Vol.1, No.3, 135-144 (2011)
doi:10.4236/ojas.201 1.13018
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
Open Journal of Anim al Sciences
Qualified “in shelter” dogs’ evaluation and training to
promote successful dog-human relationships
Marilena Sticco1, Roberto Trentini2, Pia Lucidi3*
1 Doctor in Animals’ Welfare and Protection, L’Aquila (AQ), Italy;
2 “G. Caporale” Experimental Zooprofilattico Institute of Abruzzo and Molise, Teramo, Italy;
3Department of Comparative Biomedical Science, School of Veterinary Medicine, Teramo, Italy;
*Corresponding Author: plucidi@unite.it
Received 26 July 2011; revised 5 September 2011; accepted 21 September 2011.
The phenomenon of dogs’ relinquishment in
Italy has become a social evil, although many
laws exist to regulate animal protection and
lately, the act of abandonment has become
criminalised (law n.189/2004, enforced by law
n.201/2010). Adop tion fro m shelters seems t o be
the only way to have a controlled, microchipped
population of dogs, as well as limiting con-
finement and euthanasia. After being asked to
simplify the previous Ethotest © version [13] by
many shelter operators and veterinarians, the
authors aimed at analyzing the effectiveness of
an improved model to test dogs’ behavioral ap-
titude matching the expectations of a hypo-
thetical adopter. The new version improves the
test feasibility by the elimination of a previous
computer-based program, and by the introduc-
tion of new items such as hierarchical behavior
towards food. In this study dogs housed in the
sanitary shelter of L’Aquila (Abruzzo, Italy), of
different age and sex, either sterilized or not,
and belonging to different breeds or cross-
breeds, were tested. All the dogs adopted from
the shelter were monitor ed fo r one y ear after the
adoption by both phone interviews and home
visit. The study aimed at analyzing if the shelter
dogs showed a good and consistent behavior
after adoption in the new environment. The re-
sults demon stra te d th at apart from a predictable
relinquishment and an unfortunate case of
abuse, none of the dogs adopted showed any
unwanted behaviors such as house soiling,
jumping up, separation-related and aggressive
behaviors; this made their stay in the family a
desirable, exciting experience independently of
the dog sex, age, and the family composition.
The authors stress the necessity of every shel-
ter, together with the veterinary cares, for a
professional expert at dogs’ behavior who can
efficaciously prevent behavioral problems,
eventually train the dogs and afford the pairing
with humans in a competent, qualified manner.
Keywords: Adoption; Behavioral Test; Shelter
Dogs; Training
All over the world the overpopulation of stray dogs is
a concern due to a number of dog attacks on infants and
adults [3,6]. Who is to blame for the failure of control of
stray dogs? In most cases the stray dogs overpopulation
results from housedogs and from irresponsible owner-
ship [17,20]. In fact, not all the owners sterilize their
dogs and, worst of all, they are allowed to roam at the
time of reproduction. This results in the possibility that
those dogs mate and give birth to unwanted puppies,
whose final destination is the abandonment without a
microchip for their identification.
In Italy, the law n. 281 regulates the capture and ster-
ilization of stray dogs since 1991. Consequently to the
enacting of this law, euthanasia of unwanted roaming
dogs has been forbidden, unless it can be demonstrated
that they are dangerous or incurable. This gave rise to
the establishment of a multitude of long-term shelters
where unattended dogs are placed in questionable condi-
tions, waiting for an adoption that sometime never oc-
curs. It has been estimated that the number of strays in
Italy amounted to 600,000 in 2009 (source: Italian Min-
istry of Health, www.salute.gov.it), but the problem of
unwanted dogs is a common, widespread topic all over
the world with millions of abandoned dogs ending up in
shelters [2].
In this study, the authors evaluated dogs’ traits that are
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
the requisites for living together amicably with humans
by simplifying the already published program Ethotest©,
which was realized in order “to lay the foundations for a
more flexible selection of dogs to be used as co-thera-
pists” [13]. In fact, after several shelters in different Ital-
ian cities adopted Ethotest©, many operators released
feedback asking for a revised, simplified program (here-
inafter Ethotest-R). Therefore, the aim of this study was
to develop a more simplified method of selection and to
sponsor the need for a qualified shelter caretaker to per-
form in order to improve the welfare of dogs and safety
of the adoptions. To analyze dogs’ behavioral traits, we
used the general outline of the previous Eth- otest© con-
sidering that, as proved by Hsu and Serpell [9], there are
only the following few factors stable and consitent
across different populations: stranger-directed fear,
stranger-directed aggression, owner-directed aggression,
non-social fear, dog-directed fear or aggression, train-
ability, and attachment.
2.1. Animals
The Ethotest-R version was carried out on 32 shel-
tered dogs (see Table 1); 24 of them crossbreed and 8
belonging to the following breeds: Abruzzi’s Shepherd
dog, American Staffordshire Terrier, English Setter,
German Shepherd, Maremma Hound, Pointer, and Sibe-
rian Husky. The dogs were housed at the small shelter of
L’Aquila health department (ASL 04) where we tested
all the dogs hosted. According to the Italian law for
straying control, all captured dogs are microchipped be-
fore their entry to the kennel, and then lodged in isola-
tion pens. Here, they are submitted to a blood test and
checked for anti-leishmania antibodies, vaccinated and
treated for parasites. After quarantined for a certain pe-
riod of time, they are transferred to other pens waiting
for being adopted. The pens, provided with external ar-
eas, could house several dogs together, depending on
how compatible they are with each other; otherwise
some subjects can be housed in isolation (i.e., the
American Staffordshire Terrier in our study).
This small regional shelter is unique in the Italian
scenario, because a qualified staff efficaciously cares for
the dogs. The dogs are fed on commercial dog food,
pens are cleaned more than once a day, and a qualified
veterinarian who cares for the animals’ healthiness peri-
odically visits all dogs. The veterinarian is available when-
ever an emergency arises. For any other needs, namely
ethological and physiological needs, there is a technician
qualified in Animals’ Protection and Welfare (one of the
Authors-hereinafter Operator 1); this Operator assists the
dogs by monitoring temperament and social interaction
Table 1. Characteristic of the dogs subjected to Ethotest-R.
ID n.Name Breed Age Sex SizeTime in
1Ada Mix 10 m spayed M6 m
2Bella Mix 8 y spayed M2,5 y
3BianconeAbruzzi Shepherd 11 y neutered M5 y
4Bobo Mix 10 m intact M6 m
5BraccoPointer 6 y intact M5 y
6Didi Mix 3 y spayed L2,5 y
7Eva English Setter 4 y spayed M2 y
8Frank Mix 8 y intact M4 y
9Gilda Mix 8 m spayed M3 m
10GregorioMix 10 y neutered G1 y
11GrethelMix 2 y spayed L2 y
12HanselMix 2 y intact L2 y
13Jack Am. Staff. Terrier 7 y intact L1 y
14Jamaica Mix 3 y spayed L2,5 y
15Lana Siberian Husky 5 y spayed L2 y
16Leon Mix 5 m intact L3 m
17Liz Mix 6 m intact M3 m
18LouiseMix 4 y spayed M3,5 y
19LupizzaGerman Shep. 8 y spayed L1 y
20 ManoloMaremmaHound 9 y neutered M5 y
21 Mitzy Mix 3 y spayed L3 y
22Nina Mix 7 y spayed M6 y
23OlguitaEnglish Setter 3 y spayed M5 m
24Petra Mix 3 y spayed L2,5 y
25Ripa Mix 4 y spayed M3,5 y
26Salvo Mix 2 y intact L1, 5 y
27Secco Mix 5 y intact S3 y
28SnoopyMix 5 m intact M2 m
29ThelmaMix 4 y spayed M3,5 y
30Tom Mix 5 y neutered L3 y
31Trilly Mix 2 y spayed M2 y
32Ugo Mix 6 y intact M6 y
Abbreviation: S, small; M, medium; L, large; G, giant.
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
of incoming animals and by deciding group composition,
housing, and time and occurrence of motor and social
activity (inter and intra-species).
In the morning, all the dogs are allowed to exercise
for nearly one hour in a wide area of more than thousand
square meters, which is adjacent to the external pens; the
dogs can stay in this area in compatible groups. The
dogs submitted to the simplified program of Ethotest-R
stayed at the shelter at least 2 months before being tested.
Their ages ranged from 5 months (dog #16) to 11 years
old (dog #3).
2.2. Establishment of the Evaluation Grid
In order to design a scheme for evaluating the behav-
ioral components of each dog, we only considered the
indicators that are required to a dog for living with hu-
mans in a satisfactory companionship; we explored a
few dimensions of dogs’ temperament, by referring to
the temperament traits described by Jones and Goslin
[10]. The schematic version of Ethotest-R is given in
Table 2. While Test A was carried out by the Operator 1
only, the sections B,C, and D were designed to analyze
three major behavioral traits (fearfulness, aggressiveness,
trainability) by two Operators: Operator 1, well known
to the dogs and a second Operator, stranger to the dogs,
as a control; this second Operator changed from an
evaluation to another because he/she became rapidly
known after their first entry. In order to differently load
the different sections (the items of section C were indeed
judged as more critical than B and D), the total scores of
the different items were multiplied for a coefficient that
was different from one section to another. For the same
reason, the coefficient was always higher for Operators 2.
The authors supervised the stranger Operators in order to
make his/her help consistent with the action to be taken,
from time to time. When both the Operators were needed,
they worked together (i.e. opening the fence together,
enter together, etc).
Test A
The first screening was focused on the dog aggres-
siveness/sociability. This Test was highly selective be-
cause it analyzed the aggression component in order to
clearly separate the dogs on the basis of a qualitative
measurement. According to such hypothesis, the second
item A2 has been centered on the component sociability
in another qualitative assay. Test A was used to immedi-
ately eliminate dogs that failed items 1 and 2 (the total
amount of dog’s responses must be two). The Operator 1
only has carried out this first measurement; indeed, if the
dogs behaved aggressively or fearfully towards the first
Operator, then we assumed that they would behave even
worse with unknown people.
Test B
Test B measured the fearfulness as a tract of the dog
temperament. This Test was carried out by the Operator
1, together with the Operator 2 (the latter did not carry
out item B3). It analyzed the boldness of the dog to-
wards the Operators when they opened the gate of the
fence and walked in. It also evaluated the fearful behav-
ior of each dog and how they reacted in very unfamiliar
places such as the entrance of the building, the corridors,
and the veterinary’s ambulatory, surgery, and office. The
coefficients used for the evaluations were 1.2 for the
familiar person and 1.5 for the stranger.
Test C
Test C evaluated other dog’s individual differences
associated to aggressiveness/submissiveness. For exam-
ple, in an open field, the Operator 1 carried out the en-
counter with a conspecific for the analysis of inter-dog
aggressiveness (C1), followed by the introduction of a
novel, unpredictable stimulus (C2), such as blowing a
trumpet or producing other unusual noises. Operator 1
also carried out a new item, not previously considered in
Ethotest ©, i.e. the hierarchical behavior as regard to
food (C3). This investigation was done to determine the
dogs’ social position compared with their pen compan-
ions (pack), or to the Operator (leader). Conversely, both
the Operators 1 and 2 in items C4 and 5 carried out the
study on dog suitability with human contact, such as
patting, manipulation (which can emphasize a dominant
temperament), and jumping on people. In order to pre-
dict a successful adoption, the dog’s coefficient score
had to be higher than in Test B-namely 1.5 for the Op-
erator 1 and 2.0 for the Operator 2.
Test D
Test D considered the responsiveness to training and
play of the dogs by evaluating their skills and their in-
terests in learning different commands and behave con-
fidently with humans. Dogs were examined twice, by
both the Operators, in the external fence and in the lane
way that enters the enclosure. Walking on a leash (D1),
sit down! (D2), and lie down! (D3) were scored due to
their attractiveness on future owners. Playing with the
Operators (D4) was chosen to study the disposition of
the dog to interact friendly with humans, which is diffi-
cult to find in long-term housed animals but expected to
be attractive to visitors. The Operators’ coefficient assign-
ed for this analysis was the lowest (1.0 and 1.2) because,
apart from the advantage to possess these skills, it is not
difficult for any equilibrate dog to gain them by education.
2.3. Classification of the Dogs and Cut-Off
At the end of the tests, the previous program utilized
two logical operators (IF and AND: the program is pro-
vided as supplementary data t doi: 10.1016/j.applanim. a
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. http://www.scirp.org/journal/OJAS/Openly accessible at
Table 2. Ethotest-R.
TEST A – Aggressiveness/sociability
ID animal age gender size breed
item Component Variable Behavior description Scores
A1 Aggressiveness towards people the dog snarls or threatens the operator
Operator 1 enters the
the dog approaches the operator in a friendly manner
A2 Sociability/diffidence the dog runs away avoiding the operator’s touch
the dog allows the operator to touch him/her
Totalmust be 2
TEST B – Fearfulness
item Component Variable Behavior description Operator 1 Operator 2
B1 Enterprise
Attempts of the dog to go out once the
gate of the fence is open
the dog does not go out
the dog goes out by itself
the dog goes out only when called
B2 Sociability II
When the operator enters the fence the
runs away
rushes near the operator
crouches or goes hesitantly
wags the tail and/or licks the operator’s hands
B3 Fearfulness to a
strange situation The dog is free to entry a new room
the dog does not enter
the dog enters the room cautiously and/or sniffs
the dog enters in a self-assured manner
The dog enters a new room with the
the dog does not enter, except when drawn
the dog enters only when called
the dog enters together with the operator
Outdoor: a new person is approaching
the dog runs far away
the dog goes far/or jumps on people
the dog approaches only when called
the dog approaches wagging the tail
Operator’s coefficient1.2 1.5
TEST C – Aggressiveness/submissiveness and jumping
item Component Variable Description Operator 1 Operator 2
toward other
In a open field (the exercise area) the
dog meets a conspecific, i.e. an
unknown dog approaching the
fence from the outside
the dog runs away or threatens the other dog
the dog approaches with hostile, dominant apti-
the dog approaches with appeasement signals
C2 Fearfulness
Introduction of a strong noise stimulus
the dog barks or snarls
the dog runs away frightened
the dog pays attention but does not run away
the dog stays calmly
While the dog is eating, another dog
approaches the bowl
the dog snarls and/or attacks the other dog
the dog eats voraciously
the dog eats quietly or it goes away
behavior While the dog is eating, the operator
approaches the bowl with her hand
the dog snarls
the dog eats voracious
the dog eats quietly
it goes away
The operator gently pat the dog
the dog runs away or becomes restless and
jumps on the operator
the dog allows the manipulation and wags its
C4 Aptitude to be
handled Harsher manipulation: the operator
restrains the dog with the arms on dog’s
back and pushes the dog to the ground
the dog rebels or runs away
the dog does not react to the domination of the
C5 Jumping How many time the dog jumps
on the operator
> 3
< 3
Operator’s coefficien
1.5 2
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
TEST D – Responsiveness to training and play
item Component Variable Description Operator 1 Operator 2
Walking with the operator
the dog does not walk on a leash
the dog draws
the dog draws sometime
the dog walks correctly on a leash
D1 Walking on a
Changing direction
the dog does not execute
the dog executes only when called
the dog executes
the dog executes with some distractions
D2 Sit down! How many times the operator
repeats the command
the dog does not execute
> 5
< 5
D3 Lie down! How many times the operator
repeats the command
the dog does not execute
> 5
< 5
With other dogs the dog does not play
the dog plays gladly
D4 Play With the operator
(tennis ball, squeezable toys)
the dog runs frightened
the dog shows no interested
the dog plays by himself
the dog plays with the operator
Operator’s coefficient1 1.2
2005.04.006), which needed the use of a computer to per
-form the selection. The logical operators were omitted
in this new simplified program; moreover, the scores
obtained by each dog in tests B, C, and D were not sub-
mitted to an independent evaluation, but they were con-
sidered on the whole, and the cut-off necessary value to
consider the dog for adoption was “more or equal to 40” ,
being the final score the resulting total score of the op-
erator 1 multiplied for its relative coefficient plus the
evaluation of the operators 2 multiplied for its relative
coefficient (intra-test confidence). The dogs that did not
reach the total score of 40 were considered unsuitable
for adoption; the subjects in the range 40-50 were con-
sidered fully adoptable dog and in the range more than
50 highly recommended dogs for adoption.
2.4 Adoption and Follow-Up
Before adoption the Operator 1 always interviewed
the aspiring owners to determine if the dog met their
expectation. The questioning was made in an informal
way, by asking the adopters about their experience with
dog, the time they could have spent with the pet, and
their lifestyle and dog’s eventual arrangement. In some
cases, the owners were asked to come back to the shelter
to socialize with the desired dog; in these cases, a train-
ing class was given to them, educating the adopters to a
safe and aware relationship with the animal. It consisted
of a general instruction about dog needs (behavioral,
dietary, and veterinary) and then the explanation of dog
postural and vocal signals, the effect of reinforcement
and punishment on dog’s learning, walking with a lash,
sit!, stay!, and how to avoid house soiling.
After the adoptions were successful, to determine if
Ethotest-R would predict a consistent behavior of the
dogs in the family environment, the necessity of a fol-
low-up was considered. To this aim, the Operator 1 car-
ried out a home visit two weeks after the adoption (when
possible) and always bi-monthly information by phone
call for a total of at least five surveys. The follow-up
addressed, with a yes/no Test, nasty habits that Chris-
tensen and colleagues [3] demonstrated to be unwanted
at home, such as: house soiling, jumping up, separa-
tion-related behavior, and aggressive behavior of any
type (Table 3). We intentionally omitted barking as an
unwanted trait because this behavior is often welcomed
by Italian dog owners (although not for the neighbor-
hood) against criminal offenses both in urban and coun-
tryside realities.
This interview has been taken by phone call every two
months during the first year post-adoption by Operator 1.
It focused on a few nasty habits showed by the adopted
dog including separation related behavior (3) and ag-
gressiveness of any type (4).
The results of dogs’ evaluation are given in Table 4.
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
Table 3. Follow up questionnaire.
Nasty habits yes no
1 House soiling
2 Jumping up
Object chewing
Agitation before being left alone
Stranger-directed aggression
Owner-directed aggression 4
Dog-directed aggression
After being submitted to the Test A, nine dogs out of
the original group of 32 were rejected and did not pro-
gress to the subsequent items for the insufficient pass
mark in Test A; this failure was never ascribable to ag-
gressive behavior but to their fearfulness, due to the lack
of socialization with humans in the critical period of
their life [7]. In the next B, C, and D Tests, being the
minimum requested score equal to 40, were rejected four
dogs (dogs #1, 14, 16, 31), that presented low score
mainly in the socialization items C3 (alimentary behav-
ior) and C4 (aptitude to be handled). In the group that we
considered to be a sufficient pass mark, (i.e. from 40 to
50) there were only two dogs (dogs #8, 26); then, in a
Table 4. Results.
Nr. Dog TEST A TEST B TEST C TEST D TOTAL scores Adoptability
3 Biancone 1 / / / 1 Not suitable
4 Bobo 1 / / / 1 “
11 Grethel 1 / / / 1 “
12 Hansel 1 / / / 1 “
18 Louise 1 / / / 1 “
24 Petra 1 / / / 1 “
25 Ripa 1 / / / 1 “
29 Thelma 1 / / / 1 “
32 Ufo 1 / / / 1 “
1 Ada 2 3.9 21 2 28.9 “
16 Leon 2 8.7 15 5 30.7 “
31 Trilly 2 9.9 21 4 36.9 “
14 Jamaica 2 15.6 18.5 5 39.6 “
26 Salvo 2 17.4 19.5 8,2 47.1 Suitable
8 Frank 2 11.1 21 15.4 49.5 “
28 Snoopy 2 11.4 21.5 15.4 50.4 “
17 Liz 2 15 21.5 12.6 51.1 “
13 Jack 2 21 10.2 18.6 51.8 “
21 Mitzy 2 12.6 21.5 18.4 54.5 “
2 Bella 2 21.6 16.5 15.4 55.5 “
10 Gregorio 2 21.6 15 19.2 57.8 “
20 Manolo 2 19.5 21.5 17.2 60.2 “
15 Lana 2 21.3 23 14.2 60.5 “
7 Eva 2 19.5 23 16.2 60.7 “
27 Secco 2 21.6 24.5 16.4 64.5 “
6 Didi 2 21 18 24 65 “
19 Lupizza 2 21 19.5 24 66.5 “
23 Olguita 2 29.7 18 20.8 70.5 “
9 Gilda 2 24.3 21 24.8 72.1 “
5 Bracco 2 25.8 20 24.8 72.6 “
30 Tom 2 24 23 24 73 “
22 Nina 2 22.5 22.5 27.4 74.4 “
esults after dogs’ evaluation with Ethotest-R (ordered per score).
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
higher position there were all the other dogs (15 dogs),
with scores higher than 50 and even 70, that possessed
characteristics of obedience and reliability, and certainly
suitable for adoption. After the dogs were judged,
twenty-two passed the Test program. Of them, only 12
were adopted, including one dog in a range considered
below the cutoff for adoption (dog #16). While in their
new family environment, the behavior of those dogs was
recorded during the next 12 months. The history of the
adopted dogs (alphabetically listed) and the results of the
follow-up interviews are summarized in Table 5. No
restriction was foreseen for adoption (i.e. experienced or
inexperienced owner, presence or absence of children).
The decision to commit or not each dog and to train the
owners was exclusively related to the expertise and
judgment of the authors. A mandatory training before
adoption was considered essential for the owners of dogs
#7, 13, 16, and 28. In the case of dog #16 the adopter
assured that she could have followed the mandatory
training class at home with an expert trainer and because
of the distance, the home visit was not possible.
This study aimed at giving to adult dog the chance to
be successfully housed and at suggesting to shelter
managers the possibility to introduce, or at least to con-
sult, a qualified dog-expert whose work, differently from
the sanitary rule of the veterinarian, can help in the
placement of the animals.
Being in touch with several shelters in Italy that used
a previous program [13] we were solicited to simplify
the Test, making it more suitable for a day-by-day selec-
tion of shelter dogs. In this study, any dog was known
and selected by a qualified Operator, whose expertise
was in the field of animal welfare, training, and behav-
ioral education of both owners and dogs. The Ethotest-R
program showed a better feasibility than the previous
Test: in fact, this version could be daily accomplished in
the shelter environment, provided that the dog-expert
Operators could dispose of external and unknown places
where to carry out all the items. We did not repeat the
Test to analyze the consistence of dogs’ behavior: we
found that, given that the animal behavior and human
interaction is affected by many conditions, the repetition
of the Test a few months later could have been very dif-
ficult and useless for our purpose. The results were in-
deed validated by asking the owners about dog’s behav-
ior in the new environment; the follow-up is in fact a
useful tool to understand the evolution of the animal’s
personality [22]. Differently from other authors [3], in
our study the follow-up results demonstrated that
Ethotest-R could be evaluated as a suitable adoption Test;
although only 32 dogs were evaluated (all the dogs
housed in that shelter), the sensitivity of the Test (the
proportion of true positives that are correctly identified
[23]) was 100%, i.e. every adopted dog was reported as
behaving well by their owners that responded to the Op-
erator 1 interviews. The follow-up focused on items that
have been demonstrated to be the primary reason for
returning a dog to a shelter, i.e. behavioral problems
Table 5. Follow-up of twelve dogs (ordered by dogs’ ID number) one-year post-adoption.
ID n. Dog’s name Sex Scores Adopter’s Family Presence of childrenPrevious dog/s ownership Dog housingFollow up
2 Bella F s 55.5 Woman with a dog- Yes Home N
5 Bracco M 72.6 N
9 Gilda F s 72.1
Young man - Yes Garden
7 Eva F s 60.7 Separated woman 2 teenagers No Home N
8 Frank M 49.5 Family - No Home N
13 Jack
§ M n 51.8 Man - Yes Home N
16 Leon * M 30.7 Young woman - No Home N
17 Liz F 51.1 Family 2 No Home N
19 Lupizza F s 66.5 Family - Yes Garden N
23 Olguita F s 70.5 Family 1 Yes Home N
26 Salvo M 47.1 Family 3 Yes Home N
28 Snoopy F s 50.4 Family 2 No Home N
The table represents the dogs adopted from the ASL 04 shelter after their evaluation. It underlines sex, total score obtained, adopter’s family situation and ex-
perience with dogs, and home/garden arrangement. In the last columns there are the results of the interviews taken during one-year of ownership in relation to
dog’s nasty habits (house soiling, jumping up, separation anxiety, and aggression). Sex: F = female; M = male; s = spayed; n= neutered. Follow up: N= negative
for all the nasty habits. §confiscated; *relinquished.
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
[2,5,6,14,19,26]. No noisy habits were registered during
one-year follow-up, first of all the aggressive behavior.
Despite the professional expertise, two dogs (i.e., #13
and 16) were not efficaciously paired with their owners.
The first was victim of mistreatments and confiscated,
the latter was a predictable case of relinquishment, being
a dog not suitable for adoption (total score below the
cutoff value); he left the shelter just before his owner
completed the mandatory training . Dog #13, a dog that
could have been forced to become aggressive towards
humans by the bad treatment he suffered, did not show
any sign of aggression, rather a form of depression in the
new situation. Although it has been demonstrated that
the ability to execute basal commands significantly red-
uces the likelihood of relinquishment [5,14,16,21,24,25],
promoting a “successful human-dog bond”, this was not
the fate of dog #13.
Apart from these cases, from our study it seems that
people were more attracted by educated dogs that is, in
conclusion, that dog’s behavior is more important to a
potential adopter than the dog’s physical appearance.
Every positively judged dog fulfilled the owner’s expec-
tation after adoption: there was no difference found be-
tween entire or neutered pets, and also the factor of gen-
der was not found significant [4,16,18]. In our trial, not
even inexperienced (first-time) owners had problems
with their dog (e.g. dog #7) despite our knowing that this
situation could present a risk. In our study, some families
had children or teenagers and, even if children are often
victims of aggressive dog behavior associated with bit-
ing [1,11], not one of the adopted dog showed aggressive
behavior towards humans and particularly towards chil-
dren. In other studies male dogs have been ranked as
more dangerous than females by their owners [8], but
this occurrence was denied by our results as well as by
Kobelt et al. [12].
We believe that the caregiver’s expertise (in this case,
Operator 1) can make the difference between a success-
ful adoption or not, and that the adopter’s training
pre-ownership is mandatory. Moreover, if the worries
about dog aggression towards humans and particularly
children make sometimes difficult for people the appro-
priate breed and age choice, the possibility to talk with a
professional expert can drive away doubts and fears, for
example by overcoming the damaging concept that a
fully developed dog will not be suitable for adoption as
much that the behavior that a puppy will exhibit at the
adult age is unpredictable. It appears however actually
clear that there is a constant, lively demand of reliable
methods to easily and successfully pair humans and dogs,
for the need to adopt consistently behaving dogs. The
Ethotest-R method could be a good example of dog se-
lection, provided that the Operator selecting and ranking
the animals is an authoritative expert of theoretical and
practical management of dogs. This professional rule
could be equivalent to that of a veterinary technician or
to a companion animals’ ethologist. These individuals
should absolutely know the animals they want to test, i.e.
they could not be people employed for food administra-
tion and pen cleaning. They also need to know the theo-
retical and practical aspects of inter-dog and dog-human
interaction. In our case, the Operator was a doctor in
Animal Welfare and Protection, with a three-year degree
that is consistent with the veterinary technician role of
other European and American countries. Obviously, the
possibility to have an expert integrated in the shelter
staff can shorten Ethotest-R since the uselessness to
carry out the Test A.
The prejudice existing against mature dogs makes the
adoption from shelters more difficult. How to persuade
people to choose an unattractive, sometimes depressed,
adult sheltered dog? What strategy could tip the scales in
favor of a new trend in adoptions? How to offer to those
unwanted dog the possibility to a better life, far from
these prisons? We think that different strategies should
be undertaken:
to enact regional laws forcing the owners to report
the birth of puppies from their bitches: in our country,
the relinquishment of dogs adopted from shelter is
not as frequent as expected. Any dog adopted from a
shelter can, in fact, be traced by his/her microchips.
On the contrary, it is extremely difficult to afford the
problem of abandoned dogs born from family’s
bitches; in some cases, they are abandoned far away
or given as a present. Frequently puppies are appre-
ciated until they became a demanding task but, when
they no longer suit their owners’ needs, they are re-
linquished without any remorse;
to give those shelter dogs a challenging, interesting
environment to live; we know, indeed, how the dev-
astating, depressing environment, in which aban-
doned or stray dogs are forced to live can perma-
nently invalidate their temperament [25], when it
lacks the appropriate physical, psychological and
human enrichment;
to give sheltered dogs an economical value, which
can prevent the recurrence to the shelter; (a no-cost
dog can be easily replaced by other no-cost dogs);
to accomplish a basal training by a qualified behav-
ioral caretaker that could help the dog to become
more attractive or, at least, to dispose of a easy,
pratical Test to perform a safe selection.
Moreover, it should be stressed that customer satisfac-
tion relies not only on a distant Test; the human factor
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/OJAS/
(knowledge and practice) is always behind it and it
seems to have become more and more indispensable to
efficaciously pair humans with dogs.
The Authors wish to thank the Veterinary Department of L’Aquila
ASL 04 for the logistic support, Prof. Francesca Rosati (University of
Teramo) for editing of this paper and Dr. Nicola Bernabò for his com-
ments. The work was supported by the University of Teramo, financial
support ex 60% 2009.
[1] American Veterinary Medical Association Task Force on
Canine Aggression and Human-Canine Interaction. (2001)
A community approach to dog bite prevention. Journal of
American Veterinary Medical Association, 218, 1732-
1749. doi:10.2460/javma.2001.218.1732
[2] Bollen, K. S. and Horowitz, J. (2008) Behavioral evalua-
tion and demographic information in the assessment of
aggressiveness in shelter dogs. Applied Animal Behav-
iour Science, 11 2, 120-135.
[3] Christensen, E., Scarlett, J., Campagna, M. and Houpt,
K.A. (2007) Aggressive behavior in adopted dogs that
passed a temperament test. Applied Animal Behaviour
Science, 106, 85-95. doi:10.1016/j.applanim.2006.07.002
[4] Clevenger, J. and Kass, P. H. (2003) Determinants of
adoption and euthanasia of shelter dogs spayed or neu-
tered in the University of California veterinary students
surgery program compared to other shelter dogs. Journal
of Veterinary Medical Education, 30, 372-378.
[5] Diesel, G., Pfeiffer, D.U. and Brodbelt, D. (2006) Factors
affecting the success of rehoming dogs in the UK in 2005.
Preventive Veterinary Medicine, 84, 228-241.
[6] Fatjó, J., Ruiz-de-la-Torre, J.L. and Manteca, X. (2006)
The epidemiology of behavioral problems in dogs and
cats: A survey of veterinary practitioners. Animal Welfare,
15, 179-185.
[7] Freedman, D.J., King, J.A. and Elliot, O. (1961) Critical
period in the social development of dogs. Science, 133,
1016-1017. doi:10.1126/science.133.3457.1016
[8] Guy, N.C., Luescher, A.U., Dohoo, S.E., Spangler, E.,
Miller, J.B., Dohoo, I.R. and Bate, L.A. (2001) Demo-
graphic and aggressive characteristics of dogs in a gen-
eral veterinary caseload. Animal Behaviour Science 74,
[9] Hsu, Y. and Serpell, J. A. (2003) Development and vali-
dation of a questionnaire for measuring behavior and
temperament traits in pet dogs. Journal of American Vet-
erinary Medical Association, 23, 1293-1300.
[10] Jones, A.C. and Goslin, S.D. (2005) Temperament and
personality in dogs (Canis familiaris): A review and
evaluation of past research. Applied Animal Behaviour
Science, 95, 1-53. doi:10.1016/j.applanim.2005.04.008
[11] Kahn, A., Bauche, P. and Lamoureux, J. (2003) Child
victims of dog bites treated in emergency departments: A
prospective survey. European Journal of Pediatrics, 162,
[12] Kobelt, A.J., Hemsworth, P.H., Barnett, J.L. and Cole-
man, G.J. (2003) A survey of dog ownership in suburban
Australia—conditions and behaviour problems. Applied
Animal Behaviour Science, 82, 137-148.
[13] Lucidi, P., Bernabò, N., Panunzi, M., Dalla Villa, P. and
Mattioli, M. (2005) Ethotest: A new model to identify
(shelter) dogs’ skills as service animals or adoptable pets.
Applied Anima l Behaviour Science, 95, 103-122.
[14] Luescher, A.U. and Medlock, R.T. (2009) The effects of
training and environmental alterations on adoption suc-
cess of shelter dogs. Applied Animal Behaviour Science,
117, 63-68. doi:10.1016/j.applanim.2008.11.001
[15] Marston, L. C. and Bennet, P. C. (2003) Renforcing the
bond-towards successful canine adoption. Applied Ani-
mal Behaviour Science, 83, 227-245.
[16] Marston, L. C., Bennet, P. C. and Coleman, G. J. (2005)
What happens to shelter dogs? An analysis of data for 1
year from three australian shelters. Journal of Applied
Animal Welfare Science, 7, 27-47.
[17] Mondelli, F., Prato Previde, E., Verga, M., Levi, D.,
Magistrelli, S. and Valsecchi, P. (2004) The bond that
never developed: Adoption and relinquishment of dogs in
a rescue shelter. Journal of Applied Animal Welfare Sci-
ence, 7, 253-266. doi:10.1207/s15327604jaws0704_3
[18] Patronek, G. J., Glickman, L. T. and Moyer, M. R. (1995)
Population dynamics and the risk of euthanasia for dogs
in an animal shelter. Anthrozoos, 1, 31-43.
[19] Segurson, S. A., Serpell, J. A. and Hart, B. L. (2005)
Evaluation of a behavioral assessment questionnaire for
use in the characterization of behavioral problems of
dogs relinquished to animal shelters. Journal of Ameri-
can Veterinary Medical Association, 227, 1755-1761.
[20] Selby, L.A., Rhoades, J.D., Hewett, J.E. and Irvin, J.A.
(1979) A survey of attitude towards responsible pet own-
ership. Public Health Report, 94, 380-386.
[21] Shore, E.R. (2005) Returning a recently adopted com-
panion animal: Adopter reasons for and reactions to the
failed adoption experience. Journal of Applied Animal
Welfare Science, 8, 187-198.
[22] Svartberg, K. (2005) A comparison of behavior in test
and in everyday life: Evidence of three consistent bold-
ness-related personality traits in dogs. Applied Animal
Behaviour Science, 91, 103-128.
[23] Taylor, K.D. and Mills, D.S. (2006) The development
and assessment of temperament tests for adult companion
dogs. Journal of Veterinary Behavior, 1, 94-108.
[24] Thorn, J., Templeton, K., Van Winkle, M. and Castillo, R.
(2006) Conditioning shelter dogs to sit. Journal of Ap-
plied Animal We lfare Science, 9, 25-39.
M. Sticco et al. / Open Journal of Animal Sciences 1 (2011) 135-144
Copyright © 2011 SciRes. http://www.scirp.org/journal/OJAS/Openly accessible at
[25] Tuber, D. S., Miller, D. D., Caris, K. A., Halter, R., Lin-
den F. and Hennessy, M. B. (1999) Dogs in animal shel-
ters: Problems, suggestions, and needed expertise. Psy-
chological Science, 10, 379-386.
[26] Wells, D. L. and Hepper, P. G. (2000) Prevalence of be-
havior problems reported by owners of dogs purchased
from an animal rescue shelter. Applied Animal Behaviour
Science, 69, 55-65. doi:10.1016/S0168-1591(00)00118-0