International Journal on Criminology Volume 1, Number 1, Fall 2013 | Page 63

International Journal on Criminology
The first consequence of this was on the comparison of statistics in May 2004 with
those from May 2003 as the announcement [1] from the ministry underlined at the time
Crime in France (including the total number of cases of non-road-related crimes and
offences recorded by police and gendarmeries) decreased by -10, 25% compared with
May 2003.
One year later, when the completeness of the collection of incidents recorded in May
2005 had not been damaged; their numbers experienced a rise that the OND presented in
its exchanges with the ministry of the interior as the inevitable consequence of the
shortened month of May in 2004.
As of June 2005, the ministry knew that its institutional communication, and, later
on, its process of collection of incidents recorded was going to be questioned in public
debate by the OND as statistics are often called upon to prove the merits of government
actions, or its failure according to its opponents (Mucchielli 2012).
This perspective seemed even more inevitable than the OND considering that the
announcement had been written up in complete indifference to the work carried out and
that it was without doubt going to try to defend its position newly acquired by the
organization producing official statistics and trying to impose its opinions in the
framework of a thorough and well argued initiative. And all of this in an initiative not
intended for connivance or negotiation (Ocqueteau 2012) but to affirm its independence.
The reaction of the OND took place in September 2005 in a new article 21 dedicated
to monthly statistics: it then took the decision to re-correct statistics regarding incidents
recorded in May 2004 and in other months in 2004 for which it seemed to be relevant
regarding the values observed.
The ONDRP then estimated that in 2004 “the months of January, May, July and
October included 1 to 3 days of entering less data compared to the same months in 2003”
and that the months of “March, June, August, November and December had a period of
data entering which was longer in 2004 compared with 2003 (OND 2005b, 26.)” The
ONDRP corrected monthly statistics according to the estimation of the number of days
more or less of data collection.
By applying a statistical adjustment of data published without modification by the
ministry of the interior, the OND pursued its initiative autonomous of the distribution of
official statistics, which have been inaugurated by the creation of its non cumulative
indicators.
The head of the national police also reacted in the same article observing that “the
month of May 2004 was subject to a report for recording incidents recorded as that month
was not concluded by a prolonged weekend of three holiday days. However, it is not
possible to support the idea of systematic report as it would be contrary to the reality and
to rules of recording data which were fixed. Furthermore, it is necessary to specify that
the incidents are not deleted but reported for the following month. As a result, they are
integrated into the total of the following month as well as in the total for the semester
which is consolidated. ” (OND 2005b, 25).
The disagreement over the generalization of the method for adjustment seems to be
less of a crucial point in the relations between the OND and the head of the national
police than the consensus which appears in public debate between two organizations
producing official statistics for the month of May.
This means in particular that some affirmations from announcements by the ministry
of the interior regarding May 2004 and May 2005 are in contradiction with the position
expressed in September 2005 by the head of the national police itself.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
21 http://www.inhesj.fr/fichiers/ondrp/resultats_et_methodes/lettreond5.pdf
62!