See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249743977

Realistic Evaluation of an Illicit Drug Deterrence

ProgrammeAnalysis of a Case Study

Article in Evaluation · January 2008

DOI: 10.1177/1356389007084673

CITATIONS READS

28 428

1 author:

Liliana Leone
Studio CEVAS
102 PUBLICATIONS 400 CITATIONS

SEE PROFILE

All content following this page was uploaded by Liliana Leone on 31 May 2014.

The user has requested enhancement of the downloaded file.

 Evaluation
http://evi.sagepub.com

Realistic Evaluation of an Illicit Drug Deterrence Programme: Analysis of a Case

Study
Liliana Leone
Evaluation 2008; 14; 9
DOI: 10.1177/1356389007084673

The online version of this article can be found at:

http://evi.sagepub.com/cgi/content/abstract/14/1/9

Published by:

http://www.sagepublications.com

On behalf of:

The Tavistock Institute

Additional services and information for Evaluation can be found at:

Email Alerts: http://evi.sagepub.com/cgi/alerts

Subscriptions: http://evi.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.co.uk/journalsPermissions.nav

Citations http://evi.sagepub.com/cgi/content/refs/14/1/9

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008
 Evaluation
Leone: Realistic Evaluation of an Illicit Drug
Copyright © Deterrence
2008 Programme
SAGE Publications (Los Angeles,
London, New Delhi and Singapore)
DOI: 10.1177/1356389007084673
Vol 14(1): 9 – 28

Realistic Evaluation of an Illicit Drug

Deterrence Programme
Analysis of a Case Study

LILIANA LEONE
Studio CEVAS, Rome, Italy

This article describes an evaluation conducted in a region of northern Italy,

based on the realistic evaluation approach. This case study is related to the
application at regional level of national legislative measures against personal
drug consumption, the so-called Article 75, through the implementation of
a pilot programme. Article 75 DPR 309/90, is a ‘sanctions-based’ intervention
mainly based on the mechanism ‘threat–dissuasion’, such that people who
are found in possession of illicit drugs for personal use can be temporarily
detained by the police, have a mandatory interview and abstinence test
and sometimes have an administrative penalty or a mandatory treatment.
First, the article describes the sequence of mechanisms anticipated by
decision-makers (legislator) and the mechanisms that actually emerged
from the programme evaluation; second, the outcomes resulting from the
interaction between the mechanisms and the specific contexts are analysed.
Finally, a few suggestions are proposed and the advantages of the realistic
evaluation approach are discussed.

K E Y W O R D S : evidence based; illegal drugs; prevention; realistic

evaluation; sanctions

Introduction
‘Realistic evaluation’ (Pawson and Tilley, 1997), together with the broader trad-
itions of ‘realist’ evaluation, belongs to the current trend for theory-based evalu-
ation (Chen, 1990; Stame, 2004; Weiss, 2000). However, with regard to programme
analysis, realistic evaluation focuses not on the theory underlying the programmes,
i.e. ‘why the programme should work’, but rather on the theory of the desired
changes, i.e. which mechanisms are activated through the programme, in what
circumstances and with what results.

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 9 12/21/07 11:38:52 AM

 Evaluation 14(1)
Realistic evaluation, as well as realistic synthesis (Pawson, 2006), utilizes a
‘generative’ approach to causation.
According to this perspective they are not the ‘programmes’ that work; but rather the
underlying reasons or resources that they offer subjects that generate change. Caus-
ation is also reckoned to be contingent. Whether the choices or capacities on offer in an
initiative are acted upon depends on the nature of their subjects and the circumstances
of the initiative. (Pawson, 2002)

The object of this evaluation (Leone, 2006) was an Italian sanctions-based meas-
ure (a national law named Art. 75 DPR 309/90) aimed at dissuading personal
illicit drug consumption. A two-year programme,1 ‘Pilot Programme Territorial
Government Office’, within the framework of the national legislation, was initi-
ated by the Lombardy Region (Department of Family and Social Solidarity) in
cooperation with the Territorial Government Office of the Interior Ministry in
Milan and three local health units.2 This article is based on an evaluation conducted
by the author; it was commissioned by LHU MI2 and by Lombardy Region and
ran from February to November 2004.
Following the logic of ‘realistic evaluation’, our evaluative question was not:
‘does the programme work?’. Rather, the study explored the possible connections
between contextual conditions (i.e. specific characteristics of the organizational and
institutional contexts and of the subjects) and the ‘mechanisms’ developed in the
field resulting from the interactions between the actions of the initiative and the
reactions of the target group. According to realistic evaluation, the causal power of
the programme lies in its underlying mechanism (M) and in its basic theory about
how programme resources will influence outcomes (O) (e.g. a subject’s behaviour)
in a particular context (C). Establishing the way in which context, mechanisms and
outcomes interact in this case – so-called CMO configurations – is the basis for the
evaluation reported in this article.
The evaluation’s aim was to provide regional decision-makers and the local
health units with useful indications to inform the redesign of the previous regional
programme and implement the pilot programme. Therefore, the main focus was
on the possible connection between the outcomes and the different practices and
intervention models used by the Territorial Government Office (TGO) and the
local health units (LHUs) involved in the pilots. This article focuses on the effects
of the ‘mandatory interview’, which is central to this kind of programme.

The Programme Theory: The Old Programme and the ‘Pilot

Programme’
The ‘Old’ Programme
Use of drugs is not defined in Italian law as a criminal offence. However, posses-
sion of illegal drugs is prohibited and punishable by administrative sanctions in
case of personal use, and by prison sentences in case of dealing or trafficking.3 The
measure, Article 75 of the national law DPR 309/90 (hereafter referred to as the
‘old programme’) concerning drug addiction, establishes a set of complex actions.
In summary these include: people who are found by the police while consuming

10

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 10 12/21/07 11:38:53 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
or in possession of a small quantity of illicit drugs are temporarily arrested for
a short period (a few hours), registered and summoned by the Territorial
Government Office of the Interior Ministry (TGO) to attend a mandatory inter-
view, i.e. a formal meeting with the Prefect of Police or his representative (TGO
social workers).
The mandatory interview, lasting approximately one hour, generally occurs
many months – about twelve – after the temporary police detention. This is
managed by a social worker employee of the TGO, who employs a counselling
approach. The aims are to dissuade drug consumption, to supply information
about risks and treatment for people with drug addiction and to make social diag-
noses of problematic drug consumption. Generally the procedure ends with a
‘formal invitation’ to avoid drug consumption in future.
In a minority of cases the social workers recognize persistent drug abuse prob-
lems or repeated arrests, by the police under Art. 75. These cases must undergo
a mandatory treatment programme usually provided by the Health Services for
Drug Addictions (SerT). The treatment generally consists of a programme of four
to five group sessions led by professional educators or by psychologists. Manda-
tory urine tests are used to check drug abstinence during or at the end of the
treatment programme.
An alternative to these mandatory measures, when people refuse to attend an
appointment or formally refuse a summons, is the imposition of administrative
penalties (e.g. suspending the validity of some personal documents, such as a driv-
ing licence or passport). This measure is based on the double mechanism ‘threat-
dissuasion’ and ‘sanction-dissuasion’. The measure of mandatory interview is
intended to have a preventive and dissuading effect on the whole population,
while the administrative sanctions and the mandatory treatment should dissuade
the young illicit drug consumer from repeated consumption. The dissuasive effects
of mandatory interview on drug users are due to threat or conviction. Drug test-
ing after the treatment could reduce drugs consumption by deterring (deterrence
mechanism) potential users who will not risk incurring administrative penalties.
Table 1 summarizes the measures (see four steps) included in the ‘old pro-
gramme’ under Art. 75 and the hypothesized response of the target population. It
highlights the negative sanctions and threat base of the ‘dissuasive’ mechanisms
being used. However it should be noted that these sanctions cannot be assumed
to be effective in many circumstances. Thus the probability of being arrested for
the possession or consumption of illicit drugs is very low (less than 1%) in a prov-
ince such as Milan. This does not mean that large numbers of young people are
not caught up by the measure as it targets drug behaviours of a large proportion
of the young population. In Lombardy Region 27.6 percent of 14–19 year olds in
2002 consumed illegal substances at least once in the last year (LYP  Prevalence
of Drug use in the Last Year Reported).4

The Pilot Programme

The pilot programme (Sperimentazione Prefettura) started in September 2003 and,
after the evaluation, it was extended to the whole Lombardy Region, involving
other TGOs and three LHUs (LHU A, B and C). During the years 1998–2001 the

11

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 11 12/21/07 11:38:53 AM

 Evaluation 14(1)
Table 1. Basic Programme Theory of Art. 75 and the Sequence of the Mechanisms
Hypothetically Active in the ‘Old Programme’

Step 1 Step 2 Step 3 Step 4

Identification of the Mandatory interview Administrative Drug users and

transgressors (drug users) in the TGO and sanctions for young people
and punishment of law treatment : ‘sermon’ those who do response: the
infringement: police to dissuade drug not comply with programme is
detention and summons consumption, to interview or supposed to
to a mandatory interview supply information treatment discourage
in the TGO about risks. (i.e. do not illicit drugs
Verification of observe consumption
possible problematic abstinence) (deterrence effect)
behaviour towards
the use of illicit
drugs and sometimes
prescribed mandatory
treatment in LHU
(Problem: the probability (Problem: it occurs
of being arrested is very too many months
low in Milan Province: after the temporary
less than 1%) police detention)
Mechanism: Mechanism: Mechanism:
Dissuasive effects on Dissuasive effects Dissuasive effects
drug users due to on drug users on drug use due
fear of sanctions due to fear to threat or
of new sanctions conviction
following the
‘sermon’ and
(sometimes)
the treatment

TGO was inundated with work (relating to the integration of immigrants) and at
the same time about 3000 young people attended a mandatory interview for the
violation of Art. 75. This was the most important starting point for the pilot pro-
gramme, which exceeded the numbers encountered in any other part of Italy.
In contrast to the old programme, the regional pilot programme ensured
that the interview took place on the premises of the LHU or of the so-called
‘Polyvalent Youth Centres’ (centres belonging to non-profit organizations
with regional accreditation) involved in the pilot programme. The interviewers
are social workers from the LHUs or from non-profit organizations; they are
not from the TGO. There were not any other significant differences between the
programmes.
The following points need highlighting. Both programmes – the old and the
pilot – resulted in treatment for only a small proportion of the sample (in the

12

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 12 12/21/07 11:38:53 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
three LHUs from 4 to 13 percent of the young people attending the interview)
and in both the programmes it (the ‘old’ and the ‘pilot’) was not undertaken in the
TGO. Over a two-year period the pilot programme involved 2036 young people
from 17 to 26 years of age; 75 percent of whom were summoned for consumption
of cannabis, hashish-marijuana, 9 percent for other substances such as cocaine,
MDMA, heroin, and for 16 percent we have no information. Finally, the pro-
gramme was intended to be the same in each of the three LHUs; however, from
the beginning there were evident differences relating to the territorial context
and to the way in which the non-profit organizations collaborated.
According to the actors involved in the regional pilot programme, the implicit
theory (why the programme should prove to be effective) is as follows: if the man-
datory interview is carried out in a place as LHU or of the so-called ‘Polyvalent
Youth Centres’, which is not the TGO, it is possible that young drug consumers
(note, not drug addicts) may accept more assistance and support. Even though
the interview, in both cases, takes place in a coercive context (i.e. there is no free
choice in the TGO, LHU or the Polyvalent Youth Centre) the interview in the
TGO is seen as exerting a stronger control function. It must be considered that
the local TG office historically does not provide health or social services to the
citizen and that its main task is one of public security.
The Polyvalent Youth Centres (PYCs) are managed by organizations in the
third sector. They don’t belong to the SerT – Drug Addiction Health Service,
nowadays labelled as a service for drug addicts. The PYC and the SerT gener-
ally provide detailed information on opportunities and access to rehabilitation
programmes in response to requests for help. They are decentralized and easily
accessible: three centres provide services for 2.5 million inhabitants, while only
one TGO in Milan covers a territory of 3 million inhabitants.
The pilot programme aims for ‘selective’ prevention action (Burkhart, 2003).
Although the aims are basically the same as the old programme, it places more
emphasis on the following goals:
• increase information about the risks of using illicit drugs;
• raise awareness among young drug consumers of the risks linked to the use
of drugs;
• encourage young drug consumers to access the drug addiction treatment
territorial health services.

Evaluation Questions and Methods

This section presents the evaluation questions and the research hypotheses
emerging from the reconstruction of the theory (Pawson and Tilley, 1997: 155)
underlying the programme.
Note that the programme theory and hence the underlying evaluation ques-
tions require the following:
• the intervention of the evaluator – who generally works on the basis of
his/her experiences and the substantive literature: it is very important for
them to know the intervention field, the policies concerning the subject

13

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 13 12/21/07 11:38:53 AM

 Evaluation 14(1)
of the study and other existing evaluation studies related to the specific
programme;
• the exploration of the ‘unspoken’ and undeclared theories applied by the
regional decision-makers, local responsible directors, coordinators of the
local health units and professional workers involved in the interventions
(see analysis of the documentation, reports, deliberations, administrative
acts and agreement protocols).
To this end, we analysed the official and semi-official documentation, monitoring
reports and deliberations and conducted semi-structured individual and group
interviews, with the purpose of uncovering and understanding the logic of the
programme and underlying the actions of the different actors involved. Table 2
relates the hypotheses to the evaluative questions.
The table outlines the three main research hypotheses: the circumstances of
the target group, the context in which measures occur and the composition of
the professional groups involved. These hypotheses were translated into research
questions that guided the evaluation.
The evaluation’s aim was to provide the regional decision-makers and the
local health units with useful indications to inform the redesign of the programme
and put the pilot project into operation. Therefore, the main focus was on: a) the
possible connection between the outcomes of the project, in terms of results and
advantages/disadvantages for the young beneficiaries; and b) the different prac-
tices and intervention models used by the local health units involved in the pilot
project.

Methods
The following methods were used to address the evaluation questions:
1. individual interviews with different responsible members of the services
and four group interviews with the social workers and the staff of operators
from four different locations (three LHUs and one TGO);
2. analysis of the monitoring data provided by the TGO itself (database) and
by the local health units with reference to the two-year pilot programme;
3. survey through a semi-structured telephone interview with a sample of
100 young people coming from the whole province of Milan who had been
interviewed in the previous two months in the LHUs and in the TG Office
(to compare the pilot to the old programme);
4. three focus groups with the beneficiaries of the treatment provided by the
different LHUs (while noting that the treatment follows interviews in only
10–14% of cases).
In all instances confidentiality and the protection of sensitive information were
guaranteed. The services were required to obtain written consent from the inter-
viewees who were to be contacted by the researchers. In turn the researchers
were obliged to respect privacy regulations, not supplying to any of the institu-
tions (regional authority, territorial government office, local health unit) any of

14

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 14 12/21/07 11:38:53 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
Table 2. Research Hypotheses and Evaluative Questions

Research hypotheses Evaluative questions

A. Threats of sanction do not act in the 1. Have Art. 75 measures exerted a

same way on all the population and under dissuasive effect with regard to drug
all conditions (i.e. at all times). consumption behaviours? Are there any
A1) We know that people more ‘included’ unforeseen impacts?
in society and with regular work can 2. Which individuals benefit the most
be more sensitive and reactive to from this measure and which individuals
their arrest while people with do not?
problems of social exclusion are
less influenced or may even react
aggressively.
A2) Young ‘drug users’, with more
experience of drug consumption, are
probably less frightened by the
measure and could develop
concealment strategies.
B. The context in which the interview is 3. Can the fact that the first interview
carried out affects the attitude of young takes place in the Local Health Unit
people towards sanctions under Article 75. (and not in the TGO, as foreseen by
B1) If the interview is done in the Local the legislative measure) influence the
Health Unit, the interviewee feels effectiveness and the outcomes of the
that his conduct has to do more intervention?
with his health. 4. Is there a different way in which the
B2) If a person is interviewed in the young drug consumer elaborates and
TGO, he/she feels that his/her reinterprets his/her experience
conduct has to do with depending on the context of the
compliance with the law. interview?
B3) The individuals who attend the 5. Is the accessibility of the treatment
interview in the Polyvalent Youth services within the pilot programme
Centres within the LHUs comply with improved?
the treatment to a greater extent
than those who have their first
interview in the TGO.
B4) Carrying out the interview in the
Polyvalent Youth Centres could
encourage individuals with abuse or
misuse problems to access treatment.
C. The different composition of the team – 6. Which organizational models and which
the professions involved and the context-related factors account for the
composition and mix of private and efficiency of the project?
public staff – will influence the effectiveness 7. Which working practices and which
of measures. organizational solutions are connected
with more satisfactory outcomes
according to the beneficiaries? Which
organizational model seems to guarantee
better performance of the measures used?

15

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 15 12/21/07 11:38:53 AM

 Evaluation 14(1)
the collected disaggregated data. The telephone interviews were made by ap-
pointment (generally the interviewees preferred mobile phone numbers) in order
to respect privacy.
We undertook mobile phone interviews only because the young people – whether
they were students, living with parents or whether they had a job – wanted to
avoid other people knowing about their involvement in a police drug detention.
The characteristics of the recipients, such as age group, gender, school attend-
ance, employment status and drug consumption, when registered by the police,
are common and similar in the four subgroups: three LHUs and one TGO.

The Effects of the Art. 75 and Mechanisms Activated by the

Pilot Programme
This section examines the sequence of mechanisms emerging from the pilot pro-
gramme. Overall, it can be said that the programme was consistent with expect-
ations, even though some of the mechanisms ‘jammed’ due to the implementation
procedure in different organizational and institutional contexts and due to the
different reactions of the individuals towards a programme based on the mech-
anism of threat–dissuasion.
The sequence of mechanisms illustrated earlier (see Table 1) had been subject
to changes: in fact, during implementation, the pilot programme brought out new
unforeseen and sometimes unwelcome outcomes. In any case the research con-
cerned a ‘punitive measure’ of which the young people were ‘frightened’. As a
result the target group could have given a systematic overestimation of benefits,
withholding important information. For these reasons, during the analysis and
interpretation of the data, the focus was on the coherence and correlation among
variables rather than on frequencies alone.

Effects and Impacts of the Art. 75 Measure

This section addresses first a more general question regarding the Art. 75 meas-
ure, in both the old and the pilot programmes: has the measure had a dissuading
effect on drug consumption behaviours? Are there any unforeseen impacts?
It is apparent, from the statements of young people involved in the research,
that one of the results of the Art. 75 measure, for the majority, had been increased
awareness of the consequences of illicit drug consumption. With deeper analy-
sis, negative unexpected effects in subgroups of the sample also emerged. Of the
sample, 45 percent stated that temporary police detention was highly discouraging
of the consumption of illicit drugs: 55 percent of the interviewed people do not
believe that temporary detention is ‘useless’. Therefore it seems that half of the
young people saw some utility in the Art. 75 measure. In addition 60 percent
asserted that the people the Art. 75 measures affected (those who were arrested
and had a mandatory interview) feel ‘frightened’: whether this emotion could have
been associated with negative, positive or other effects will be discussed later.
With regard to mandatory interviews in Art. 75, we found that 60–74 percent
(varying across the three different LHUs) of the sample believed that the interview

16

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 16 12/21/07 11:38:54 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
helped them to think about the use of substances, that it wasn’t a waste of time,
and that it could be useful for people who consume illicit drugs such as cocaine or
heroin and this measure could have a dissuasive effect for occasional consumption
of illicit substances. Some indications also emerged about indirect positive impact:
one-third of the sample stated that their group of friends was also indirectly influ-
enced by the Art. 75 measure (see Var 117: The interview dissuades the friends).
Previously, I said that negative unexpected effects also emerged, i.e. ‘hidden
and opposition effects’, in a subgroup of the sample. It was believed by 55 per-
cent of the sample that one consequence of the Art. 75 measure was that young
people decided more carefully where to consume, choosing places that are less
frequented by the police.
Another proportion of individuals expressed indifference, 9 percent of the
subjects stated that they weren’t helped to reflect, and 15 percent of them developed
hostility and unwelcome outcomes, which favoured ‘concealment’ rather than ad-
dressing the problem of drug-taking.
Negative unforeseen outcomes were noted for a proportion of individuals who
were more at risk, most likely to be cocaine users, who seem to have become
‘oppositional’ and probably obtained less benefits from the programme. Of the
sample, 7 percent stated that the experience could lead to a negative reaction: an
increase in the consumption of illicit drugs. As confirmed by evaluative literature
on criminal prevention programmes, many punitive measures seem to get the
worst results among target groups at risk of social exclusion processes: unemploy-
ment, low education as well as ‘drug misuse’.
Considering the effects of the pilot programme related to Art. 75, taking into
account those who did not attend the interview, and the results obtained during
the two-year period (based on analysis of the monitoring data of TGO) we find
that there are diversified effects in relation to three different subgroups of effects.
Figure 1 presents the intermediate results and the outcomes of the application
of the Art. 75 measure in the pilot programme. The pilot programme was carried
out between September 2002 and June 2004 with a population of 2677 individuals.
This figure summarizes the entire process relating to the Art. 75 measure.

Subgroup A This consisted of around 23 percent of the cases, made up of those

who did not attend the mandatory interview. These were defined as ‘unknown
persons’ who did not benefit from any intervention related to the process because
we did not have any information at our disposal to enable us to assess the
usefulness of the process to them. They avoided the ‘sermon-information’ step
(the interview in the LHU) and opted for the administrative sanction. There is
marked difference between LHUs: in the LHU A (the city) we noted 30 percent
of people refusing the mandatory interview, in the provincial LHUs, this figure
ranged from 15 (LHU B) to 18 percent (LHU C).

Subgroup B This consisted of 61 percent of cases who complied with the

procedure subsequent to the temporary police detention/registration by the police.
For these individuals, the procedure was suspended after the first interview or was
continued through treatment programs (generally four to five group meetings).

17

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 17 12/21/07 11:38:54 AM

 Evaluation 14(1)

Subgroup B) 61%
People summoned to usefulness of the
interview - Art. 75 in the Local measure
Health Units - 2677
2064 attended the interview
(77%) Subgroup C)
15% undesired
effects
613 did not attend the
interview (23%)
271 were sent to
treatment
groups (10.1%)
Subgroup A)
unknown effects
23%) imposition of
sanction 217 (8%) 185 individuals 48 dropouts
undergo treatment (1.8%)
1576 formal
(6.9%)
invitations
(59%)
113 positive outcome
group treatment
(4.2%)

Figure 1. Outcomes – Measure Art. 75 in the Pilot Programme (%s calculated in relation
to the total number [2677] of individuals who were sent to the interview in the LHUs)

Subgroup C This consisted of around 15 percent of cases – individuals who,

after having initially complied with the obligations (interview and/or treatment),
developed undesirable effects and/or became oppositional, for example: increased
drug consumption and improved ability to avoid controls.
Note that all the data were obtained from the monitoring documents, while the
estimation regarding subgroups B and C were extrapolated from the evaluation
research, which was conducted on a sample of 100 individuals.
While the pilot programme ‘Sperimentazione Prefettura’ only marginally
affects the subgroup ‘Unknown persons’ and there is no way to ascertain the pos-
sible effects of the Art. 75 measure (dissuading effects resulting from the threat of
being arrested/registered), it emerged that the measure might have had positive
outcomes for Subgroup B. For instance there was an increase in the treatment
offers by LHUs and a waiting-time reduction for mandatory interviews

What are the Mechanisms?

We found that the reactions to the ‘mandatory interview Art. 75’ are diverse and
can be classified into four different types of mechanism within an overall typology.
These are activated as a consequence of the interaction between the programme
and particular subgroups of users.
We also found that the mechanisms developed, and therefore the outcomes, are
statistically correlated to the place, the institutional and social context in which
the mandatory interview took place (i.e. TGO or the PYC of the LHUs).

18

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 18 12/21/07 11:38:54 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
During the data analysis phase, six variables were identified corresponding to the
dimension ‘usefulness of the treatment’ and two variables corresponding to the
dimension ‘experience of fear’. Combining these two dimensions we can identify
the four possible reaction types (A, B, C, D) summarized in Figure 2.
Subsequently, the data were examined (using Datastat software) through an
analysis of the raw frequencies and the percentages for the whole sample and dif-
ferentiated by LHU and TGO. In order to analyse the connections and to con-
struct contingent tables, the variables were dichotomized. To calculate the degree of
statistical significance, chi-squared tests and the R Phi tests were used, these being
adequate when variables are measured on the nominal and ordinal scale, as in this
case. A factorial analysis was conducted in order to identify mechanisms and reac-
tions typologies underlying attitudes and opinions collected during the survey.
Thirteen variables relating to the previously mentioned dimensions and to the
undesired and desired reactions subsequent to the procedure were selected; a
factorial analysis was conducted5 (see automatic rotation of the axes) using only
one of the two modalities (‘YES’) of the variable.
In Figure 3 we combined the four theoretical types illustrated previously with
factors emerging from the factorial analysis. The first type, ‘oppositional and
critical individuals’, is composed of those who are angry and express poor appre-
ciation of the experience and opposition: 20 percent of individuals react in this
way to the pilot programme and show a greater tendency to hide his/her own con-
sumption practice and may even change the type of illicit drug for more harmful
drugs, such as cocaine.6 In this subgroup we find people who express their feelings
of social discomfort; they are often cocaine addicts who underwent treatment
or were sanctioned. This subgroup is intended to be the privileged recipients of
the interventions planned for the pilot scheme. Individuals belonging to this type
are likely to have had the interview in the TGO and not in the Polyvalent Youth
Centres of the Local Health Units.
The second type is made up of ‘frightened’ individuals who reported feelings of
discomfort and the dissuasive effects of the experience. They are individuals who
are frightened by the Art. 75 measure, they scarcely reflect on the suggestions
they are given during the mandatory interview and tend to be reticent. It can be
argued that, during all phases of the procedure (from the police detention to the
interview) one should avoid the use of intimidatory and dissuasive tools, because
it potentially generates reticence in the individuals under treatment and it doesn’t
help them to use the experience as an opportunity for personal development and
raising awareness. Also in this case we find that the factor ‘frightened’ and the
location of the interview in the TGO are connected.

Usefulness Yes Usefulness No

Fear YES A) Fear and usefulness of the treatment B) Scarce usefulness and fear

Fear NO C) Lack of fear and usefulness D) Scarce usefulness and no fear

Figure 2. Typology of Expected Responses

19

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 19 12/21/07 11:38:54 AM

 Evaluation 14(1)

USEFULNESS YES USEFULNESS NO

Dissuasive A) Interviewee feels frightened, they see the B) They don’t express much appreciation,
effect interview as useful, they feel uneasy veiled criticism, they are complacent
YES

II. frightened

IV.
indifferent:
useful to

III. reIflective
others

Dissuasive
effect C) They express
NO both appreciation I. oppositional
Attention!
and criticism,
they distinguish risks
according to illicit substances and D) They express poor appreciation and
consumption modalities oppositional attitudes

Figure 3. The Theoretical Types and the Factors Previously Identified

The third type consists of ‘reflective’ individuals, who viewed the Art. 75
measure, particularly the interview, as useful. They are people who express both
appreciation and criticism and make a distinction between risks according to the
modalities of different illicit substances and their consumption. These people
develop trust in the organizers of the interview and are at ease during the inter-
view. This suggests that a warm and friendly atmosphere, and a non-judgemental
context encourages the individual to reflect upon their own personal experience
of the use of illicit drugs and their consumption, thus helping them to reduce the
risks to their health.
The fourth type consists of ‘indifferent’ people, which includes those who con-
sidered the experience of Art. 75 not useful, even though they expressed only veiled
criticism. In these subjects the punitive dimension and fear of sanctions were pre-
dominant; while the psychological effects on their life experience was very weak.

The Relation between Contexts and Outcomes

This section focuses attention on the following questions:
• Can the fact that the first interview takes place in the LHUs (and not in
the TGO) change the effects of intervention? Is there improvement to the
accessibility of the treatment services?
• Is there a different way in which the young drug consumer interprets his/her
experience in relation to the place where the interview is carried out?

20

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 20 12/21/07 11:38:54 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
• Are younger people (15–18) more receptive to the punitive measure than
older ones (19–26)?
• Are there any differences correlated to the age, place where they come from
(city/province), drug consuming experience, level of education?
The pilot programme increases the accessibility of the territorial health services
for young drug consumers (but not the drug-addicted). A relationship exists be-
tween attending the interview in the LHU and regarding it as a place that inter-
viewees can recommend to their friends who are experiencing the same problems.
The three different PY Centres of the LHUs are mentioned as an ‘Advice place’
in 11–33.3 percent of the cases and the TGO is mentioned only in one case. These
data confirm that the contact was successful and that a trustful relationship be-
tween the young person and service operators of the LHU had been developed.
We can provide a positive answer to the second question. The institutional and
organizational context of the interview affected significantly the way the young
consumers perceived the measure in all its aspects: from the police detention to
the subsequent interview. The hypothesis was confirmed. The use of illicit sub-
stances was more likely to be considered as a ‘health problem’ if the interview
took place in the LHU context.
The most striking difference about the experience of the interviewees was in
their prevalent emotions: the interviewees in the TGO systematically reported ‘the
experience of fear’ (see Table 3) while those interviewed in the LHUs reported ‘the
experience of ease’. The connection between the experience of feeling at ease and
undergoing the interview in the LHUs (67.6 percent in the LHUs against 8.5-chi
squared  6.79, df (degree of freedom)  1 ** p  .01) is statistically significant.
The data suggest that a strong sense of fear is linked to the following contrast-
ing effects:
• a negative opinion toward drug consumption;
• an increase in defence mechanisms (evasive answers);
• compliance during the interview (Hawks et al., 2002: 41).
There is a relationship, even though not statistically significant, between ‘Interview
in the TGO’ and young drug consumers ‘becoming shrewd’: the young consumers
now try to avoid being caught by police (46.1% TGO against 32.4% LHU).

Table 3. Was the Interview/Procedure Frightening?

LHU TGO Total

Yes 37 23 60
% Col. 50.0 88.5 60.0
No 37 3 40
% Col. 50.0 11.5 40.0
Total (cards) 74 26 100

Chi squared  11.86, df (degree of freedom)  1 ***

(p  .001)

21

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 21 12/21/07 11:38:55 AM

 Evaluation 14(1)
Table 4. Comparison of Two Subgroups of Interviewees: Metropolitan vs Provincial
Residents

Variables LHU Milan Other LHUs and

city ( % ) TGO ( % )

108. Police detention dissuades 32.35 51.52

116. The interview dissuades from occasional 85.29 92.42
consumption
117. The interview dissuades the friends 26.47 34.85
111. The interview is a waste of time 26.47 12.12
112. They become more shrewd consumers 38.24 63.64
114. As a reaction to the interview 23.53 25.76
increased consumption
110. The interview helps reflection 79.41 95.45
123. The consumption of illicit drugs 79.41 92.42
seen as a health problem

Table 4 compares the group of young people who come from the city of Milan
with the rest of the sample, who come from the rest of the province and the
suburbs. The table shows weaker dissuading effects (Var 108) and more nega-
tive judgements regarding the utility of the measure for the group from the city
(Milan). At the same time in both the subgroups (city and province) we observe a
similar rate of negative effects such as increasing substance consumption (as nega-
tive reactions to police detention and mandatory interviews) and ‘concealment
effects’ such as disguising drug consumption behaviours (Var 112). These findings
can be partially explained by context differences in norms of social control. In a
metropolitan area such as Milan, there is a high rate of drug consumption among
young people. They are less recognizable when consuming and have less respect
towards the institution and punitive measures. In the province the young people
are more likely to experience pressures of social control (managed by police and
other institutions) and become more careful when consuming illicit drugs.
In Table 5 the results are shown to differ in relation to the educational level of
the interviewees. Young people with a high level of education (diploma or uni-
versity degree) show better perception of risks associated with illicit substance
consumption. They think that society considers the consumption of illicit drugs a
health care problem (not only a legal problem) and during the interview they are
better able to make a clear distinction of the risks (e.g. risks of road accidents) in
relation to different substances (21.67% vs 2.50%).
Police detention is explicitly considered as discouraging by 38.33 percent of
those who have a lower level of education, whereas those with a high degree of
education (diploma or degree) are less inclined to consider police detention as
discouraging (12.50%).
It should be noted that the small sample size is a limit in the present study,
because it does not allow other in-depth analysis of the connections between

22

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 22 12/21/07 11:38:55 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
Table 5. Comparison of Two Subgroups of Interviewees: High vs Low Level of
Education

Variables Chi Square High (%) Low (%)

8. The temporary police detention discourages 7.94 38.3 12.5

33. Distinction of risks connected to drug use 7.32 21.7 2.5
21. Society considers illicit substances consumption 7.24 43.3 17.5
as a health problem

Chi squared df  1 * (p  .01)

target group characteristics (age, sex, ‘user’/‘non-user’, etc.) and the place/context
of the interview.
Finally the last evaluative questions regarding different ways of implementing
the same programme were as follows:
• Which organizational models and which context-related factors explain the
effectiveness – or lack of effectiveness – of the programme?
• Has there been improvement in the accessibility of the treatment services in
the pilot programme?
• Which working practices and which organizational solutions are associated
with more satisfactory outcomes according to the beneficiaries? Which
organizational models are associated with better programme performance?
Accessibility to the services, professional qualifications of the operators, their dif-
ferent strategies and treatment flexibility partly explained the successful implemen-
tation of the treatment and the different outcomes in the three local health units.
The way in which the actors understood the programme is crucial. We observed
significant differences among the LHUs: the use of sanction-based behaviour and
of administrative measures rather than mandatory treatment programmes (both
measures pertaining the individual rights) proved to be quite variable.
The percentage of individuals who had an administrative sanction imposed on
them in the three local health units varies from a minimum of 4.6 percent (LHU
B) to a maximum of 13 percent (LHU C); similarly the percentage of individ-
uals who were sent for treatment varies from 10.7 to 16 percent. The differences
related to the dropout from treatment process (from 4% in LHU B to 24% in
LHU C) are noteworthy, i.e. those who did not attend the group treatment were
sanctioned.
This variability is not due to the characteristics of the users of the three local
health units, but is mostly due to:
• the decisions and orientation of the professional operators, who have ample
space for discretion within a framework of the measure;
• their perception of the possible utility of the measure, i.e. how much all the
staff agreed with the legislative measure;
• the treatment programmes offered.

23

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 23 12/21/07 11:38:55 AM

 Evaluation 14(1)
The LHU C team was, for example, convinced that sending young drug consumers
for mandatory treatment was useless and moreover there was a lack of interaction
between the social workers of the public health services for drug addicts (SerT
of LHU) and the educators of the not-for-profit organizations. In this LHU the
young users were obliged to attend a double interview and they encountered
more difficulty in setting up appointments. The latter contributed to the largest
expenditure within the treatment programme of the LHU.

CMO Configuration Identified in the Pilot Programme

Sufficient data were collected to assert that the involvement of the public health
services in cooperation with non-profit organizations can represent a good solu-
tion to improve some drug prevention outcomes of a deterrence measure such as
Art. 75.
The fact that the interview takes place in the LHUs encourages access to the
network of territorial health and social services by new drug users who generally
do not consider themselves to have consumption or addiction problems. They can
ask for support from the health services; these drug users technically aren’t ‘drug
addicted’, they could be simple consumers or problematic users. This is a target
group strongly recommended for selective prevention in international guidelines
(Burkhart, 2003).
If we observe the most relevant differences in the two main contexts – on the
one hand, the three LHUs and on the other the TGO – we can simplify the two
CMO configurations highlighted by this evaluation as follows.
Table 6 summarizes only some aspects of the CMO configuration that emerged
during the study: in terms of mechanisms we have identified how the operators
have interpreted their task (e.g. the tendency to impose sanctions), and the reac-
tions of the recipients (see factors previously identified).
The sequence of mechanisms activated in the programme partly corresponds
to the sequence hypothesized by the decision-makers at the outset; while parallel
mechanisms have emerged, which affect the total outcomes in both negative and
positive values. Some of the outcomes obtained led us to believe that, although
being the result of dissuasion–punishment oriented policies, this measure indeed
brings about ‘harm reduction’ results (i.e. a reduction of risks related to substance
consumption).
Additionally, we observed different rates of sanctions for mandatory treat-
ment, which are not due to factors such as different characteristics of the users,
but to discretion in the interpretation of the measure by the service operators.
Each team, during the interview and in the following treatment, ‘interpreted
the programme’ according their assessment of the usefulness of the sanctions in
affecting addiction and use of illicit substances.
Some factors correlated with greater acceptance to the mandatory interview
and with a low dropout rate during treatment; they are the treatment organiza-
tional model (i.e. LHU B) and the logistic solution of the PYC.
The availability of a more flexible and tailored offer, in relation to the different
phases of consumption and degree of addiction, allows a better compliance and

24

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 24 12/21/07 11:38:55 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
Table 6. CMO Configuration

Place of Interview Responses Results

Territorial Government Feeling of discomfort during Greater presence of

Office of Interior Ministry interview, they become reactions such as:
Strong presence of a hierarchy shrewder and avoid being concealment (they
Values: control and public caught by police while become more shrewd)
order security taking illicit substances. by the target (65% vs
Presence of police and soldiers Strong effect of frightening 53% in LHUs)
in the access facilities and experience of fear
The team is composed of social (88.5% of TGO vs
workers that do not perform 50% in LHUs)
any treatment but only Feeling of fear which affects
interview retrospectively the
Only 1 TGO for a population moment of the police
of 3 million inhabitants detention which is
considered as ‘dissuasive’
(65.38% vs 37.8% in LHUs)
Only a proportion of the
beneficiaries state they
feel at ease (38.5% vs
67.6% in LHUs)
Low number of individuals
are sent to treatment
(3.8% vs 24.3% in LHUs)
LHU, generally a Polyvalent Feeling at ease (67.6%) Increased ability within
Centre and not the Health Less strong reaction of fear: the target group to
Service for Drug only in 50% of beneficiaries. differentiate the risks
Addictions (SerT) The LHU context increases related to the
There is an interaction between the degree of accessibility consumption of
the social workers involved and develops a more substances and to
in the Art.75 interview and positive atmosphere in one’s health
the team involved in the which to establish a Use of illicit substances
treatment programme. relationship of help is more likely to be
and support considered as a health
Development of a trusting problem
relationship with the Greater access and
Polyvalent Centre of information from the
the LHU health territorial
services
Less negative impacts
such as ‘opposition’
and ‘concealment’

participation in the treatment programme. Evidence suggests that it is necessary

to identify an unbiased place, i.e. ‘a place not socially labelled and not labelled as
a setting for drug addicts’. Additionally, it emerged that time plays an important

25

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 25 12/21/07 11:38:55 AM

 Evaluation 14(1)
role: there is a reduction in the usefulness of the mandatory interview and treat-
ment if the procedure is carried out a considerable time after police temporary
detention.
In conclusion, regarding the general utility of the measure, some doubts have
emerged: half of the sample stated that the measure ‘has a dissuasive effect’ but
also admitted to concealment. Subgroup C (around 15 percent of subjects) de-
veloped undesirable effects and/or became oppositional, for example, increased
drug consumption and improved ability to avoid controls.
It was not possible to include people who refused to have contact with the
public services and chose administrative penalties instead (23 percent of the
people with a police detention under Art. 75), because their names were not public.
A specific evaluation of this subgroup is still required.

Final Observations
The political debate around limits and strengths of punitive measures against per-
sonal drug consumption is always very heated. In the EU Member States legal
approaches to the use and possession of drugs for personal use range from toler-
ance of the use of certain drugs to penal sanctions for any use of any substance.
A recent overview states:
. . . in many countries, personal use of illicit drugs is considered a relatively minor of-
fence, incompatible with custodial sanctions. However, it would be a mistake to inter-
pret this as a ‘relaxation’ or a ‘softening’ of drug laws in the EU. And many of the 10
new EU Member States still consider use or possession for personal use as a crimi-
nal offence punishable by sanctions of ‘deprivation of liberty’ (e.g. imprisonment).
(EMCDDA, 2005)

The programme evaluated here was developed from a national sanctions-based

measure that aimed to discourage and punish personal drug consumers. Several
elements have been introduced to make the measure ‘less punitive’ and more
functional in obtaining health prevention effects.
After the evaluation, the pilot programme was extended to the entire
Lombardy Region, involving other TGOs and LHUs and some recommendations
that emerged from the evaluative research were taken into account.
The construct of CMO configuration, elaborated utilizing realistic evaluation,
was very useful to orient the evaluation design (e.g. to find the right evaluation
questions) and to search for an explanation as to how the complexity of elem-
ents and dimensions were interacting. Thanks to the concept of ‘mechanism’ we
attempted to explain ‘how’ the subjects reacted to similar interventions developed
in several contexts, and why different effects were observed.
The differences in circumstances were multifaceted and observations were
limited, so the analysis concentrated on only some differences in the context (i.e.
institutional nature and organizational differences of the services that delivered
the programme) rather than others. In the future there needs to be study in detail
of the effects of a punitive–dissuasive measure, such as Art. 75, on personal drug
consumption, also on the target group of people who chose the administrative
penalties and refused to attend the mandatory interview.
26

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 26 12/21/07 11:38:55 AM

 Leone: Realistic Evaluation of an Illicit Drug Deterrence Programme
Evaluators through meta-evaluations and realist synthesis need to consider
also the negative effects of punitive measures aimed at regulating the demand for
illicit drugs when comparing legislative measures in different countries. The most
important systematic reviews to date do not offer a clear indication about this
specific point, although there are indications that the prevention approach had to
be multifaceted, had to avoid single knowledge-oriented interventions and should
be based on educational strategies, such as ‘life skills’ programmes (Brounstein
et al., 1998; Hawks et al., 2002; NIDA, 2003).
One advantage of the realistic evaluation approach is its idea of the cumulative
process of knowledge production. Instead of asking ‘Does the programme work?’,
it is possible to improve gradually the understanding of ‘outcomes’ patterns as
well as other different CMO configurations emerging from other evaluations of
similar interventions as suggested by the concept of ‘family of mechanisms’ pro-
posed as part of realistic syntheses (Pawson, 2006).

Notes
1. Agreement protocol Jan. 2002: ‘Sperimentazione Prefettura-Avvio della sperimentazione
coordinata per il trattamento di persone segnalate ai sensi artt.75 e 121 del T.U. delle
leggi in materia di tossicodipendenze – DPR 309/90’.
2. Territorial Government Office of Milan (Prefettura di Milano), Local Health Unit,
LHU MI2 (ASL Azienda Sanitaria Locale MI2 – Melegnano Leader in the pilot pro-
gramme, LHU Città di Milano, and LHU MI3 – Monza). Evaluation Report available
at www.cevas.it/casi/index.htm
3. EMCDDA website. See country profile: Italy, ‘drug use and possession’, http://eldd.
emcdda.europa.eu/index.cfm?fuseaction  public.content&sLanguageISO 
EN&nNodeID  5174
4. Relazione Annuale al Parlamento sullo Stato delle Tossicodipendenze In Italia 2003,
Ministero del Lavoro e delle Politiche Sociali, 608, 2003. Report ESPAD (European
School Survey Project on Alcohol and other Drugs) 2003.
5. The variance has been: Fat. 1: 0.284; Fat. 2: 0.259; Fat. 3: 0.246; Fat. 4: 0.211. general vari-
ance: 0.448.
6. ‘If I want to take drugs, I do it in hiding, and if I have to risk, I prefer taking cocaine to
a joint’ (statement by a boy).

References
Brounstein, P. J., J. M. Zweig and S. E. Gardner (1998) Science-Based Practices in Substance
Abuse Prevention: A Guide, Substance Abuse and Mental Health Services Administra-
tion, Centre for Substance Abuse Prevention (CSAP). Available at: http://prevention.
samhsa.gov (accessed 31 Aug. 2005).
Burkhart, G. (2003) Report on Selective Prevention in the European Union and Norway,
Selective Prevention: First Overview on the European Situation, Lisbon: EMCDDA
European Monitoring Centre for Drugs and Drug Addiction. Available at: http://www.
emcdda.eu.int/index.cfm?fuseaction  public.Content&nNodeID  1569&sLanguage
ISO  EN (accessed 31 Aug. 2005).
Chen, H. (1990) Theory Driven Evaluations. Newbury Park, CA: SAGE.
27

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd 27 12/21/07 11:38:56 AM

 Evaluation 14(1)
European Monitoring Centre for Drugs and Drug Addiction (2005) Illicit Drug Use in the
EU: Legislative Approaches (11/2/2005), Lisbon: EMCDDA. Available at: http://eldd.em-
cdda.eu.int/index.cfm?fuseactionpublic.Content&nNodeID5175&sLanguageISO
EN (accessed 1 May 2006).
Hawks, D., K. Scott, N. McBride, P. Jones and T. Stockwell (2002) Prevention of Psycho-
active Substance Use: A Selected Review of What Works in the Area of Prevention. Geneva:
World Health Organization (WHO) – Mental Health: Evidence and Research Depart-
ment of Mental Health and Substance Dependence. Geneva, Switzerland Available
at: http://www.who.int/substance_abuse/publications/en/prevention_substance_use.pdf
(accessed 31 Aug. 2005).
Leone, L. (2006) ‘Promoting the Utilization of Effectiveness Evidence and Guidelines in
the Sector of Health Promotion’, paper presented at the Joint International Conference
UKES–EES, ‘Evaluation in Society: Critical Connections’, London, 4–6 Oct.
National Institute on Drug Abuse (2003) Preventing Drug Abuse among Children and
Adolescents: A Research-Based Guide for Parents, Educators, and Community Leaders.
Washington, DC: NHI National Health Institute, Publication No. 04–4212 (A). Available
at: http://www.nida.nih.gov/Prevention/Prevopen.html (accessed 31 Aug. 2005).
Pawson, R. (2002) ‘Evidence-Based Policy: The Promise of “Realist Synthesis”’, Evalu-
ation 8(3): 340–58.
Pawson, R. (2006) Evidence Based Policy: A Realistic Perspective. London: SAGE.
Pawson, R. and N. Tilley (1997) Realistic Evaluation. London: SAGE.
Stame, N. (2004) ‘Theory Based Evaluation and Types of Complexity’, Evaluation 10(1):
48–76.
Weiss, C. (2000) ‘Which Links in Which Theories Shall we Evaluate?’, in P. J. Rogers,
T. Hacsi, A. Petrosini and T. A. Huebner (eds) Program Theory in Evaluation: Chal-
lenges and Opportunities, New Directions for Evaluation, 87, pp. 35–45. San Francisco,
CA: Jossey-Bass.

L I L I A N A L E O N E is Director of Studio CEVAS (Consulenza e Valutazione

nel Sociale). Her evaluation practice is concentrated in the domains of social and
health programmes and public care services. Please address correspondence to:
Studio CEVAS,Via Calpurnio Fiamma 9, 00175 Roma, Italy. [email: Ieone@cevas.it;
website: www.cevas.it]

28

Downloaded from http://evi.sagepub.com at CNTRL INST INDIAN LANG on December 11, 2008

084673_EVI_9-28.indd
View publication stats 28 12/21/07 11:38:56 AM