oods
A icle
In es iga ing I alian Consume P e e ences o
Di e en Cha ac e is ics o P o olone Valpadana
Using he Conjoin Analysis App oach
Niculina Iudi a Sampalean 1, Tiziana de-Magis is 2,3,* and Daniele Rama 1
1Depa men o Ag i-Food Economics, Facul y o Ag icul u e, Food and En i onmen al Sciences,
Campus Piacenza-C emona, Ca holic Uni e si y o Sac ed Hea , ia Emilia Pa mense, 84,
29122 Piacenza (PC), I aly; [email p o ec ed] (N.I.S.); [email p o ec ed] (D.R.)
2Unidad de Economía Ag oalimen a ia y de los Recu sos Na u ales, Cen o de In es igación y Tecnología,
Ag oalimen a ia de A agón, 50059 Za agoza, Spain
3Ins i u o Ag oalimen a io de A agón (IA2), CITA-Uni e sidad de Za agoza, 50013 Za agoza, Spain
*Co espondence: [email p o ec ed]; Tel.: +34-976-71-6352
Recei ed: 29 Oc obe 2020; Accep ed: 23 No embe 2020; Published: 25 No embe 2020


Abs ac :
The objec i e o his pape was wo old. Fi s , we es ima ed consume p e e ences o an
I alian cheese (P o olone Valpadana) wi h espec o se e al a ibu es and le els, such as p ice, o igin
ce i ica ion, p oduc ion sys em, ‘ ee om’ labelling, and b and. Second, we iden i ied consume
clus e s wi h simila p e e ences o a ious cheese cha ac e is ics. P e e ences we e es ima ed using
he conjoin analysis me hod. Then, a clus e analysis was used o classi y consume s in o di e en
( h ee) clus e s ollowed by a ma ke simula ion. In all h ee clus e s, he a ibu e mos p e e ed
by I alian consume s was he b and o he cheese: consume s p e e ed o pu chase P o olone
cheese ha ing he lowes p ice, p oduced by Au icchio, bea ing a Eu opean Union (EU) quali y
ce i ica ion, p oduced o ganically, and non-lac ose- ee. The esul s o ou s udy p o ide help ul
in o ma ion o ood companies o be e segmen ing hei ma ke and a ge ing hei consume s,
as well as e ec i ely p omo ing hei p oduc s using b ands, ce i ica ions as o ganic and lac ose- ee.
This s udy con ibu es o he li e a u e on consume p e e ence o he EU labelling scheme ( olun a y
and manda o y). To ou knowledge, his is he i s s udy o in es iga e his combina ion o mul iple
labels displayed on he on o I alian cheese packaging.
Keywo ds: P o olone Valpadana; cheese; conjoin analysis; clus e analysis; ma ke sha es
1. In oduc ion
Consume s ha e become highly conscious o ood p oduc cha ac e is ics when shopping.
They sea ch o p oduc s o highe quali y made wi h en i onmen ally iendly p oduc ion p ocesses
ha ha e connec ions wi h he e i o y and ha p e en heal h issues [
1
–
4
]. To keep up wi h his
change in consume p e e ences, a g ea e ange o p oduc s ha e eme ged on he ma ke . As a esul ,
his abundan a ie y o p oduc s has made ood shopping a mo e di icul ask o consume s [5].
Some ood-labelling schemes enable manu ac u e s o disclose he quali ies o ood p oduc s o
consume s [
6
]. Thus, he use o labels allows ood companies o signal quali y o he p esence o speci ic
desi able bu no ob ious a ibu es [
7
]. In his con ex , du ing he pas wo decades, he Eu opean
Union (EU) has launched di e en ood labelling egula ions ha ood companies can adop on a
olun a y basis. The aim o hese ood-labelling schemes is o es ablish he equi emen s o he
use o pa icula labels and he p ocedu es o con olling he quali y o labelled p oduc s o p o ec
consume s om being misled. Two examples o hese egula ions a e EEC Regula ion No. 834/2007,
conce ning o ganic p oduc ion and he labelling o o ganic p oduc s, and Regula ion No. 2081/92
Foods 2020,9, 1730; doi:10.3390/ oods9121730 www.mdpi.com/jou nal/ oods
Foods 2020,9, 1730 2 o 16
adop ed by he Eu opean Commission and u he amended wi h se e al ex s which culmina ed in
EEC Regula ion No. 1151/2012, which in oduced he quali y labels P o ec ed Designa ion o O igin
(PDO) and P o ec ed Geog aphical Indica ion (PGI) labels. The PDO and PGI labels we e designed as
quali y cues o educe asymme ic in o ma ion and o educe consume unce ain y associa ed wi h
ood pu chases ega ding desi able p oduc cha ac e is ics. In pa icula , he PDO label may be used
in he case o ood p oduc s ha a e p oduced, p ocessed, and p epa ed in a gi en geog aphical a ea
using local know-how. (EEC Regula ion No. 1151/2012).
Many s udies on consume p e e ences o ood labelling ha e been conduc ed since he 2000s,
and hey ha e especially ocused on consume willingness o pay (WTP) o mul iple on -o -package
(FOP) ood labelling schemes [1,3,8–10] and some ha e ocused on cheese p oduc s [11–13].
Al hough cheese is one o he mos impo an ag icul u al ood p oduc s in I aly, accoun ing o
87.4% o he o al alue o I alian dai y p oduc s sold ab oad [
14
], s udies on consume p e e ences
o ood labelling on cheese p oduc s a e sca ce [
12
,
15
–
17
]. In gene al, hese s udies ha e epo ed
ha PDO labels a e p e e ed o e o ganic labelling; nu i ional claims a e p e e ed o e o ganic
labelling; nu i ional claims a e p e e ed o e he nu i ional ac panel. Ye , mos s udies ha e
also demons a ed ha p e e ences o ood labelling schemes a e he e ogeneous ac oss consume s,
wi h gende , age, educa ional le el, and ha he deg ee o en i onmen al conce n in luenced consume
WTP o di e en cheeses [11,18].
A he Eu opean le el, 53 o 247 ypes o cheese designa ed as PDO o PGI a e I alian
p oduced. Fo I aly, hese cheeses ep esen 57% o he na ional p oduc ion alue and 50% o
expo s.
Likewise, in 2018,
sales o Geog aphical Indica ion (GI) cheeses g ew by 2% in quan i y and
1.5% in alue compa ed wi h non-GI cheeses; hose cheeses dec eased by 1.6% in olume and inc eased
by 0.6% in alue [
19
]. In his con ex , G ana Padano PDO and Pa migiano Reggiano PDO oge he
accoun o 59.3% o he olume and 66.3% o he alue o o al sales o GI cheeses in la ge-scale
dis ibu ion [19].
Fou egions o no he n I aly (Lomba dy included) accoun o 65% o he PGI p oduc ion alue;
76 PDO/PGI ce i ied p oduc s come om Lomba dy, and 9896 PDO/PGI en e p ises a e loca ed in he
egion, gene a ing a p oduc ion alue o €1958 million [19].
P o olone Valpadana is one o he mos consumed GI cheeses in I aly, wi h a consump ion alue
o
€
77 million and ep esen ing 1.1%, by weigh , o he o al I alian GI cheese consump ion, mo e han
he well-known Peco ino Toscano PDO o Mon asio PDO [19].
Is a ’s ‘Repo on Family Consump ion Expense’ shows ha he a e age mon hly household
expendi u e in I aly in 2017 was
€
2563.94. O his o al a e age expendi u e,
€
457.12 was dedica ed o
ood and non-alcoholic d inks. Spending on cheese, milk, and eggs was
€
58.26 ( hese a e measu ed as
an agg ega e), which ep esen s 2.27% o he o al a e age mon hly household expendi u e and 12.75%
o he o al a e age mon hly expendi u e o ood and non-alcoholic d inks. P o olone is a spun-pas e
semiha d cheese made om whole cows’ milk wi h na u al acidi y om e men a ion. Milk is collec ed
in he a ea o o igin wi hin 60 h o milking and hen unde goes mild hea ea men un il he cheese
is pas eu ised. The ma u a ion pe iod can a y as ollows: o up o 6 kg, he minimum ma u a ion
pe iod is 10 days; o o e 6 kg, he minimum ma u a ion pe iod is 30 days; o o e 15 kg o only he
piquan a ie y, he minimum ma u a ion pe iod is 90 days; o o e 30 kg o he p oduc labelled
PVS (P o olone Valpadana S ong) he ma u a ion pe iod is o e 8 mon hs ( o he piquan a ie y).
The cheese may also be smoked, o med in o di e en shapes (one o i s mos impo an cha ac e is ics)
and be o med in o blocks o a ious weigh [20].
P o olone Valpadana was ecognized as a PDO p oduc in 1996. Since hen, all i s p oduce s mus
s ic ly comply wi h he p oduc ion egula ions. Today, P o olone Valpadana PDO is p oduced by
au ho ised cheesemake s and can also be sold by packaging companies. In all cases, e e y single shape
mus be ma ked wi h he Conso ium’s adema k and he PDO adema k, which a e unequi ocal
gua an ees o quali y [21].
Foods 2020,9, 1730 3 o 16
P o olone Valpadana is a ypical p emium cheese wi h unmis akable quali ies; i is easy o
compa e wi h o he gene ic p oduc s (e.g., spun-pas e cheese), bu consume s mus ecognise i s
unique quali ies.
In ligh o he o egoing con ex , he objec i e o his s udy was wo old. The i s objec i e was o
assess I alian consume p e e ences o P o olone Valpadana ha bea s mul iple ood labels, such as
hose making claims ela ed o hei o igin (PDO), me hod o p oduc ion (o ganic), and heal hiness
( ee om lac ose). Because o he inc easing complexi y o consume p e e ence, he second aim o
his s udy was o examine he e ogenei y in p e e ences o labelled cheeses on he basis o consume s’
sociodemog aphic and pe sonal cha ac e is ics.
To achie e hese objec i es, we conduc ed an online su ey o 246 I alian cheese consume s and
used bo h a conjoin and clus e analysis o analyse ou da a.
Ou s udy con ibu es o he consume cheese p e e ence li e a u e because his is he i s
s udy o in es iga e consume p e e ences o hese mul iple ood labels on I alian cheese using an
app oach combining conjoin analysis and clus e analysis. Fu he mo e, ou indings ha e ma ke ing
implica ions because hey disclose use ul in o ma ion o I alian s akeholde s along wi h he cheese
supply chain ha can help p oduce s and e aile s, among o he s, design he bes s a egies o
di e en ia e cheeses by adop ing a ious labelling schemes o di e en segmen s o consume s.
Li e a u e Re iew
Consume p e e ences owa ds di e en ood labels and di e en combina ions o ood
labels ha e been widely in es iga ed du ing he las decade o se e al ood p oduc s.
Fo ins ance,
Be nabeu e al. [22].
analysed he p ice, o igin, ype and p oduc ion sys em o wine,
while
E aach e al. [9]
examined he ela i e impo ance o Eu opean o igin labels (PDO) associa ed
wi h ex insic (p ice and packaging) and in insic (colo ) cues o Spanish oli e oil on consume
p e e ences. Ap ile e al. [
8
] p o ided e idence abou ex a- i gin oli e oil consume s’ WTP. Likewise,
de Magis is and G acia [
23
] in es iga ed consume s’ p e e ences and WTP o almonds ca ying
di e en sus ainabili y labels (e.g., ood miles and o ganic ce i ica ions). Simila ly, Mesias e al. [
24
]
used p ice, o igin, p oduc ion sys em and labelling as a ibu es o s udy p e e ences o bee in
Ex emadu a. Finally, Cilla e al. [
25
] and Ga a aglia and Ma iani [
26
] in es iga ed consume s’
p e e ences o cu ed ham in Spain.
These s udies in gene al showed ha he PDO and o ganic a ming labels a e mos impo an .
In pa icula , Ap ile e al. [
8
] showed ha he highes p emium p ice is associa ed wi h PDO labels,
ollowed by o ganic a ming labels, and hen PGI labels. E aach e al. [
9
] indica ed ha p ice and
o igin labelling (PDO label) we e he a ibu es ha mos a ec consume s’ p e e ences. De Magis is
and G acia [
23
] sugges ed ha consume s we e willing o pay a posi i e p ice p emium o ‘locally
g own’ and ‘o ganically p oduced’ a ibu es. Finally, Mesias e al. [
24
] showed ha he o igin o he
p oduc is he mos impo an a ibu e in he choice o bee , ollowed by quali y labelling, p oduc ion
sys em and p ice. In he li e a u e, some s udies ha e ocused on consume p e e ences o se e al ood
labels appea ing on cheeses sold in Eu ope. Fo example, Ga a aglia and Ma coz [
12
] in es iga ed he
Fon ina cheese bea ing he PDO ce i ica ion. The au ho s epo ed di e ences in consume p e e ences
depending on he esponden ’s place o esidence. Likewise, Skubic e al. [
13
] conduc ed a s udy in
Slo enia and ound ha p ice was he mos impo an a ibu e o consume s. De Magis is and
G acia [
27
] ound ha Spanish consume s we e willing o pay mo e o PDO cheese han o o ganic o
ligh cheeses. Mo eo e , gende , age, educa ional le el, and he deg ee o he en i onmen al conce n
in luenced consume WTP o di e en cheeses.
Se e al s udies ha e in es iga ed consume p e e ences o cheese p oduc s based on p oduc ion,
p ocessing, ading o b anding. Fo example, Tende o and Be nabeu [
28
] demons a ed ha consume s
mos highly alue he PDO label as a gua an ee o quali y and ood sa e y and ha hey p e e cheeses
ha a e app op ia ely p iced, aged, and i possible, ce i ied. Simila ly, Be nabeu e al. [
29
] in es iga ed
he pe cei ed quali y o cheese om Cas illa-La-Mancha compa ed wi h ha o cheeses om he es
Foods 2020,9, 1730 4 o 16
o Spain and o eign cheeses. The au ho s ound ha consume s a ached he mos impo ance o he
o igin o he cheese, ollowed by he cheese ype, p ice, and p oduc ion sys em (o ganic). Mo eo e ,
Be nabeu e al. [
29
] e ealed ha o ganic p oduc ion sys ems had he lowes ela i e impo ance as
well as a nega i e in luence on consume p e e ences o cheese.
Napoli ano e al. [
17
] assessed he e ec o in o ma ion conce ning o ganic p oduc ion o Peco ino
cheese on consume WTP. They showed ha consume s we e willing o pay a highe p ice han local
e ail p ice o o ganic cheese han egula cheese (
€
3.00/100 g). Mo eo e , in o ma ion ega ding
o ganic a ming was ound o be he main de e minan o cheese p e e ence and consume WTP.
Some s udies ha e ocused on consume WTP o cheeses wi h FOP nu i ional claims. Fo ins ance,
G acia and de Magis is [
3
] de e mined ha he mos p e e ed ood label was he PDO indica ion,
closely ollowed by he nu i ional ac panel and he EU o ganic logo. Likewise, de Magis is and
Lopez-Galan [
11
]. In es iga ed consume WTP o cheeses bea ing ‘ educed a ’ and ‘low sal ’ claims
in Spain. The au ho s epo ed ha Spanish cheese consume s we e willing o pay a p emium o
packages o cheese wi h educed- a and low-sal claims.
Ano he impo an claim ela ed o he heal h o consume s ha may be p esen on he FOP
o cheese is he ‘lac ose- ee’ claim. This claim con ains impo an in o ma ion o consume s wi h
ood in ole ances (e.g., lac ose in ole ance). Ha mann e al. [
30
] in es iga ed how ‘ ee- om labelling’
shapes consume pe cep ions o ood p oduc s and whe he he absence o an ing edien is conside ed
an indica o o imp o ed nu i ional alue o he p oduc . The au ho s concluded ha a ‘lac ose- ee’
label was highly co ela ed wi h he pe cep ion o heal hiness as well as he in en ion o pay a p emium
p ice, imagining ha consume s a ibu e inapp op ia e heal h bene i s o p oduc s ee o lac ose.
Lac ose- ee cheese occu s when cheese is aged mo e han abou one mon h, as bac e ia added om
he beginning o he milk ‘diges ’ lac ose con e ing i in o lac ic acid; he la e con ibu es o cheese
as e and a oma, and i s o all i induces p o ein coagula ion, he e o e he cheese p oduc ion [31].
Finally, s udies on consume p e e ence o p i a e b ands o cheese a e sca ce. To ou knowledge,
only A ini [
15
] conduc ed such a s udy, compa ing consume WTP o cheeses bea ing Conso ium
labels and EU PDO labels. The indings showed ha he p esence o he manu ac u e ’s b and was
compa a i ely impo an when shopping o g a ed cheese (18.91%) and o dina y cheese (18.24%).
The esul s u he indica ed ha li le a en ion was paid o he p i a e label o he i m p oducing o
ma ke ing he ypical p oduc s a issue; howe e , he Conso ium label played an impo an ole in
eassu ing consume s abou he quali y o he pu chased p oduc . Fo Pa migiano Reggiano cheese,
as many as 75% o he in e iewed consume s looked o he Conso ium label when shopping [15].
2. Ma e ials and Me hods
2.1. Conjoin Analysis and Choice Task P ocedu e
Conjoin analysis is one o he mos widely used ma ke ing esea ch me hods o analysing
consume ade-o s and is used o examine su ey esponses conce ning p e e ences and in en ions o
buy; i is a me hod o simula ing how consume s migh eac o changes in cu en p oduc s o new
p oduc s in oduced in o an exis en compe i i e a ay [
32
]. The model assumes ha he subs i u e
p oduc s no ion can be de ined as a se ies o speci ic le els o a common se o a ibu es. I also
assumes ha he o al u ili y he consume de i es om a p oduc is de e mined by he u ili y o
pa -wo h con ibu ed by each a ibu e le el. The conjoin analysis s a s wi h he consume ’s o e all
o global judgemen s abou a se o complex a ibu es. I hen pe o ms a decomposi ion o he o iginal
e alua ions o he consume in o sepa a e and compa ible u ili y scales, acco ding o which he o iginal
o e all judgemen s can be econs i u ed [33].
Foods 2020,9, 1730 5 o 16
The mos commonly used composi ion ule adop ed in his s udy is addi i e because i
is he one conside ing mos (80% o 90%) o he a ia ion in p e e ence in almos all cases.
Following Mesias e al., 2005 [24] we used he addi i e composi ion ule:
Y=
n
X
i=1
Xij X
j=1
ij
whe e Y is he o al u ili y, Vij is he u ili y associa ed wi h le el j (j =1, 2
. . .
, m) o a ibu e i (i =1, 2
. . .
, n), and Xij is a dummy a iable ha akes he alue o 1 (o 0) in he case o p esence (o absence)
o he j h le el o he i h a ibu e.
Fo he quali a i e a ibu es, he pa -wo h model was used because o i s lexibili y [
31
]. Fo he
p ice a ibu e, a Linea Less ela ionship was used because in gene al, highe p ices co espond o
lowe u ili y o p e e ences.
He e, he conjoin analysis (CA) p ocedu e consis ed o a choice ask o es ablish he ade-o s
ha I alian consume s make be ween p ice, quali y ce i ica ion (PDO), sys em p oduc ion (o ganic),
‘ ee om’ labelling, and b and o P o olone Valpadana Dolce.
I alian consume s we e p esen ed 16 a ie ies o P o olone Valpadana Dolce wi h di e en
combina ions o a ibu es le els. The a ibu es (and hei le els) used in he s udy a e lis ed in
Table 1and we e iden i ied (as mos impo an when buying cheese: p ice (
€
2.20/200 g,
€
2.86/200 g,
and
€
3.20/200 g), quali y ce i ica ion (wi h o wi hou PDO), p oduc ion sys em (o ganic o
con en ional), b and (Au icchio, La e ia So esina (bo h na ional b ands), Coop (p i a e label),
o no b and) and lac ose con en (lac ose- ee o non-lac ose- ee). To selec he p ice le els, in o ma ion
on he P o olone Valpadana sold in di e en I alian supe ma ke s was used, and h ee p ice le els
we e se up (
€
2.20/200 g,
€
2.86/200 g, and
€
3.20/200 g). The mos equen lowes p ice o a package
o 200 g was a ound
€
2.20. The e o e, his p ice was chosen as he lowes p ice. The second le el
was 30% highe han he mos equen lowes p ice, and he hi d le el was 45% highe han he
mos equen lowes p ice. Mo eo e , he Coop p i a e label was chosen since i could ep esen any
I alian p i a e label o P o olone Valpadana. As o he na ional b ands, wo b ands we e selec ed:
Au icchio, he leade
b and [
34
], and La e ia So esina, ano he na ional b and o P o olone Valpadana
cheese. In o he wo ds, any o he p i a e label and na ional b and could ha e been used ins ead o
COOP and La e ia So esina.
Table 1. A ibu es and hei co esponding le els.
A ibu es Le el
P ice
2.20 €
2.86 €
3.20 €
Eu opean Quali y Ce i ica ion
(P o ec ed Designa ion o O igin)
Yes
No
P oduc ion Sys em O ganic
Con en ional
Lac ose con en Con ains Lac ose
Lac ose-F ee
B and
La e ia So esina
Au icchio
Coop
No B and
Wi h hese i e a ibu es and hei 13 le els, 96 po en ial p o iles we e ob ained. Because his is
a pa icula ly high numbe o show o consume s, an o hogonal design (SPSS) was applied o he
p o iles o educe he numbe o 16, a numbe mo e likely o encou age pa icipa ion o consume s.

Foods 2020,9, 1730 6 o 16
An example o one o he 16 ob ained p o iles is shown in Figu e 1.
Foods 2020,9, 0 6 o 16
An example o one o he 16 ob ained p o iles is shown in Figu e 1.
Foods 2020, 9, x FOR PEER REVIEW 6 o 16
The esponden s e alua ed a pic u e ha con ained he di e en a ibu e le els in his s udy.
The packaging used was he same as ha on he ma ke , bu some a ibu e-le el combina ions shown
a e no p esen in he ma ke place. Once he p o ile’s ca ds we e o med, hey we e shown o he
esponden s who e alua ed hem, gi ing a sco e om 1 o 10 acco ding o hei s a ed p e e ence.
Pa icipan s had he op ion o epea he same numbe on di e en ca ds. The numbe 1 and 10
co esponded o he lowes and highes le els o p e e ence, espec i ely ( hus employing he
comple e p o ile me hod). The main a gumen in suppo o he ull-p o ile app oach is ha i gi es
a mo e ealis ic desc ip ion o s imuli by de ining he le els o each o he a ibu es and conside ing
he po en ial en i onmen al co ela ion be ween ac o s in eal s imuli [35].
Figu e 1. Example o e alua ed cheese p o ile.
Using SPSS so wa e, he u ili y es ima es gene a ed by he conjoin analysis we e used in a k-
means clus e analysis o classi y he consume s in o homogeneous p e e ences g oups and classi y
hem in o clus e s
2.2. Da a Ga he ing
A sel -adminis e ed s uc u ed elec onic su ey was comple ed by 245 people in I aly. The
su ey was dis ibu ed online (Facebook, Messenge , Wha sApp, LinkedIn) be ween Sep embe 2019
and Ma ch 2020. All subjec s ga e hei in o med consen o inclusion be o e hey pa icipa ed in he
s udy. The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki, and he p o ocol was
app o ed by he E hics Commi ee o Ca holic’s Uni e si y o he Sac ed Hea . A pilo s udy wi h
Valpadana cheese consume s was conduc ed o es he alidi y and he comp ehension o he su ey.
Table 2 shows he cha ac e is ics o he inal sample o esponden s.
Table 2. Sociodemog aphic dis ibu ion o he collec ed sample in compa ison wi h he I alian
popula ion (%).
Va iables Le els Sample Popula ion
Age (in yea s) 18–35 24
36–55 53
>55 23
Gende Female 81 51 *
Male 19 49 *
Educa ion Elemen a y and middle school 2
High school Diploma 41
Figu e 1. Example o e alua ed cheese p o ile.
The esponden s e alua ed a pic u e ha con ained he di e en a ibu e le els in his s udy.
The packaging used was he same as ha on he ma ke , bu some a ibu e-le el combina ions shown
a e no p esen in he ma ke place. Once he p o ile’s ca ds we e o med, hey we e shown o he
esponden s who e alua ed hem, gi ing a sco e om 1 o 10 acco ding o hei s a ed p e e ence.
Pa icipan s had he op ion o epea he same numbe on di e en ca ds. The numbe 1 and 10
co esponded o he lowes and highes le els o p e e ence, espec i ely ( hus employing he comple e
p o ile me hod). The main a gumen in suppo o he ull-p o ile app oach is ha i gi es a mo e
ealis ic desc ip ion o s imuli by de ining he le els o each o he a ibu es and conside ing he
po en ial en i onmen al co ela ion be ween ac o s in eal s imuli [35].
Using SPSS so wa e, he u ili y es ima es gene a ed by he conjoin analysis we e used in a
k-means clus e analysis o classi y he consume s in o homogeneous p e e ences g oups and classi y
hem in o clus e s.
2.2. Da a Ga he ing
A sel -adminis e ed s uc u ed elec onic su ey was comple ed by 245 people in I aly. The su ey
was dis ibu ed online (Facebook, Messenge , Wha sApp, LinkedIn) be ween Sep embe 2019 and
Ma ch 2020. All subjec s ga e hei in o med consen o inclusion be o e hey pa icipa ed in he
s udy. The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki, and he p o ocol was
app o ed by he E hics Commi ee o Ca holic’s Uni e si y o he Sac ed Hea . A pilo s udy wi h
Valpadana cheese consume s was conduc ed o es he alidi y and he comp ehension o he su ey.
Table 2shows he cha ac e is ics o he inal sample o esponden s.
App oxima ely 81% o he esponden s we e women and only 19% we e men. Compa ed wi h
he I alian popula ion, women we e o e ep esen ed, which migh be explained by he ac ha , in
gene al, women o e see ood shopping o households. Roughly 24% o he pa icipan s we e less han
35 yea s old, 53% we e be ween 36 and 55 yea s old, and 23% we e olde han 55 yea s. The sample
may ha e skewed owa d a younge demog aphic because he ques ionnai e was dis ibu ed online,
and he olde popula ion ends o no ha e access o he In e ne o compu e skills. Roughly 43% o
he pa icipan s had comple ed high school deg ees, and 57% had a uni e si y deg ee. Mo e han hal
we e ma ied (53%), which is simila o he I alian popula ion o e all (47%). In mo e han 40% o ou
sample, he amily was composed o h ee o ou membe s, which is also compa able wi h he I alian
popula ion o e all (35%). Mo e han hal he esponden s (66%) we e om sou he n o island egions,
wi h he cen al egion being he leas ep esen ed e i o y (7%) o pa icipan s compa ed wi h 20%
o he I alian popula ion. Fo pa icipan occupa ions, mos (40%) we e o ice wo ke s, ollowed by
housewi es (14%), eelance s (13%), s uden s o unemployed pe sons (p esen in he same p opo ion
Figu e 1. Example o e alua ed cheese p o ile.
The esponden s e alua ed a pic u e ha con ained he di e en a ibu e le els in his s udy.
The packaging used was he same as ha on he ma ke , bu some a ibu e-le el combina ions shown
a e no p esen in he ma ke place. Once he p o ile’s ca ds we e o med, hey we e shown o he
esponden s who e alua ed hem, gi ing a sco e om 1 o 10 acco ding o hei s a ed p e e ence.
Pa icipan s had he op ion o epea he same numbe on di e en ca ds. The numbe 1 and 10
co esponded o he lowes and highes le els o p e e ence, espec i ely ( hus employing he comple e
p o ile me hod). The main a gumen in suppo o he ull-p o ile app oach is ha i gi es a mo e
ealis ic desc ip ion o s imuli by de ining he le els o each o he a ibu es and conside ing he
po en ial en i onmen al co ela ion be ween ac o s in eal s imuli [35].
Using SPSS so wa e, he u ili y es ima es gene a ed by he conjoin analysis we e used in a
k-means clus e analysis o classi y he consume s in o homogeneous p e e ences g oups and classi y
hem in o clus e s.
2.2. Da a Ga he ing
A sel -adminis e ed s uc u ed elec onic su ey was comple ed by 245 people in I aly. The su ey
was dis ibu ed online (Facebook, Messenge , Wha sApp, LinkedIn) be ween Sep embe 2019 and
Ma ch 2020. All subjec s ga e hei in o med consen o inclusion be o e hey pa icipa ed in he
s udy. The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki, and he p o ocol was
app o ed by he E hics Commi ee o Ca holic’s Uni e si y o he Sac ed Hea . A pilo s udy wi h
Valpadana cheese consume s was conduc ed o es he alidi y and he comp ehension o he su ey.
Table 2shows he cha ac e is ics o he inal sample o esponden s.
App oxima ely 81% o he esponden s we e women and only 19% we e men. Compa ed wi h
he I alian popula ion, women we e o e ep esen ed, which migh be explained by he ac ha , in
gene al, women o e see ood shopping o households. Roughly 24% o he pa icipan s we e less han
35 yea s old, 53% we e be ween 36 and 55 yea s old, and 23% we e olde han 55 yea s. The sample
may ha e skewed owa d a younge demog aphic because he ques ionnai e was dis ibu ed online,
and he olde popula ion ends o no ha e access o he In e ne o compu e skills. Roughly 43% o
he pa icipan s had comple ed high school deg ees, and 57% had a uni e si y deg ee. Mo e han hal
we e ma ied (53%), which is simila o he I alian popula ion o e all (47%). In mo e han 40% o ou
sample, he amily was composed o h ee o ou membe s, which is also compa able wi h he I alian
popula ion o e all (35%). Mo e han hal he esponden s (66%) we e om sou he n o island egions,
wi h he cen al egion being he leas ep esen ed e i o y (7%) o pa icipan s compa ed wi h 20%
o he I alian popula ion. Fo pa icipan occupa ions, mos (40%) we e o ice wo ke s, ollowed by
housewi es (14%), eelance s (13%), s uden s o unemployed pe sons (p esen in he same p opo ion
Foods 2020,9, 1730 7 o 16
o 5%), eache s (2%), and en ep eneu s (2%). Conce ning annual income, 26% o he esponden s
ea ned less han
€
10,000 pe yea , 31% ea ned be ween
€
10,000 and
€
20,000 pe yea , and 11% ea ned
mo e han €40,000 pe yea .
Table 2.
Sociodemog aphic dis ibu ion o he collec ed sample in compa ison wi h he I alian
popula ion (%).
Va iables Le els Sample Popula ion
Age (in yea s) 18–35 24
36–55 53
>55 23
Gende Female 81 51 *
Male 19 49 *
Educa ion
Elemen a y and middle school
2
High school Diploma 41
Uni e si y deg ee 57
Ma i al s a us Bachelo /Maiden 33 42 *
Ma ied 53 47 *
Di o ced 9 3 *
In a ela ionship 3
Widow 1
Family membe s 1 12 33 *
2 36 27 *
3–4 40 35 *
>4 12 5 *
Geog aphical Dis ibu ion No h 27 46 *
Sou h and Islands 66 34 *
Cen e 7 20 *
Occupa ion O ice wo ke 40
F eelance 13
S uden 5
Housewi e 14
Teache 2
En ep eneu 2
Unemployed 5
O he 19
A e age annual income (€)<10.000 26
10,000–20,000 31
20,000–40,000 32
40,000–50,000 4
>50,000 7
* ISTAT (Na ional S a is ics Ins i u e) da a ex ac ed in Ap il 2020. h p://da i.is a .i /Index.aspx?Que yId=18460.
Foods 2020,9, 1730 8 o 16
3. Resul s
The esul s sec ion is di ided in o wo pa s. The i s pa desc ibes he esul s o consume
p e e ences o di e en P o olone Valpadana cheese p o iles and he clus e s o med by hese
p e e ences. The second pa , using he simula ion p ocess, iden i ies ma ke sha es o he mos
p e e ed cheeses as well as o some exis en cheeses on he ma ke .
3.1. Cheese Consume P e e ence S uc u e
Table 3p esen s consume p e e ences o a ious FOP ood labelling. We ound ha b and is
he mos highly alued a ibu e (58.24%), ollowed by p ice (13.89%), p oduc ion sys em (10.65%),
lac ose con en (8.94%), and inally, he quali y ce i ica ion (8.29%). Addi ionally, based on he u ili y
es ima es o each le el o a ibu es, consume s p e e ed o pu chase P o olone cheese ha ing he
lowes p ice, bea ing he Au icchio b and, bea ing he EU quali y ce i ica ion, ha ing been o ganically
p oduced, and non-lac ose- ee.
Table 3. Mean pa -wo h and ela i e impo ance o all esponden s.
A ibu es Le els U ili y Es ima e Impo ance Values (%)
Quali y Ce i ica ion NO PDO −0.077 8.292
PDO 0.077
P oduc ion Sys em O ganic 0.144 10.652
Con en ional −0.144
B and
No B and −0.287
58.239
Au ichio 0.966
Coop −0.378
La e ia So esina −0.301
Lac ose Con en Con ains Lac ose 0.039 8.941
Lac ose-F ee −0.039
P ice
2.2 −0.068
13.877
2.86 −0.136
3.2 −0.204
(Cons an ) 6.723
No es: Pea son’s R =0.992; Kendall’s au =1.000.
Using SPSS so wa e, he u ili y es ima es gene a ed by he conjoin analysis we e used in a
k-means clus e analysis o classi y he consume s in o homogeneous p e e ences g oups and classi y
hem in o clus e s. Th ee clus e s o consume s we e iden i ied wi h di e en p e e ence s uc u es.
The analysis o a iance (ANOVA) showed ha all clus e s di e ed signi ican ly om each o he
wi h espec o p ice (p<0.05), lac ose con en (p<0.1) as well as o he o he a iables we e he
signi icance was he highes (p<0.01).
The u ili y es ima es and hei ela i e impo ance we e hen calcula ed o each o he le els
o he a ibu e. The esul s, oge he wi h he size o each clus e (exp essed as %), a e lis ed in
Table 4(A and B).
In all h ee clus e s, he a ibu e mos p e e ed by he I alian consume s was he b and o he cheese,
and his a ibu e’s highes ela i e impo ance was in clus e 2 (77.16%) ollowed by clus e 1 (62.83%).
Foods 2020,9, 1730 9 o 16
Table 4. A. Mean pa -wo h o he clus e s. B. Rela i e impo ance o he clus e s.
A
A ibu es Le els Clus e 1 (44%) Clus e 2 (22%) Clus e 3 (34%)
P ice 2.20 €−0.14947 −0.01865 −0.08053
2.86 €−0.1943 −0.02424 −0.10469
Eu opean Quali y Ce i ica ion
(P o ec ed Designa ion o O igin)
Yes −0.0089 0.0658 0.2014
No 0.0089 −0.0658 −0.2014
P oduc ion Sys em O ganic 0.0656 0.0668 0.2971
Con en ional −0.0656 −0.0668 −0.2971
Lac ose con en Con ains Lac ose 0.0171 0.0708 0.0519
Lac ose-F ee −0.0171 −0.0708 −0.0519
B and
La e ia So esina −0.3137 −0.8606 0.0878
Au icchio 0.9577 2.7149 −0.1778
Coop −0.5533 −1.0091 0.2683
No b and −0.0900 −0.8440 −0.1800
B
A ibu es Rela i e Impo ance
Clus e 1 Clus e 2 Clus e 3
P ice 13.57 6.49 19.13
Eu opean Quali y Ce i ica ion
(P o ec ed Designa ion o O igin) 6.43 4.53 13.40
P oduc ion Sys em 8.79 5.30 16.67
Lac ose Con en 8.38 6.52 11.23
B and 62.83 77.16 39.57
Rega ding he la ges and i s clus e (44%), he mos p e e ed a ibu e was he b and (62.83%),
ollowed by p ice (13.57%). Howe e , hese consume s equally p e e ed he p oduc ion sys em
and lac ose con en (8.79% and 8.38%), whe eas he quali y ce i ica ion was he leas p e e ed
a ibu e (6.43%).
The second clus e consis ed o 22% o he sample who mos p e e ed he b and o cheese
(77%); lac ose con en and p ice exhibi ed simila ela i e impo ance o p e e ence (6.52% and
6.49%, espec i ely). As in clus e 1, EU quali y ce i ica ion was he leas alued a ibu e (4.53%).
Compa able o he i s and he second clus e , in he hi d clus e ( ep esen ing 34% o all consume s),
he mos p e e ed a ibu e was b and, ollowed by p ice (19.13%); howe e , in con as o he o he
wo clus e s, he lac ose con en (11.23%) was he leas p e e ed a ibu e.
Finally, as expec ed, he le el o p ice in all clus e s exhibi ed a nega i e alue, especially o he
highes p ice (
€
3.20/200 g). Mo eo e , he u ili y assigned o he ‘ ee om lac ose’ le el exhibi ed a
nega i e alue in all clus e s, sugges ing ha consume s did no p e e lac ose- ee cheese.
Consume s in clus e 1 showed a clea p e e ence o cheese wi hou a quali y ce i ica ion, in
con as o clus e s 2 and 3. Likewise, ega ding he b and a ibu e, cheese wi h he Au icchio b and
p esen ed he highes u ili y in he i s and second clus e s bu nega i e u ili y in he hi d clus e ,
in con as wi h La e ia So esina and Coop, bo h o which exhibi ed posi i e u ili y. Thus, he esul s
sugges ha consume s in clus e 1 mos p e e ed he b and Au icchio while in he clus e 3, consume s
mos p e e ed he cheese b and Coop.
Rega ding lac ose con en , in clus e 2, consume s anked he lac ose con en in he cheese as he
second mos impo an a ibu e in con as o consume s in clus e s 1 and 3. As o he p oduc ion
Foods 2020,9, 1730 16 o 16
38.
K ys allis, A.; Ness, M. Consume p e e ences o quali y oods om a Sou h Eu opean pe spec i e: A conjoin
implemen a ion on G eek oli e oil. In . Food Ag ibus. Manag. Re . 2005,8, 62–91.
39.
Fo opoulos, C.; K ys allis, A.; Ness, M. Wine p oduced by o ganic g apes in G eece: Using means-end chains
analysis o e eal o ganic buye s’ pu chasing mo i es in compa ison o he non-buye s. Food Qual. P e e .
2003,14, 549–566. [C ossRe ]
40.
Ba nes, V. B ands, Geog aphical O igin, and he Global Economy: A His o y om he Nine een h Cen u y o
he P esen . J. Leg. His . 2019,40, 326–328. [C ossRe ]
41.
Bou dieu, P. Dis inc ion: A Social C i ique o he Judgemen o Tas e; Ha a d Uni e si y P ess: Camb idge, MA,
USA, 1984; ISBN 0674212770.
42.
Topçu, U.C.; Kaplan, M. Willingness o Buy Fo eign P oduc s in Rela ion o E hnocen ic Tendencies and
Wo ldminded A i udes o Consume s. P ocedia Soc. Beha . Sci. 2015,207, 157–164. [C ossRe ]
Publishe ’s No e:
MDPI s ays neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional
a ilia ions.
©
2020 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access
a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion
(CC BY) license (h p://c ea i ecommons.o g/licenses/by/4.0/).