peter Bassett is an observer of media, politics and public institutions, writing on how narrative replaces scrutiny.