New flaws in Garda figures emerge as 89 homicides go unrecorded

Policing Authority expresses ‘alarm’ at latest lapses in record keeping

The Republic’s homicide figures are missing 89 killings because of mistakes in the way homicides have been recorded by Garda members.

The Policing Authority said the figure was “alarming”. It was also concerned that the problem had never been flagged with it before.

The 89 homicides that were not counted in the homicide data occurred between 2003 and May, 2017. All of them were cases of dangerous driving causing death.

The disclosure came as senior gardaí appeared before the Policing Authority in public session on Thursday.

READ MORE

Defending the force’s recent record, Ms O’Sullivan said while trust levels in the Garda had dropped to an all time low in April 2014 - around the time she took over as interim commissioner - but it had reached an all time high of 91 per cent in the first quarter of this year.

Asked by authority member Dr Moling Ryan why she felt trust in the Garda had increased she said: “I think putting the focus on the concept of service and changing the attitudes of members on the ground,” she said, adding putting victims at the centre of Garda work had also been important.

However, Garda management were still unable to explain why the number of alcohol breath tests carried out between 2012 and 2016 had been inflated from one to two million in the official Garda figures.

A second interim report on the matter has been received in recent days, with a final report due before the end of July.

However, authority member Maureen Lynott said while the reports outlined the actions taken by the Garda to address the problem, there was still no assessment of how the matter had occurred.

Garda Commissioner Nóirín O’Sullivan said Deputy Commissioner John Twomey would have been able to speak about the matter but could not attend Thursday’s meeting. She also said the reports were “interim reports, it’s not finalised”.

“It’s a combination of factors, it’s very hard to weigh up which factors are which,” she said, adding the final report was imminent.

However, chair of the Policing Authority Josephine Feehily said her concern and that of her colleagues was that a second interim report had just been received and no explanation was available for how the inflating occurred.

“One would have thought if the final report is imminent we’d be getting a list of what the (Garda) is looking at; the beginnings of some analysis. But there is no analysis at all.”

Data errors

It was the errors with the homicide data that occupied most time at Thursday’s meeting.

Authority member Vicki Conway told a team of senior Garda members, led by Ms O’Sullivan, that she was concerned about what other problems were about to emerge with homicide data.

She also wondered were there concerns about, for example, the accuracy of data around serious injuries sustained in road traffic collisions.

Ms Feehily said she was given the impression at the last public meeting two months ago that a report on other categories of homicide was almost complete. However, it had still not been received.

“We were led to believe it would be available by now,” Ms Feehily said.

Assistant Commissioner Eugene Corcoran said the delay had arisen because of a need to carry out further checks.

He added “in order to be sure that we’re completely and confidently reassuring the public” checks were continuing to Garda districts. They were seeking to establish the level of investigation carried out at the time into those homicide cases where recording issues had been identified.

“We expect in the next four weeks or so to have concluded our work,” Mr Corcoran said.

At the last hearing, in April, it emerged a review of homicides committed between 2013 and 2015 had been carried out.

Issues had emerged in 41 killings, though those problems were not set out in any detail.

Domestic murders

Dr Gurchand Singh, the Garda’s head of analytics, said at April’s meeting he could recall his analysts finding “three or four” incidents in which domestic murders were counted in the crime statistics as non fatal offences against the person.

In explaining how the separate 89 cases of dangerous driving causing death had not been counted as homicides in the crime statistics, he said the errors had occurred in three ways.

Firstly, the manner in which some dangerous driving causing death cases were filed omitted mention of the deceased person. The injured party should have been “flagged” as part of the data entry process when the crime had been recorded. And the absence of “flagging” meant those cases were not captured in the end of year homicide figures.

Secondly, in some cases of dangerous driving causing death, other crimes were also committed. And the entire incident was categorised as one of the other, more minor, crimes rather than the homicide offence.

Thirdly, other cases involved multiple fatalities but in the official figures only one of those fatalities was recorded.

Dr Singh said “the bulk” of the cases had occurred in the early years of the period under review.

He added a process of education and training was underway to ensure the recording and data entry errors did not happen again. The issue had never been identified despite previous Garda and Central Statistics Office reviews of crime data.

“To me, I think it is a process issue in terms of how people were putting on the incidents on (the Garda database),” Dr Singh said.

“There is no attempt to hide a homicide. There’s no attempt to declassify a homicide to a more minor offence.”

Conor Lally

Conor Lally

Conor Lally is Security and Crime Editor of The Irish Times