Mendoza-Mendoza, Adel; Visbal-Cada id, Delimi o; De La Hoz-Domínguez, En ique
A icle
Classi ica ion o La in Ame ican and Ca ibbean coun ies
based on mul idimensional de elopmen indica o s: A
mul i a ia e empi ical analysis
Economies
P o ided in Coope a ion wi h:
MDPI – Mul idisciplina y Digi al Publishing Ins i u e, Basel
Sugges ed Ci a ion: Mendoza-Mendoza, Adel; Visbal-Cada id, Delimi o; De La Hoz-Domínguez,
En ique (2025) : Classi ica ion o La in Ame ican and Ca ibbean coun ies based on
mul idimensional de elopmen indica o s: A mul i a ia e empi ical analysis, Economies, ISSN
2227-7099, MDPI, Basel, Vol. 13, Iss. 6, pp. 1-21,
h ps://doi.o g/10.3390/economies13060178
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329458
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Academic Edi o : Fabio Clemen i
Recei ed: 11 Ap il 2025
Re ised: 27 May 2025
Accep ed: 1 June 2025
Published: 17 June 2025
Ci a ion: Mendoza-Mendoza, A.,
Visbal-Cada id, D., & De La
Hoz-Domínguez, E. (2025).
Classi ica ion o La in Ame ican and
Ca ibbean Coun ies Based on
Mul idimensional De elopmen
Indica o s: A Mul i a ia e Empi ical
Analysis. Economies,13(6), 178.
h ps://doi.o g/10.3390/
economies13060178
Copy igh : © 2025 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 ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
Classi ica ion o La in Ame ican and Ca ibbean Coun ies Based
on Mul idimensional De elopmen Indica o s: A Mul i a ia e
Empi ical Analysis
Adel Mendoza-Mendoza 1,* , Delimi o Visbal-Cada id 2and En ique De La Hoz-Domínguez 3
1Indus ial Enginee ing P og am, Facul y o Enginee ing, Uni e sidad del A lán ico,
Ba anquilla 080001, Colombia
2Indus ial Enginee ing P og am, Facul y o Enginee ing, Uni e sidad del Magdalena,
San a Ma a 470004, Colombia; [email p o ec ed]
3S a is ical and Quan i a i e Me hods Resea ch G oup (GEMC), Uni e sidad del Magdalena,
San a Ma a 470004, Colombia; [email p o ec ed]
*Co espondence: [email p o ec ed]
Abs ac : This s udy de elops a mul idimensional classi ica ion o La in Ame ican and
Ca ibbean coun ies based on a mul idimensional se o economic, social, echnological,
and en i onmen al indica o s. This s udy de elops a mul idimensional assessmen o
he pe o mance o La in Ame ican and Ca ibbean coun ies, aking in o accoun he
ollowing indica o s o he pe iod 2017–2022: educa ion expendi u e (% o GDP), heal h
expendi u e (% o GDP), GDP pe capi a (cons an USD), CO
2
emissions pe capi a (me ic
ons), ene gy consump ion pe capi a (kWh), in e ne use s (% o popula ion), mobile
phone subsc ip ions (pe 100 inhabi an s), and he Global Inno a ion Index (GII). Ini ially,
h ough he applica ion o p incipal componen analysis (PCA), he objec i e was o educe
he complexi y o he da a se and e eal he main s uc u al dimensions. Subsequen ly,
clus e analysis was used o classi y coun ies acco ding o sha ed de elopmen pa e ns.
To achie e his, he a e age o he indica o s o he 2017–2022 pe iod was used as a basis,
which enabled he educ ion in sho - e m dis o ions and he cap u e o s uc u al ends.
The esul s e eal he exis ence o dis inc g oups, wi h coun ies wi h highe le els o
digi al connec i i y, in es men in human capi al, and economic dynamism expe iencing
mo e a o able de elopmen condi ions.
Keywo ds: p incipal componen analysis (PCA); clus e analysis; mul idimensional
assessmen ; socioeconomic indica o s
1. In oduc ion
To unde s and he di e en socioeconomic and en i onmen al dynamics in La in
Ame ican coun ies, compa a i e analysis o di e en indica o s becomes he ool. This
ype o compa ison cons i u es a mul idimensional p oblem. These assessmen s no only
p o ide a comp ehensi e pe spec i e on he cha ac e is ics and challenges aced by na ions
bu also o e an empi ical basis o public policymaking (Pina ,2019;Ba ska e al.,2020;
Guillén-Fe nández,2024
). Se e al echniques ha e p o en use ul in iden i ying pa e ns
and classi ying elemen s based on mul iple indica o s, among which he ollowing mul-
i a ia e analysis echniques s and ou : p incipal componen analysis (PCA) and clus e
analysis (Kama i & Schul z,2022;K zy´sko e al.,2022).
The egion ha includes he coun ies o La in Ame ica and he Ca ibbean is cha -
ac e ized by he he e ogenei y ha occu s in di e en ields, so in economic, social, and
Economies 2025,13, 178 h ps://doi.o g/10.3390/economies13060178
Economies 2025,13, 178 2 o 21
en i onmen al e ms, he e is a high a iabili y in he espec i e indica o s, which is why
he need a ises o use me hodologies ha ha e he capaci y o measu e his di e si y. The
a ailabili y o da a om in e na ional and na ional o ganiza ions o e s an oppo uni y
o explo e hese di e ences om a quan i a i e pe spec i e (San os e al.,2023). The e-
o e, o ex ac ele an in o ma ion ha allows s uc u ing esul s om he complexi y
o mul idimensional da a, ad anced echniques a e equi ed so ha his in o ma ion can
be unde s ood by di e en go e nmen en i ies. (Ray e al.,2021;Guo e al.,2022). The e-
o e, his s udy add esses he ollowing esea ch ques ions: Wha a e he main ac o s
ha explain he di e ences be ween coun ies in he egion? How a e hese coun ies
g ouped acco ding o sha ed cha ac e is ics? How can hese esul s con ibu e o s a egic
decision-making?
Fo his s udy, indica o s we e selec ed ha ake in o accoun coun ies’ de elopmen
in a eas such as educa ion, heal h, he economy, he en i onmen , and echnology. This
selec ion o indica o s seeks o answe he p e iously o mula ed esea ch ques ions. In his
analysis, he da a we e examined using p incipal componen analysis (PCA) o iden i y he
main ac o s explaining a iabili y in he egion and clus e analysis o g oup coun ies ac-
co ding o common pa e ns. This p oposed me hodological app oach con ibu es o be e
isualiza ion and in e p e a ion o esul s, acili a ing decision-making by hose esponsible
o designing go e nmen policies, due o he iden i ica ion o simila i ies when conside ing
he exis ing ela ionship be ween mul iple indica o s (Ko i ,2024;Alkhay a e al.,2020).
2. Li e a u e Re iew
2.1. P incipal Componen Analysis (PCA)
P incipal componen analysis (PCA) is a echnique ha educes he numbe o di-
mensions when he e is a la ge da a se ; o his pu pose, i ans o ms he a iables ha
a e co ela ed in o a new, smalle se called p incipal componen s (Jolli e,2002). This
acili a es he in e p e a ion and isualiza ion o da a in mul idimensional s udies, since,
wi hou losing he mos signi ican a iabili y, he o iginal in o ma ion can be comp essed
(Salem & Hussein,2019;Gewe s e al.,2021).
PCA is based on he decomposi ion o eigen alues o he a iance o co ela ion ma ix
o he o iginal da a; hus, he i s componen ound ies o explain he g ea es a iabili y
o he da a, while he ollowing componen s p og essi ely explain less a iabili y. This
way, p oblems o o e i ing and mul icollinea i y a e a oided by educing dimensions,
which is key in mul i a ia e da a analysis (Jolli e & Cadima,2016).
The applica ion o PCA has been widesp ead in knowledge in a ious ields, including
chemis y, biology, economics, and social sciences (Younes e al.,2021;Abson e al.,2012;
Dai e al.,2021;Zhang e al.,2020). PCA is used abundan ly in all o ms o analysis because
i is a simple, nonpa ame ic me hod o ex ac ing ele an in o ma ion om con using
da a se s. I has also been used o assess coun ies by analyzing in o ma ion on a ious
de elopmen indica o s. Fo example, h ough PCA, key ac o s ha explain he di e ences
in economic g ow h and social well-being ha e been iden i ied, as e idenced in global
compe i i eness s udies (Ku ek e al.,2022;Noman e al.,2024). In he e alua ion o he
en i onmen al pe o mance o di e en egions, PCA has acili a ed he iden i ica ion o
highly signi ican componen s o ecological e iciency, as shown in some en i onmen al
sus ainabili y s udies (Jiang e al.,2018;Almulhim,2024).
Addi ionally, PCA enables he c ea ion o composi e indices ha cap u e he o e all
le el, which can be used o compa e di e en coun ies and egions, allowing policymake s
o iden i y he bes p ac ices and a eas o imp o emen . Due o his, he applica ion o
PCA has been e y impo an o he cons uc ion and de elopmen o aluable composi e
Economies 2025,13, 178 3 o 21
indices such as he Global Compe i i eness Index (GCI) and he Human De elopmen
Index (HDI) (Asongu & Nwachukwu,2017;Qazi,2022).
2.2. Clus e Analysis
Clus e analysis co e s s a is ical echniques o disco e pa e ns wi hin a da a se
wi h he objec i e o g ouping he clus e uni s in such a way ha he elemen s wi hin
a clus e ha e a high deg ee o ela ionship o associa ion wi h each o he , while he
clus e s a e di e en be ween hem (Romesbu g,2004). This me hod is widely used in
a ious ields, ini ially in he disciplines o biology and ecology; in ecen yea s, i has also
been used in a eas such as economics, social sciences, and heal h sciences
(Spoo ,2023
;
Van Leeuwen & Koole,2022
). In he ield o economic and social de elopmen , conglome -
a e analysis has become a key ool o classi ying egions and coun ies, aking in o accoun
di e en indica o s o economic, social, and en i onmen al ypes, allowing o an in eg a ed
iew om a mul idimensional pe spec i e. (Ne i e al.,2017;K ishnan & Fi oz,2023).
Clus e ing me hods can be di ided in o wo g oups: pa i ions and hie a chical, each
wi h i s ad an ages and limi a ions. Pa i ion me hods, such as K-means, seek o assign
each obse a ion o one o he p ede ined g oups, minimizing he in e nal a iance o
he conglome a e. The hie a chical g ouping algo i hms a e based on he ep esen a-
ion o he da a as a hie a chy o nes ed clus e s h ough he inclusion o he se o se s
(MacQueen,1967
). In he pa i ion app oach, i s pe o mance depends on he op imal
numbe o conglome a es ha is de e mined by he elbow me hod o he c i e ia o he sil-
houe e (
Kau man & Rousseeuw,2009
). Hie a chical me hods build an a bo ized g ouping
s uc u e, beginning wi h each da a poin in i s own clus e and successi ely me ging he
mos simila clus e pai s o o m a clus e hie a chy (Wa d,1963). Finally, densi y-based
me hods, such as DBSCAN (Es e e al.,1996), ha e been used o de ec pa e ns in noisy
da a se s wi h complex s uc u es.
In he classi ica ion o coun ies, clus e analysis has had di e en applica ions wi hin
which we can highligh he ollowing: Ahlbo n and Schweicke (2019) used a mac ocon-
glome a e app oach o classi y 115 coun ies acco ding o hei economic s uc u es and
le el o de elopmen , which p o ides an in eg al pe spec i e o global economic di e -
ences and simila i ies. Simila ly, S ielkowski e al. (2024) applied conglome a e analysis o
desc ibe he di e en cha ac e is ics o disca ding and mi iga ion policies o clima e change
in se e al coun ies, p o iding in o ma ion o policy o mula ion.
2.3. In eg a ion o PCA and Clus e Analysis
The simul aneous use o PCA and conglome a e analysis has been a alid me hodolog-
ical s a egy o mul idimensional s udies. PCA educes he dimensionali y o he da a se
h ough he co ela ions be ween a iables; his educ ion op imizes he quali y o g oup
algo i hms and hus imp o es he in e p e a ion o he esul s (Jolli e & Cadima,2016).
This s a egy has been applied in a ious con ex s o he analysis o coun ies and
egions, being an e icien echnique o he analysis o mul idimensional da a se s. This is
how he simul aneous use o hese wo mul i a ia e analysis echniques is seen in se e al
s udies. Jansson e al. (2022), h ough PCA and he K-means algo i hm, educed he
dimensionali y and he g ouping o mul i a ia e ock da a. P ana a e al. (2023) applied
PCA along wi h g oup algo i hms such as K-means and DBSCAN o companies g ouped
in he s ock ma ke o Indonesia. K ishnan and Fi oz (2023) used PCA and conglome a e
analysis o e alua e egional en i onmen al quali y, iden i ying pa e ns ha can used o
basing public policies and sus ainable de elopmen s a egies. T ipa hi e al. (2025) used
PCA and K-means o segmen he ma ke s o he e ail sec o , imp o ing he unde s anding
o consume beha io and op imizing ma ke ing s a egies.
Economies 2025,13, 178 4 o 21
The e o e, his s udy, wi h he combined use o PCA and conglome a e analysis, is a
classi ica ion o La in Ame ican and Ca ibbean coun ies om a mul idimensional poin
o iew, aking in o accoun key indica o s o educa ion, heal h, economy, en i onmen ,
and echnology. Wi h his me hodological in eg a ion, i is hoped ha he classi ica ion
ob ained will be much mo e obus .
2.4. Recen O e iew o Mul idimensional De elopmen in La in Ame ica
In ligh o he mos ecen s udies, La in Ame ican de elopmen has been app oached
om pe spec i es ha anscend adi ional mac oeconomic me ics o inco po a e di-
mensions o subjec i e well-being, job quali y, gende , aging, and ene gy sus ainabili y.
This me hodological e olu ion esponds o he need o cap u e wi h g ea e p ecision he
s uc u al inequali ies ha pe sis in he egion.
Based on he same idea, C uz-Ma ínez (2014) p oposed a Mul idimensional Wel-
a e Index in eg a ing ins i u ional dimensions o co e age and esul s, allowing o a
be e cha ac e iza ion o he wel a e s a e in La in Ame ica h ough ac o analyses ha
explain mo e han 75% o he a iance. Simila ly, San illán e al. (2020) o mula ed a
Mul idimensional Ene gy Po e y Index (MEPI) o se en La in Ame ican coun ies, which
showed a high Human De elopmen Index (HDI) does no necessa ily co ela e wi h low
ene gy dep i a ion. Hence, demons a ing he need o include ene gy access pa ame e s in
mul idimensional indexes.
F om he ield o gende inequali y, Medina-He nández e al. (2021) employed he
HJ-Biplo echnique o examine he economic, physical, and poli ical au onomy o women
in 17 coun ies, inding ha hese dimensions in e ac in an in e dependen manne . Thei
s udy no only con i ms he use ulness o mul i a ia e app oaches bu also jus i ies he
inclusion o gende indica o s in s uc u al de elopmen analyses.
Likewise, B a o-Sanzana e al. (2023) emphasized, in a sys ema ic e iew ega ding
he social clima e in La in Ame ica, he absence o app op ia e ools o assess inclusion, co-
hesion, and pe cei ed sa e y, which unde sco es he need o mul idimensional app oaches
a en i e o a ying educa ional and cul u al amewo ks.
In he labo sphe e, González e al. (2021) p oposed a syn he ic index o dep i a ion
in employmen quali y (QoE) o six Cen al Ame ican coun ies. Using he Alki e–Fos e
me hod, hei analysis shows ha mo e han 60% o labo dep i a ion comes om non-
mone a y ac o s such as ins abili y, in o mali y, and poo wo king condi ions. This inding
ein o ces he a gumen ha he quali y o employmen should be an in eg al pa o
de elopmen me ics.
Finally, Ama an e and Colacce (2022) add essed mul idimensional po e y among
olde adul s in i e La in Ame ican coun ies, highligh ing he impo ance o dimensions
such as heal h, social secu i y, and housing. Thei s udy demons a es ha dep i a ions
pe sis in o old age and ha an in o mal wo k his o y has las ing e ec s, ea i ming he
need o a ge ed measu es o his popula ion.
Toge he , hese wo ks demons a e a g owing consensus: he La in Ame ican eali y
canno be adequa ely explained h ough unidimensional indica o s. The e o e, his s udy
alls wi hin his in eg a i e pa adigm by combining obus s a is ical echniques wi h a
concep ual amewo k ha ecognizes he s uc u al complexi y o de elopmen in La in
Ame ica and he Ca ibbean.
3. Me hodology
This s udy analyzes he mul idimensional pe o mance o La in Ame ican coun ies
using mul i a ia e analysis echniques. Da a om o icial sou ces such as he Wo ld Bank,
he Economic Commission o La in Ame ica and he Ca ibbean (ECLAC), and he Wo ld
Economies 2025,13, 178 5 o 21
Economic Fo um (WEF) we e used, ensu ing compa abili y and quali y o he in o ma ion.
The selec ion o a iables esponds o he need o cap u e key dimensions o coun ies’
de elopmen , encompassing economic, social, and en i onmen al indica o s.
The ollowing a iables we e analyzed o he pe iod o 2017–2022: educa ion spend-
ing (% o GDP), heal h spending (% o GDP), GDP pe capi a (cons an USD), CO
2
emissions
pe capi a (me ic ons), ene gy consump ion pe capi a (kWh), in e ne use s (% o popula-
ion), mobile phone subsc ip ions (pe 100 inhabi an s), and he Global Inno a ion Index
(GII). The a e age alues o each indica o we e used o he analysis o minimize he
e ec s o empo al a iabili y.
The se o indica o s ha we e s udied in his esea ch is ailo ed o add ess he
exis ing gap ega ding de elopmen measu emen s in La in Ame ica and he Ca ibbean,
cap u ing i s mul i-dimensional p og ess in bo h classical and con empo a y app oaches o
well-being. The ange includes educa ion expendi u e (% o GDP) and heal h expendi u e
(%), which indica e he e o s by he go e nmen owa ds human capi al o ma ion,
which is a undamen al componen owa ds a aining sus ainable de elopmen . The e is
as li e a u e suppo ing hese wo indica o s as cons i uen s o social p oduc i i y and
ins i u ional quali y (Ba ska e al.,2020;Al-Wo a i,2024).
GDP pe capi a is commonly used o income le el measu emen , se ing as a p oxy
o es ima ing ci izens’ economic well-being in a coun y. Meanwhile, pe capi a CO
2
emissions and pe capi a ene gy consump ion a e measu es o p oduc i i y and en i-
onmen al sus ainabili y, which a e impo an in he con ex o sus ainable de elopmen
(Jiang e al.,2018;Almulhim,2024).
The pene a ion o in e ne use s (% o he popula ion) and mobile subsc ip ions
pe 100 inhabi an s allow o he e alua ion o he le el o digi al connec i i y, associa ed
wi h echnological inclusion and access o in o ma ion, ac o s ecognized as enable s
o mode n de elopmen (Pick e al.,2021;Pé ez-Mo o e e al.,2020). In he end, he
Global Inno a ion Index (GII) is a composi e indica o ha shows how well coun ies
can c ea e and use knowledge, which is a key pa o economies ha ely on inno a ion
(Asongu & Nwachukwu,2017).
We may ob ain a ull pic u e o p og ess in he a ea by combining hese eigh a i-
ables. This pic u e includes economic, social, en i onmen al, and echnological aspec s.
This selec ion is based no only on hei a ailabili y and in e na ional compa abili y bu
also on hei empi ical and heo e ical ele ance wi hin he ield o mul idimensional
de elopmen s udies.
The use o p incipal componen analysis (PCA) and hie a chical clus e ing was se-
lec ed a e e alua ing h ee c i e ia: sui abili y o he objec i es, p eceden s in he li e -
a u e, and s a is ical obus ness. Fi s ly, he pu pose o iden i ying la en s uc u es and
subsequen ly segmen ing coun ies equi es a echnique ha educes collinea i y wi h-
ou losing ele an in o ma ion; PCA mee s his equi emen by concen a ing mo e han
70% o he a iance in a educed se o componen s, op imizing he subsequen classi i-
ca ion (
Jolli e & Cadima,2016
). Secondly, ecen egional s udies (
K ishnan & Fi oz,2023
;
Jiménez-P eciado e al.,2024)
e idence how clus e ing di ec ly on o iginal a iables ends
o c ea e g oups sensi i e o scaling and noise. By i s p ojec ing he da a in o he ac o ial
space, his isk is educed and he economic in e p e abili y o he clus e s is imp o ed.
Finally, al e na i es such as K-means clus e analysis on aw da a o models based on
syn he ic indica o s—DEA o composi e indices—we e disca ded because, al hough use ul,
hey impose assump ions o con exi y o compensabili y ha could mask key he e o-
genei ies. The Wa d hie a chical me hod, on he o he hand, allows o isualizing he
nes ed ela ionship be ween coun ies and elies on dis ance me ics ha maximize in-
e nal homogenei y. The Ba le and KMO es s, as well as alida ion wi h he co-do
Economies 2025,13, 178 6 o 21
me hod, suppo he me hodological decision and ensu e he eplicabili y and obus ness
o he indings.
S udy Popula ion
Thi y- wo coun ies in La in Ame ica and he Ca ibbean we e included based on da a
a ailabili y. The coun ies analyzed a e An igua and Ba buda, A gen ina, he Bahamas,
Ba bados, Belize, Boli ia, B azil, Chile, Colombia, Cos a Rica, Cuba, Dominica, he Do-
minican Republic, Ecuado , El Sal ado , G enada, Gua emala, Guyana, Hai i, Hondu as,
Jamaica, Mexico, Nica agua, Panama, Pa aguay, Pe u, Sain Ki s and Ne is, Sain Vincen
and he G enadines, Sain Lucia, Su iname, T inidad and Tobago, U uguay, and Venezuela.
P incipal componen analysis (PCA) was used o educe dimensionali y and iden i y
he unde lying ac o s ha explain he g ea es a iabili y ac oss coun ies. This me hod
ans o ms he o iginal a iables in o a se o unco ela ed componen s, p ese ing he
maximum amoun o in o ma ion possible. Be o e applying PCA, he da a we e no malized
o elimina e scale e ec s, using s anda diza ion based on he mean and s anda d de ia ion.
The ollowing c i e ia we e e alua ed o p incipal componen selec ion:
•Kaise ’s C i e ion: Componen s wi h eigen alues g ea e han 1 a e e ained.
•
Explained Va iance: Componen s ha explain a leas 70% o he o al a iabili y
a e selec ed.
•Sc ee Plo : Iden i ica ion o he in lec ion poin on he eigen alue cu e.
To segmen he coun ies, an agglome a i e hie a chical clus e analysis was applied
using Wa d’s me hod, which minimizes wi hin-g oup a iance and op imizes classi ica ion.
The op imal numbe o clus e s was de e mined using he elbow me hod, applying Euclidean
dis ance and Wa d’s linkage c i e ion. This s a egy allowed o he iden i ica ion o s uc u al
pa e ns and he g ouping o coun ies acco ding o simila de elopmen ajec o ies.
Va ious es s we e ca ied ou o ensu e he obus ness o he esul s:
•Sample Adequacy: Ba le ’s es o sphe ici y.
•
PCA Robus ness: Compa ison o esul s wi h al e na i e no maliza ions o he a iables.
•
Clus e ing S abili y: Assessmen by epea ing he analysis wi h di e en dis ance me ics.
The combined use o PCA and clus e analysis s eng hens he obus ness o he classi-
ica ion, as i elimina es edundancies in he o iginal da a and imp o es he in e p e a ion
o he esul ing g oups. By applying clus e ing o he educed p incipal componen space,
i is ensu ed ha he clus e s e lec s uc u al di e ences a he han simple indi idual
a ia ions in he o iginal a iables. This me hodology p o ides a clea e iew o he ela-
ionships be ween coun ies, allowing o he o mula ion o ecommenda ions based on
empi ical e idence.
The me hodological app oach adop ed in his s udy ensu es igo ous and eplicable
analysis, aligned o bes p ac ices in empi ical esea ch on mul idimensional de elopmen .
The combina ion o ad anced s a is ical echniques wi h he use o a e age da a ensu es
s able esul s and allows o s uc u ed compa isons be ween coun ies in he egion. Based
on his classi ica ion, common de elopmen pa e ns can be iden i ied and di e en ia ed
s a egies es ablished o imp o e pe o mance in key a eas such as educa ion, heal h, he
economy, and en i onmen al sus ainabili y.
4. Resul s
4.1. Resul s o P incipal Componen Analysis
The Fac oMineR package ( e sion 2.11) o he R s a is ical so wa e (Husson e al.,2023)
was used o pe o m he PCA. A. The PCA g aphs we e gene a ed using he Fac oex a
package (Kassamba a & Mund ,2016) o he R s a is ical so wa e (R Co e Team, 2024).
Economies 2025,13, 178 7 o 21
Table 1shows, o he i s h ee componen s, he associa ed eigen alues, he a iance
explained by each componen , and he cumula i e pe cen age o a iance explained using
p incipal componen analysis (PCA).
Table 1. Eigen alue and a iabili y explained.
Eigen alues Pe cen age o
Va iance
Cumula i e Pe cen age
o Va iance
Comp. 1 3.5657 44.5711 44.5711
Comp. 2 1.6655 20.8192 65.3903
Comp. 3 0.9364 11.7055 77.0958
The eigen alues ep esen he p opo ion o in o ma ion e ained by each p incipal
componen (see Figu e 1). The PCA esul s show ha h ee ac o s explain 77.096% o
he in o ma ion con ained in he eigh o iginal a iables. The p- alue o Ba le ’s Tes
o Sphe ici y is 6.89
×
10
−17
, so he null hypo hesis “Ho: The co ela ion ma ix equals
he iden i y ma ix” should be ejec ed, indica ing ha p incipal componen s analysis
is app op ia e.
Figu e 1. Eigen alues and explained a iabili y (PCA).
Figu es 2and 3show he co ela ions be ween he a iables and he p incipal com-
ponen s. The co ela ion ci cle shown in he igu es allows us o isualize he co ela ions
be ween he quan i a i e a iables and be ween hese and he p incipal componen s. I can
be seen ha he i s p incipal componen is s ongly associa ed wi h he a iables ene gy,
GDP, in e ne use s, and GII. On he o he hand, he second p incipal componen is s ongly
co ela ed wi h he a iables educa ion and heal h. Simila ly, he a iables Educa ion and
Heal h a e qui e co ela ed wi h each o he , as is he case be ween he a iables ene gy,
GDP, in e ne , and GII. The mobile phone subsc ip ions and CO
2
a iables a e associa ed
wi h he hi d p incipal componen . This can also be seen in Table 2.
Economies 2025,13, 178 8 o 21
Figu e 2. Coo dina es o a iable g oups wi h he p incipal componen s (MFA).
Figu e 3. Co ela ion be ween o iginal a iables wi h he p incipal componen s (PCA).
Table 2. Associa ion o he g oups o a iables wi h he PCA’s dimensions.
Va iable Con ibu ions (%) Quali y o Rep esen a ion
Cosines-Squa ed
PC1 PC2 PC3 PC1 PC2 PC3
Educa ion 0.014 46.609 5.269 0.0005 0.7763 0.0493
Heal h 1.972 37.971 1.574 0.0703 0.6324 0.0147
GDP 21.215 1.532 1.245 0.7565 0.0255 0.0117
CO28.506 3.890 53.713 0.3033 0.0648 0.5030
Ene gy 23.309 3.237 6.822 0.8311 0.0539 0.0639
In e ne 20.226 0.541 1.849 0.7212 0.0090 0.0173
Mobile 9.403 0.059 24.024 0.3353 0.0010 0.2250
GII 15.355 6.160 5.505 0.5475 0.1026 0.0515
The coo dina es o he a iables (also e e ed o as loadings o ac o loadings) indica e
he s eng h and di ec ion o he ela ionship be ween each o iginal a iable and he
p incipal componen s (dimensions) in PCA. These coo dina es e eal how much each
Economies 2025,13, 178 15 o 21
Table 5. The i e speci ic obse a ions o each clus e .
Clus e 1
Hai i Venezuela
4.0230 3.1253
Clus e 2
Cuba El Sal ado Nica agua Hondu as Boli ia
4.6306 3.1464 2.7973 2.7417 2.5856
Clus e 3
Dominican
Republic G anada Sain Lucia Pe u Sain Vincen and
he G enadines
3.1216 2.8962 2.7669 2.6545 2.5304
Clus e 4
Bahamas Chile An igua and
Ba buda U uguay Cos a Rica
4.7637 3.8848 3.7516 3.3213 2.9801
5. Discussion
The esul s ob ained using p incipal componen analysis (PCA) con i m he ele ance
o educing he dimensionali y o a b oad se o a iables, as i allows o cap u ing he
majo i y o a iabili y wi h a small numbe o componen s. This app oach is pa icula ly
use ul when wo king wi h mul iple indica o s ha encompass economic, social, and en i-
onmen al dimensions (Coope e al.,2007). In he p esen s udy, he selec ion o p incipal
componen s was guided by s a is ical c i e ia (eigen alue > 1, explained a iance, and he
sc ee plo ), ensu ing ha he la ges p opo ion o he o iginal in o ma ion was p ese ed.
The use o hie a chical agglome a i e analysis using Wa d’s me hod, applied o PCA
sco es, allowed o he iden i ica ion o obus and s uc u ally cohe en g oupings o coun-
ies. This s a egy, based on Euclidean dis ance and alida ed using he elbow me hod
and he silhoue e index, no only acili a ed isually in e p e able segmen a ion bu also
elimina ed edundancies and collinea i ies be ween indica o s. Compa ed wi h app oaches
such as K-means o models based on obse ed a iables (Pick e al.,2021), he hie a -
chical s uc u e o e ed a mo e lexible and obus classi ica ion o he La in Ame ican
con ex , ein o cing he use ulness o his me hodological combina ion in mul idimen-
sional de elopmen s udies (Mö ing-Ma ínez e al.,2024;Jiménez-P eciado e al.,2024;
K ishnan & Fi oz,2023
). This s a egy minimizes biases ha could a ise om simul ane-
ously including co ela ed a iables, p o iding a clea e iew o he unde lying s uc u e
in he da a.
The indings sugges ha na ions wi h highe le els o a iables linked o inno a-
ion, digi al connec i i y (in e ne and mobile phone use s), and GDP pe capi a end o
clus e oge he , e lec ing de elopmen pa e ns ocused on economic di e si ica ion and
echnology adop ion. On he o he hand, coun ies loca ed in lowe -pe o ming clus e s
sha e ai s such as less access o digi al se ices o educed in es men in heal h and
educa ion, which is consis en wi h he li e a u e linking human capi al accumula ion and
echnological ad ancemen wi h g ea e g ow h and well-being. (Pé ez-Mo o e e al.,2020;
Al-Wo a i,2024).
The obus ness o he esul s was ein o ced by conduc ing Ba le ’s es o sphe ici y
o suppo he use o PCA. This educed he likelihood o o e i ing o unde i ing he ue
s uc u e o he da a. Addi ionally, eplica ing he analysis using di e en dis ance me ics
Economies 2025,13, 178 16 o 21
(e.g., Euclidean and Manha an dis ance) would ha e been an addi ional s ep o compa e
he s abili y o he clus e s (Plaza-Díaz e al.,2020;Haining e al.,2022;Niu e al.,2022).
This wo k o e s a comp ehensi e me hodological app oach in compa ison wi h o he
p e ious s udies in he con ex o La in Ame ica and he Ca ibbean.
San iago e al. (2020
)
examined he ole o globaliza ion and economic eedom in g ow h h ough ARDL
models, bu hey did no add ess la en s uc u es o conside egional segmen a ion.
Ahmed e al. (2021)
cons uc ed an ICT index using p incipal componen analysis (PCA)
and link i o en i onmen al sus ainabili y; howe e , hey do no a emp o g oup coun ies
o alida e he da a s uc u e. Pick e al. (2021) used K-means clus e ing o s udy he digi al
di ide in he egion, bu hei analysis is based only on obse ed a iables, and no in e nal
alida ion es s a e men ioned. In con as , his s udy applies PCA ollowed by hie a chical
clus e analysis, allowing o a de ailed iden i ica ion o g oupings ac oss coun ies. This
combina ion o e s a b oade iew o de elopmen in he egion by in eg a ing a iables
ela ed o heal h, educa ion, economic pe o mance, sus ainabili y, and digi al inno a ion.
The combina ion o p incipal componen analysis (PCA) and clus e analysis has
bo h absolu e and ela i e me hodological ad an ages o e o he me hods used in coun y
classi ica ion s udies. PCA is di e en om me hods like da a en elopmen analysis
(DEA) o mul i a ia e eg ession models since i can combine a lo o mul idimensional
da a wi hou making any assump ions abou how p oduc i e some hing is o how hings
a e ela ed.
F om a compa a i e pe spec i e, s udies ha exclusi ely use me hods like K-means
end o simpli y he s uc u al complexi y o coun ies by elying on a bi a ily selec ed a i-
ables o di ec agg ega ions. In con as , PCA empi ically iden i ies he la en dimensions
ha explain he a iabili y, which imp o es he quali y o he subsequen clus e ing.
Addi ionally, he use o hie a chical analysis (Wa d) on he p incipal componen s
gua an ees g ea e obus ness in segmen a ion compa ed o models ha ope a e di ec ly
on he o iginal a iables, which a e o en co ela ed wi h each o he . This educ ion in
mul icollinea i y and he emphasis on he in e nal s uc u e o he da a p o ide esea che s
and p o essionals wi h a mo e in e p e able, eplicable, and use ul classi ica ion o public
policy design.
Toge he , he PCA–Clus e combina ion p esen s i sel as a solid me hodological
al e na i e, especially in con ex s whe e cap u ing complex s uc u al pa e ns is equi ed,
such as in he case o La in Ame ica and he Ca ibbean. This compa a i e pe spec i e
ein o ces he p ac ical u ili y o he s udy, pa icula ly o hose esponsible o egional
planning and analysis.
These esul s show ha PCA can help g oup coun ies and ind clus e s. E en hough
PCA is no a clus e ing me hod on i s own, i can s ill make he esul s o hie a chical
clus e ing be e by showing pa e ns in he da a when looking a how coun ies pe o m
in di e en a eas.
In e p e ación In eg al de Los Conglome ados: Desempeño, Es uc u a y Fundamen os Teó icos
The classi ica ion in o i e clus e s no only highligh s quan i a i e di e ences among
he coun ies o La in Ame ica and he Ca ibbean bu also allows o he iden i ica ion
o comp ehensi e de elopmen p o iles, suppo ed by his o ical ajec o ies, ins i u ional
con igu a ions, economic s uc u es, and sociocul u al ounda ions. This sec ion aims o
analyze he eme ging pa e ns in ela ion o he heo e ical li e a u e on mul idimensional
de elopmen , go e nance, and s uc u al ans o ma ion.
Clus e 4, composed o U uguay, Chile, Cos a Rica, and Panama, ep esen s a g oup
o high s uc u al pe o mance. These coun ies combine high pe capi a income, high
le els o inno a ion, obus echnological connec i i y, and widesp ead access o essen ial
Economies 2025,13, 178 17 o 21
public se ices. F om he heo y o human capi al and knowledge-o ien ed de elopmen
models (Asongu & Nwachukwu,2017;Alkhay a e al.,2020), his g oup illus a es how
sus ained in es men in educa ion, science, and go e nance can ansla e in o angible
well-being ou comes. Ins i u ionally, hey show g ea e s a e e iciency, egula o y s abili y,
and ela i e social cohesion. In p oduc i e e ms, mo e di e si ied economies a e obse ed,
wi h he capaci y o inse hemsel es in o global alue chains, sus aining a i uous cycle
be ween in e nal capabili ies and ex e nal oppo uni ies.
A he opposi e end, Clus e 1, consis ing o Hai i and Venezuela, e lec s a pa e n
o mul idimensional low pe o mance. The laws show up in low le els o social in es -
men , a ailing educa ional sys em, bad connec ions, and he b eakdown o ins i u ions.
Guillén-Fe nández (2024
) and San os e al. (2023) said ha wi hou basic go e nance and
s a e legi imacy, de elopmen indica o s no only s op mo ing o wa d, bu hey can e en
mo e backwa d o a long ime. These si ua ions a e in line wi h wha hey said. Bo h
coun ies ha e weak o b oken p oduc ion s uc u es, ely hea ily on emi ances o o he
ou side esou ces, and canno c ea e long- e m alue. They a e dealing wi h a lo o dis us ,
a lo o people mo ing away, and a collapse o he social con ac , all o which make hei
s uc u al ulne abili y wo se.
Clus e 2, which con ains Boli ia, Hondu as, Nica agua, and Gua emala, does okay
bu no g ea . Some social indices, such as spending on heal h o educa ion, show edis ibu-
i e measu es, bu hese do no always lead o mo e income, c ea i i y, o en i onmen al
sus ainabili y. The main ac i i ies ha make hese coun ies p oduc i e a e p ima y ac i i-
ies, which ha e poo p oduc i i y and li le added alue.
The s uc u al limi a ions desc ibed by Kama i and Schul z (2022) a e e lec ed in
pe sis en in o mali y, dependence on na u al esou ces o emi ances, and limi ed digi al
ans o ma ion. On he ins i u ional le el, he e is agili y in public managemen , low ax
p essu e, and unequal se ice co e age. E hnic and cul u al di e si y, al hough a po en ial
weal h, o en does no in eg a e e ec i ely in o de elopmen p ocesses.
Clus e 3, mo e di e se, g oups coun ies like Pe u, he Dominican Republic, Ecuado ,
and Pa aguay, wi h in e media e pe o mance. These coun ies show pa ial p og ess in
connec i i y, heal h, and GDP pe capi a, bu wi hou s uc u al consolida ion. F om he
pe spec i e o Pick e al. (2021) and Pé ez-Mo o e e al. (2020), hese a e economies ha
ansi ion be ween adi ional models and p oposals mo e o ien ed owa ds inno a ion,
al hough wi h ins i u ional limi a ions o sus ain he e o ms. Economic s uc u es end o
be dual: mode n sec o s (mining, ou ism, inancial se ices) coexis wi h la ge a eas o
in o mali y o u al backwa dness. A he sociocul u al le el, go e nance is mo e s able
han in Clus e 2 bu is s ill condi ioned by poli ical cycles, egional agmen a ion, o
dis us owa ds he eli es.
Clus e 5, composed exclusi ely o T inidad and Tobago, cons i u es an a ypical case.
I s economic pe o mance is high in e ms o pe capi a income and ene gy consump-
ion, bu i is based on an economy in ensi e in ossil esou ces. F om Almulhim’s (2024)
pe spec i e, his ype o economy aces he “pa adox o plen y”: g ow h wi h low di e si i-
ca ion and isks o ex e nal ola ili y. Al hough i s social indica o s a e accep able and i s
ins i u ions ela i ely s able, he model is ulne able o in e na ional p ice shocks and aces
long- e m sus ainabili y challenges. The coun y’s cul u al and eligious di e si y, along
wi h egional he e ogenei y, poses addi ional challenges o social cohesion.
In summa y, his in eg a ed in e p e a ion shows ha empi ical classi ica ion is no
only use ul o desc ibing bu also o explaining s uc u al di e ences in egional de elop-
men . The conglome a es e lec di e gen pa hs, whe e pe o mance is insepa able om
ac o s such as ins i u ional quali y, economic s uc u e, and social ab ic. Fa om being a
echnical segmen a ion, his ypology cons i u es an in e p e a i e ool ha allows us o
Economies 2025,13, 178 18 o 21
unde s and how capaci ies, cons ain s, and oppo uni ies a e a icula ed in he di e en
na ional con ex s o La in Ame ica and he Ca ibbean.
F om his s udy, i ollows ha La in Ame ica and he Ca ibbean canno be analyzed
as a homogeneous de elopmen block, as he e a e clea ly di e en iable s uc u al pa e ns
ha go beyond indi idual coun y s a is ics. The classi ica ion in o conglome a es, based
on mul iple dimensions—economic, social, echnological, and en i onmen al—allows o
he p ecise iden i ica ion o g oups o coun ies acing simila challenges and, he e o e,
could bene i om di e en ia ed and collabo a i e egional s a egies.
The analysis shows ha coun ies wi h g ea e digi al connec i i y, in es men in
inno a ion, and human capi al end o achie e highe le els o de elopmen , while hose
wi h low s a e capaci y and limi ed social in es men p esen mo e ad e se condi ions.
Fu he mo e, by inco po a ing a sociocul u al eading o he esul s, i is ecognized ha he
nume ical indica o s e lec , in many cases, his o ical ajec o ies, ins i u ional dynamics,
and deeply oo ed cul u al con ex s.
Fo all he abo e easons, his s udy o e s a use ul and adap able me hodological ool
o classi ying coun ies based on comp ehensi e c i e ia, p o iding empi ical e idence
o suppo a ge ed public policies. A he same ime, i in i es us o e hink egional
de elopmen om a logic o segmen ed and con ex ualized coope a ion ins ead o uni e -
sal ecipes.
6. Conclusions
In his s udy, we p oposed a me hodology o unde s and he pe o mance o La in
Ame ican and Ca ibbean coun ies in se e al key a eas. To do so, we combined p incipal
componen analysis (PCA) wi h hie a chical clus e analysis. The objec i e is o g oup
coun ies based on indica o s ela ed o educa ion, heal h, economy, en i onmen , and
echnology. This combina ion allows us o analyze complex in o ma ion mo e easily and
iden i y pa e ns ha migh no be ob ious a i s glance.
The esul s show ha coun ies wi h highe GDP pe capi a le els, be e Global
Inno a ion Index (GII) sco es, g ea e digi al connec i i y, and highe pe capi a ene gy
consump ion end o appea in g oups wi h g ea e o e all de elopmen . On he o he
hand, coun ies wi h less access o he in e ne and mobile de ices, lowe GII alues, and
lowe in es men in heal h and educa ion end o be loca ed in clus e s wi h less a o able
condi ions. This ype o g ouping helps build a clea e pic u e o he egion and can se e
as a s a ing poin o de eloping public policies be e adap ed o he eali ies o each
g oup o coun ies.
I should be no ed ha he s udy is based on o icial da a, using a e age alues om
he 2017–2022 pe iod. This helps educe he impac o sho - e m changes. Howe e , he e
a e some limi a ions, especially ega ding he lack o da a o some coun ies o ac o s
ha a e di icul o measu e, such as deep-sea ed s uc u al p oblems. E en so, he me hod
used is obus and could be applied o o he egions o simila analysis.
In u u e wo k, i would be in e es ing o compa e hese indings wi h he esul s o
o he me hods, such as da a en elopmen analysis (DEA), hyb id models like PCA-DEA,
and ac o analysis combined wi h DEA. This ype o compa ison could help imp o e he
design o s a egies o each g oup o coun ies.
Au ho Con ibu ions: W i ing— e iew and edi ing, w i ing—o iginal d a , alida ion, and concep-
ualiza ion: A.M.-M., D.V.-C. and E.D.L.H.-D. Me hodology: A.M.-M. and D.V.-C. Fo mal analysis:
A.M.-M. and D.V.-C. Resou ces: A.M.-M. and D.V.-C. Da a cu a ion: A.M.-M. and E.D.L.H.-D.
So wa e: E.D.L.H.-D. and D.V.-C. All au ho s ha e ead and ag eed o he published e sion o
he manusc ip .
Economies 2025,13, 178 19 o 21
Funding: This esea ch ecei ed no ex e nal unding.
Ins i u ional Re iew Boa d S a emen : No applicable.
In o med Consen S a emen : No applicable.
Da a A ailabili y S a emen : Da a will be made a ailable on eques .
Con lic s o In e es : The au ho s decla e no con lic s o in e es .
Re e ences
Abson, D. J., Dougill, A. J., & S inge , L. C. (2012). Using p incipal componen analysis o in o ma ion- ich socio-ecological
ulne abili y mapping in Sou he n A ica. Applied Geog aphy,35(1–2), 515–524. [C ossRe ]
Ahlbo n, M., & Schweicke , R. (2019). Economic sys ems in de eloping coun ies—A mac o clus e app oach. Economic Sys ems,
43(3–4), 100692. [C ossRe ]
Ahmed, Z., Na haniel, S. P., & Shahbaz, M. (2021). The c i icali y o in o ma ion and communica ion echnology and human capi al in
en i onmen al sus ainabili y: E idence om La in Ame ican and Ca ibbean coun ies. Jou nal o Cleane P oduc ion,286, 125529.
[C ossRe ]
Alkhay a , M., Aljnidi, M., & Aljoumaa, K. (2020). A compa a i e dimensionali y educ ion s udy in elecom cus ome segmen a ion
using deep lea ning and PCA. Jou nal o Big Da a,7(1), 9. [C ossRe ]
Almulhim, T. (2024). Assessmen o g een de elopmen le el pe o mance in G20 coun ies: A dynamic e alua ion amewo k. Heliyon,
10(18), e37622. [C ossRe ]
Al-Wo a i, Y. M. (2024). Technology in public heal h educa ion in de eloping coun ies. In Handbook o medical and heal h sciences in
de eloping coun ies: Educa ion, p ac ice, and esea ch (pp. 1–20). Sp inge In e na ional Publishing.
Ama an e, V., & Colacce, M. (2022). Mul idimensional po e y among olde people in i e La in Ame ican coun ies. Social Indica o s
Resea ch,159(3), 945–965. [C ossRe ]
Asongu, S. A., & Nwachukwu, J. C. (2017). The compa a i e inclusi e human de elopmen o globaliza ion in A ica. Social Indica o s
Resea ch,134, 1027–1050. [C ossRe ]
Ba ska, A., J˛ed zejczak-Gas, J., Wy wa, J., & Kononowicz, K. (2020). Mul idimensional assessmen o he social de elopmen o EU
coun ies in he con ex o implemen ing he concep o sus ainable de elopmen . Sus ainabili y,12(18), 7821. [C ossRe ]
B a o-Sanzana, M. V., Va ela, J., Te án-Mendoza, O., & Rod iguez-Ri as, M. E. (2023). Measu ing school social clima e in La in Ame ica:
The need o mul idimensional and mul i-in o man es s—A sys ema ic e iew. F on ie s in Psychology,14, 1190432. [C ossRe ]
[PubMed]
Coope , W. W., Sei o d, L. M., & Tone, K. (2007). Da a en elopmen analysis: Comp ehensi e ex wi h models, applica ions, e e ences and
DEA-sol e so wa e (Vol. 2, p. 489). Sp inge .
C uz-Ma ínez, G. (2014). Wel a e s a e de elopmen in La in Ame ica and he Ca ibbean (1970s–2000s): Mul idimensional wel a e
index, i s me hodology and esul s. Social Indica o s Resea ch,119, 1295–1317. [C ossRe ]
Dai, P. F., Xiong, X., & Zhou, W. X. (2021). A global economic policy unce ain y index om p incipal componen analysis. Finance
Resea ch Le e s,40, 101686. [C ossRe ]
Es e , M., K iegel, H. P., Sande , J., & Xu, X. (1996, Augus 2–4). A densi y-based algo i hm o disco e ing clus e s in la ge spa ial da abases
wi h noise. Kdd (Vol. 96, No. 34, pp. 226–231), Po land, OR, USA.
Gewe s, F. L., Fe ei a, G. R., A uda, H. F. D., Sil a, F. N., Comin, C. H., Amancio, D. R., & Cos a, L. D. F. (2021). P incipal componen
analysis: A na u al app oach o da a explo a ion. ACM Compu ing Su eys (CSUR),54(4), 1–34.
González, P., Sehnb uch, K., Apablaza, M., Méndez Pineda, R., & A iagada, V. (2021). A mul idimensional app oach o measu ing
quali y o employmen (QoE) dep i a ion in six cen al Ame ican coun ies. Social Indica o s Resea ch,158(1), 107–141. [C ossRe ]
Guillén-Fe nández, Y. B. (2024). Socioeconomic ac o s de e mining mul idimensional child po e y g oups in cen al Ame ica: A
measu emen p oposal om he wellbeing app oach using a comp ehensi e se o child en’s igh s. Child Indica o s Resea ch,
17(5), 2175–2217. [C ossRe ]
Guo, H., Liang, D., Sun, Z., Chen, F., Wang, X., Li, J., Zhu, L., Bian, J., Wei, Y., Huang, L., Chen, Y., Peng, D., Li, X., Lu, S., Liu, J., &
Shi azi, Z. (2022). Measu ing and e alua ing SDG indica o s wi h Big Ea h Da a. Science Bulle in,67(17), 1792–1801. [C ossRe ]
[PubMed]
Haining, K., Gajwani, R., G oss, J., Gumley, A. I., Ince, R. A. A., Law ie, S. M., Schul ze-Lu e , F., Schwannaue , M., & Uhlhaas, P. J.
(2022). Cha ac e izing cogni i e he e ogenei y in indi iduals a clinical high- isk o psychosis: A clus e analysis wi h clinical
and unc ional ou come p edic ion. Eu opean A chi es o Psychia y and Clinical Neu oscience,272(3), 437–448. [C ossRe ]
Husson, F., Josse, J., Lê, S. F., & Fac oMineR, J. M. (2023). Mul i a ia e explo a o y da a analysis and da a mining wi h R Package ‘Fac oMineR’.
A ailable online: h ps://c an. -p ojec .o g/web/packages/Fac oMineR/index.h ml (accessed on 27 May 2025).
Economies 2025,13, 178 20 o 21
Husson, F., Josse, J., & Pages, J. (2010). P incipal componen me hods-hie a chical clus e ing-pa i ional clus e ing: Why would we
need o choose o isualizing da a. Applied Ma hema ics Depa men ,2010, 1–17.
Jansson, N. F., Allen, R. L., Skogsmo, G., & Ta akoli, S. (2022). P incipal componen analysis and K-means clus e ing as ools du ing
explo a ion o Zn ska n deposi s and indus ial ca bona es, Sala a ea, Sweden. Jou nal o Geochemical Explo a ion,233, 106909.
[C ossRe ]
Jiang, Q., Liu, Z., Liu, W., Li, T., Cong, W., Zhang, H., & Shi, J. (2018). A p incipal componen analysis based h ee-dimensional
sus ainabili y assessmen model o e alua e co po a e sus ainable pe o mance. Jou nal o Cleane P oduc ion,187, 625–637.
[C ossRe ]
Jiménez-P eciado, A. L., C uz-Aké, S., & Venegas-Ma ínez, F. (2024). Iden i ica ion o pa e ns in CO
2
emissions among 208 coun ies:
K-Means clus e ing combined wi h PCA and non-linea -SNE isualiza ion. Ma hema ics,12(16), 2591. [C ossRe ]
Jolli e, I. T. (2002). P incipal componen analysis o special ypes o da a (pp. 338–372). Sp inge .
Jolli e, I. T., & Cadima, J. (2016). P incipal componen analysis: A e iew and ecen de elopmen s. Philosophical T ansac ions o he
Royal Socie y A: Ma hema ical, Physical and Enginee ing Sciences,374(2065), 20150202. [C ossRe ]
Kama i, A., & Schul z, C. P. L. (2022). A combined p incipal componen analysis and clus e ing app oach o explo ing eno mous
eno a ion design spaces. Jou nal o Building Enginee ing,48, 103971. [C ossRe ]
Kassamba a, A., & Mund , F. (2016). Package “ ac oex a”. Ex ac and isualize he esul s o mul i a ia e da a analyses. A ailable online:
h ps://c an. -p ojec .o g/web/packages/ ac oex a/index.h ml (accessed on 27 May 2025).
Kau man, L., & Rousseeuw, P. J. (2009). Finding g oups in da a: An in oduc ion o clus e analysis. John Wiley & Sons.
Ko i , E. K. (2024). Compa a i e clus e ing and isualiza ion o socioeconomic and heal h indica o s: A case o Kenya. Socio-Economic
Planning Sciences,95, 101961. [C ossRe ]
K ishnan, S., & Fi oz, M. (2023). Assessmen o egional en i onmen al quali y using clus e analysis. En i onmen al De elopmen ,
45, 100832. [C ossRe ]
K zy´sko, M., Nijkamp, P., Ra ajczak, W., & Woły´nski, W. (2022). Mul idimensional economic indica o s and mul i a ia e unc ional
p incipal componen analysis (MFPCA) in a compa a i e s udy o coun ies’ compe i i eness. Jou nal o Geog aphical Sys ems,
24(1), 49–65. [C ossRe ]
Ku ek, K. A., Heijman, W., an Ophem, J., G˛edek, S., & S ojny, J. (2022). Measu ing local compe i i eness: Compa ing and in eg a ing
wo me hods PCA and AHP. Quali y & Quan i y,56(3), 1371–1389.
MacQueen, J. (1967). Some me hods o classi ica ion and analysis o mul i a ia e obse a ions. In P oceedings o he i h Be keley
symposium on ma hema ical s a is ics and p obabili y, olume 1: S a is ics (Vol. 5, pp. 281–298). Uni e si y o Cali o nia P ess.
Medina-He nández, E., Fe nández-Gómez, M. J., & Ba e a-Mellado, I. (2021). Gende inequali y in La in Ame ica: A mul idimensional
analysis based on ECLAC indica o s. Sus ainabili y,13(23), 13140. [C ossRe ]
Mö ing-Ma ínez, G., Senzeybek, M., & Jochem, P. (2024). Clus e ing he Eu opean Union elec ic ehicle ma ke s: A scena io analysis
un il 2035. T anspo a ion Resea ch Pa D: T anspo and En i onmen ,135, 104372. [C ossRe ]
Ne i, L., D’Agos ino, A., Regoli, A., Pulselli, F. M., & Coscieme, L. (2017). E alua ing dynamics o na ional economies h ough clus e
analysis wi hin he inpu -s a e-ou pu sus ainabili y amewo k. Ecological Indica o s,72, 77–90. [C ossRe ]
Niu, T., Huang, W., Zhang, C., Zeng, T., Chen, J., Li, Y., & Liu, Y. (2022). S udy o deg ada ion o uel cell s ack based on he collec ed
high-dimensional da a and clus e ing algo i hms calcula ions. Ene gy and AI,10, 100184. [C ossRe ]
Noman, A. A., Rehman, F. U., Khan, I., & Saeed, T. (2024). T ade compe i i eness and sus ainable ene gy p ac ices: The mode a ing
ole o ecological oo p in . En i onmen al and Sus ainabili y Indica o s,21, 100331. [C ossRe ]
Pé ez-Mo o e, R., Pon ones-Rosa, C., & Núñez-Chicha o, M. (2020). The e ec s o e-go e nmen e alua ion, us and he digi al
di ide in he le els o e-go e nmen use in Eu opean coun ies. Technological Fo ecas ing and Social Change,154, 119973. [C ossRe ]
Pick, J., Sa ka , A., & Pa ish, E. (2021). The La in Ame ican and Ca ibbean digi al di ide: A geospa ial and mul i a ia e analysis.
In o ma ion Technology o De elopmen ,27(2), 235–262. [C ossRe ]
Pina , M. (2019). Mul idimensional well-being and inequali y ac oss he Eu opean egions wi h al e na i e in e ac ions be ween he
well-being dimensions. Social Indica o s Resea ch,144(1), 31–72. [C ossRe ]
Plaza-Díaz, J., Molina-Mon es, E., So o-Méndez, M. J., Mad igal, C., He nández-Ruiz, Á., Vale o, T., Villoslada, F. L., Leis, R., de Vic o ia,
E. M., Mo eno, J. M., O ega, R. M., Ruiz-López, M. D., Va ela-Mo ei as, G., & Gil, Á. (2020). Clus e ing o die a y pa e ns and
li es yles among spanish child en in he EsNuPi s udy. Nu ien s,12(9), 2536. [C ossRe ]
P ana a, K. S., Gunawan, A. A., & Gaol, F. L. (2023). De elopmen clus e ing sys em IDX company wi h k-means algo i hm and
DBSCAN based on undamen al indica o and ESG. P ocedia Compu e Science,216, 319–327. [C ossRe ]
Qazi, A. (2022). Adop ion o a p obabilis ic ne wo k model in es iga ing coun y isk d i e s ha in luence logis ics pe o mance
indica o s. En i onmen al Impac Assessmen Re iew,94, 106760. [C ossRe ]
Ray, P., Reddy, S. S., & Bane jee, T. (2021). Va ious dimension educ ion echniques o high dimensional da a analysis: A e iew.
A i icial In elligence Re iew,54(5), 3473–3515. [C ossRe ]
Romesbu g, H. C. (2004). Clus e analysis o esea che s. Lulu P ess.
Economies 2025,13, 178 21 o 21
Salem, N., & Hussein, S. (2019). Da a dimensional educ ion and p incipal componen s analysis. P ocedia Compu e Science,163, 292–299.
[C ossRe ]
San iago, R., Fuinhas, J. A., & Ma ques, A. C. (2020). The impac o globaliza ion and economic eedom on economic g ow h: The case
o La in Ame ica and Ca ibbean coun ies. Economic Change and Res uc u ing,53(1), 61–85. [C ossRe ]
San illán, O. S., Cedano, K. G., & Ma ínez, M. (2020). Analysis o ene gy po e y in 7 La in Ame ican coun ies using mul idimensional
ene gy po e y index. Ene gies,13(7), 1608. [C ossRe ]
San os, G. F. D., Vi es Ve ga a, A., Fuen es-Albu quenque, M., de Sousa Filho, J. F., Pai a, A. S., Useche, A. F., Yamada, G., Al a o, T.,
F iche, A. A. L., And ade, R. F. S., Ba e o, M. L., Caia a, W. T., & Diez-Roux, A. V. (2023). Socioeconomic u ban en i onmen in
La in Ame ica: Towa ds a ypology o ci ies. Sus ainabili y,15(8), 6380. [C ossRe ]
Spoo , J. M. (2023). Imp o ing cus ome segmen a ion ia classi ica ion o key accoun s as ou lie s. Jou nal o Ma ke ing Analy ics,11(4),
747–760. [C ossRe ]
S ielkowski, W., Chyg yn, O., D ozd, S., & Koibichuk, V. (2024). Sus ainable ans o ma ion o ene gy sec o : Clus e analysis o he
sus ainable de elopmen s a egies o selec ed Eu opean coun ies. Heliyon,10(19), e38930. [C ossRe ]
T ipa hi, S., Bachmann, N., B unne , M., Tuezuen, A., Thienemann, A. K., Pöch age , S., & Jodlbaue , H. (2025). E alua ion o clus e ing
wi h PCA o ma ke segmen a ion: A s udy using simula ed and su oga e da a. P ocedia Compu e Science,253, 2063–2075.
[C ossRe ]
Van Leeuwen, R., & Koole, G. (2022). Da a-d i en ma ke segmen a ion in hospi ali y using unsupe ised machine lea ning. Machine
Lea ning wi h Applica ions,10, 100414. [C ossRe ]
Wa d, J. H., J . (1963). Hie a chical g ouping o op imize an objec i e unc ion. Jou nal o he Ame ican S a is ical Associa ion,58(301),
236–244. [C ossRe ]
Younes, K., Mouh ady, O., Chaouk, H., Obeid, E., Rou ayel, R., Mogh abi, A., & Mu shid, N. (2021). The applica ion o p incipal
componen analysis (PCA) o he op imiza ion o he condi ions o ab ica ion o elec opunk nano ib ous memb ane o
desalina ion and ion emo al. Memb anes,11(12), 979. [C ossRe ]
Zhang, H., Cheng, S., Li, H., Fu, K., & Xu, Y. (2020). G oundwa e pollu ion sou ce iden i ica ion and appo ionmen using PMF
and PCA-APCA-MLR ecep o models in a ypical mixed land-use a ea in Sou hwes e n China. Science o he To al En i onmen ,
741, 140383. [C ossRe ]
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .
