Tou ism & Managemen S udies, 12(1) (2016), 181-187 DOI: 10.18089/ ms.2016.12119
181
To wha ex en does human capi al di e si y mode a e he ela ionship be ween HRM p ac ices
and o ganiza ional pe o mance: E idence om Spanish i ms
En qué medida la di e sidad de capi al humano mode a la elación en e las p ác icas de GRH y el endimien o o ganizacional:
E idencia de las emp esas españolas
Ra ael T igue o-Sánchez
Uni e si y o Se ille, School o Economics and Business, Depa men o Business Adminis a ion and Ma ke ing, 41018, Se illa,
Spain, igue [email protected]
Jesús C. Peña-Vinces
Uni e si y o Se ille, School o Economics and Business, Depa men o Business Adminis a ion and Ma ke ing, A enida Ramón y
Cajal nº 1, 41018, Se illa, Spain, [email p o ec ed]
Me cedes Sánchez-Apellániz
Uni e si y o Se ille, School o Economics and Business, Depa men o Business Adminis a ion and Ma ke ing, 41018, Se illa,
Spain, [email p o ec ed]
Abs ac
The pu pose o his esea ch s udy is o explo e he mode a ing e ec
he di e si y o human capi al may ha e on he ela ionship be ween
HRM p ac ices and business pe o mance. To his end, ac o s
de e mining o human capi al di e si y ha e been used.
Wi h a sample o mo e han one hund ed Spanish companies we ha e
ca ied ou a ac o analysis-p incipal axis ac o ing wi h a imax
o a ion- on iden i ied HRM p ac ices and pe cei ed o ganiza ional
pe o mance as ac o s wi h good ac o loadings, and consis en wi h
he p oposed model.
Ou indings indica e ha he human capi al ac o s such as “educa ion
le el," “ unc ional specializa ion” and “leng h o se ice” condi ion he
e ec s o HRM-p ac ices on o ganiza ional pe o mance.
The li e a u e pays li le a en ion o non-linea models. Examining
he ac o s' de e mining human capi al di e si y cas s some ligh on
he black box o he ela ionship be ween human esou ces p ac ices
and o ganiza ional pe o mance.
Keywo ds: Employee di e si y, human capi al, HRM p ac ices,
o ganiza ional pe o mance, Spanish i ms.
Resumen
El p opósi o de es e es udio de in es igación es explo a el e ec o
mode ado que puede ene la di e sidad de capi al humano sob e la
elación en e las p ác icas de Ges ión de Recu sos Humanos (GRH) y
endimien o de la emp esa. Pa a alcanza es e obje i o, se han u ilizado
los ac o es de e minan es de la di e sidad del capi al humano. Con una
mues a de más de 100 de emp esas españolas y un análisis ac o ial con
una o ación a imax, hemos e aluado las p ác icas de GRH y el
desempeño o ganizacional pe cibido. Dichas a iables han mos ado
buenas ca gas ac o iales, en consonancia con el modelo p opues o.
Nues as conclusiones indican que los ac o es de capi al humano ales
como: el ni el educa i o, la especialización uncional y la an igüedad en
el pues o, condicionan los e ec os de las p ác icas de GRH y el
desempeño o ganizacional.
El es udio de la elación p ác ica de GRH- desempeño o ganizacional y la
e aluación de los ac o es que de e minan la di e sidad del capi al
humano nos a oja algo de luz sob e la llamada caja neg a. Debido a que,
la li e a u a p es a poca a ención a los modelos no lineales.
Palab as cla e: Di e sidad de los empleados, capi al humano,
p ác icas de GRH, endimien o o ganizacional, emp esas españolas.
1. In oduc ion
In ecen yea s, he e has been a g owing in e es among
esea che s on he e ec o employee di e si y on
o ganiza ional pe o mance (Jackson, Joshi & E ha d , 2003;
Van Knippenbe g & Schippe s, 2007). Jackson e al. (2003)
ha e no ed ha al hough esea ch s udies analyzing he
impac o di e si y on pe o mance a e based on a clea ly
s uc u ed se o heo e ical app oaches; one o hei
weaknesses is ha only a ew o hem do analyze how
di e si y a ec s pe o mance.
Di e si y e e s o he di e ences among indi iduals based on
any a ibu e ha may lead o a pe cep ion ha someone is
di e en om onesel (Williams & O’Reilly, 1998) o among
in e dependen membe s o a wo k uni (Jackson e al., 2003).
A ibu es o in e es may be hose ha a e quickly de ec ed
when we mee someone o he i s ime (demog aphic
a ibu ed such as age, gende o e hnici y); o he unde lying
less pe cei able a ibu es, only become appa en when we
ge o know an indi idual (pe sonali y, knowledge, alues); and
he e a e o he ones which a e ound be ween hese wo
ex emes o anspa ency (such as educa ion le el o
p o essional expe ience), which a e known as “human capi al”
due o hei ela ionship o p o essional knowledge, skills and
know-how (Becke , 1964; T igue o-Sánchez, Peña-Vinces &
Sánchez-Apellániz, 2011).
Acco ding o Van Knippenbe g and Schippe s (2007), he
dissimila indings we can ind in his esea ch ield could lead
us o assume ha he p e ailing esea ch s udies in his ield
ha e ocused on he “main e ec s," such as comp omise,
sa is ac ion, es ing ela ions be ween di e en dimensions o
di e si y and ou comes, neglec ing po en ially mode a ing
a iables. Al hough he mode a ing na u e o some a iables
o he ela ion be ween di e si y and pe o mance has been
analyzed, as i is he case wi h leade ship o wi h eam
expe ience o e ime (Ha ison, P ice, Ga in & Flo ey, 2002),
ew esea ch s udies ha e conside ed di e si y as a
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
182
mode a ing a iable be ween Human Resou ces (HR) policies
and p ac ices and co po a e pe o mance. This s udy aims o
ill his esea ch gap. Consequen ly, he objec i e o his
esea ch is o analyse he mode a ing e ec ha employee
di e si y has in he ela ionship be ween HR policies ocusing
on commi men and o ganiza ional pe o mance.
Mo eo e , mo e han hal o Spanish i ms ollow an isomo phic
app oach; in o he wo ds, hese companies end o imi a e o
copy he p ac ices (HRM p ac ices) o o he businesses bu in
ac , hey a e no awa e o how employee di e si y a ec s hei
o ganiza ional pe o mance. In his sense, ou in es iga ion is
p esen ed as an al e na i e o his p oblem.
Fu u e esea ch will bene i om he ole o di e si y as a
mode a ing elemen in HR managemen (HRM) and o how i
can con ibu e o business success con ibu ing o he
s eng hening o he Resou ce-Based View, acco ding o which
HR is a sou ce o compe i i e ad an age.
2. Resea ch model and cons uc de ini ions.
2.1 High commi men HRM p ac ices
Since he 1990s, i is possible o ind in he li e a u e a g owing
in e es o HRM and in pa icula , o he ela ionship be ween
HRM and pe o mance (Gues , 1997; 2011). The li e a u e
sugges s ha Human Resou ces (HR) p ac ices a e di ec ly and
indi ec ly linked o he collec i e beha io o employees, ha
hey may ha e an impac on hem and hei a i udes (Al-Jaba i,
2013), which in u n can se e as media o s be ween such
p ac ices and co po a e pe o mance.
Conce ning he o ien a ion o HR p ac ices, Wal on (1985)
indica es he need o change om con ol p ac ices o
commi men p ac ices as he basis o managing he wo k o
indi iduals. Fo his au ho , he e is no choice; he p esc ibes a
commi men s a egy i we in end o p ospe . The idea ha
ele a ed commi men HR managemen p ac ices ha e an
impac on a i udes owa ds wo k h ough employee
pe cep ion o h ough hei expe iences is suppo ed by he
Social Exchange Theo y (C opanzano & Mi chell, 2005), and
he signaling “ heo y” (Os o & Bowen, 2000). These heo ies
p opose ha high commi men HR p ac ices ha e an impac
on employees because hey suppo hem; hey wo k as
“signs” o he in en ions o he o ganiza ion owa ds hem.
2.2 Pe cei ed o ganiza ional pe o mance and i s
ela ionship o HRM p ac ices
The e a e mul iple esea ch s udies ha ha e add essed he
ela ionship exis ing be ween HR managemen and
o ganiza ional pe o mance (OP) as well as hei measu emen
(Ba ena-Ma ínez, López-Fe nández & Rome o-Fe nández,
2013). Howe e , he e is no clea consensus on how o
measu e i (Gues , 2011). Gues (1997) has aised doub s
abou he “cause-e ec ” ela ionship be ween HRM as inpu
and esul s based on inancial pe o mance. “Despi e o he
a ac i e o using economic indica o s in any a emp o
con ince op manage s o he impac o HRM, we need o use
a wide ange o measu es i we wan o unde s and how and
why HRM has an impac on inancial pe o mance”(Gues ,
1997: 274).
The use o “mo e closely- ela ed” indica o s, especially hose
on which he wo k o ce may ha e an impac a e heo e ically
mo e plausible (since HRM aims a imp o ing he di ec
con ibu ion o employees o pe o mance) and
me hodologically hey a e easie o ela e. We need
pe o mance indica o s ha a e much close in e ms ha hey
indeed a ec HRM p ac ices. In his line, he e is a clea op ion
o he use o non- inancial measu es when analyzing
e ec i eness o HRM p ac ices (Bon is, C ossan & Hulland,
2002; T igue o-Sánchez, Peña-Vinces, González-Rendón &
Sánchez-Apellániz, 2012).
An impo an mechanism ha has been p oposed sugges s
ha he impac o HR p ac ices on pe o mance akes place
h ough indi idual employees (Ne es, Gal ão & Pe ei a, 2013).
Some au ho s emphasize he c ucial ole o employee
a i udes and beha io s o ansla e HR p ac ices o i m
pe o mance. In line wi h his mo e cen al ole o employees,
hese au ho s emphasize he need o including employee
pe cep ions in HR esea ch s udies (Gues , 2011).
Gues (2011) says ha he models o igina ing in o ganiza ional
and social psychology make us e lec on he ac ha wha is
ac ually impo an is no he p esence o HR p ac ices, bu
a he he pe cep ion o he in en ion unde lying such
p ac ices. The way in which p ac ices a e in e p e ed can
shape he esponse o hem.
In HR s a egic managemen esea ch pe cep ions on
co po a e pe o mance o he HR depa men o o manage s
ha e been equen ly used as pe o mance indica o s being
conside ed as a easonable subs i u e o objec i e
pe o mance measu es (Den Ha og, Boon, Ve bu g & C oon,
2013). Measu emen o pe cei ed o ganiza ional pe o mance
e e s o impo an aspec s such as p oduc quali y, clien
sa is ac ion and de elopmen o new p oduc s. In he same
line Den Ha og e al. (2013) ha e es ablished ha he exis ing
measu es o sensed pe cei ed o ganiza ional pe o mance
gene ally co e di e en pe o mance aspec s, adding o
p oduc quali y o clien se ice and sa is ac ion o he aspec s
ela ed o epu a ion. In his esea ch s udy and acco ding o
Bon is e al. (2002) he i ems used o measu e he “pe cei ed
pe o mance measu es” cons uc include he classi ica ion o
he u u e pe spec i es o he business, mee ing clien s’ needs
and global assessmen o business pe o mance.
In his esea ch s udy, we expec HR p ac ices ocusing on
commi men o co ela e posi i ely wi h pe cei ed
o ganiza ional pe o mance measu es – global assessmen o
business pe o mance, mee ing clien s’ needs, u u e
pe spec i e o he company and epu a ion. I is expec ed ha
commi men -o ien ed p ac ices will encou age employee
pa icipa ion in decision-making in hei g oup, sea ch o
ag eemen s among hem and p e en ion o in e pe sonal
con lic s (Bowen & Os o , 2004). This is likely o a o g ea e
employee in eg a ion and he de elopmen o posi i e
a i udes and beha io s owa ds he o ganiza ion. Such
in eg a ion and beha io s a e expec ed o be associa ed o
highe -pe o mance le els (Den Ha og e al., 2013).
The e o e, we expec pe cei ed o ganiza ional pe o mance o
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
183
be highe whe e commi men -o ien ed HRM p ac ices a e
implemen ed.
2.3 Employee di e si y as mode a ing a iable
Al hough he li e a u e sugges s ha he e is a ela ionship
be ween HR p ac ices and o ganiza ional pe o mance, he e
a e s ill open ques ions on how his ela ion akes place, on
which a e he a iables exis ing in he “black box". The e o e,
he e is a need o u he esea ch s udies conside ing ac o s
ha may in e ene and/o mode a e he ela ion be ween HR
p ac ices and o ganiza ional pe o mance. Acco ding o Gues
(2011), he commi men le el de eloped by employees in
esponse o HR p ac ices canno be homogeneous. The
changing alues o he wo k o ce, he gene a ion hey belong
o, hei wo k expe ience and hei a ying pe sonal
ci cums ances may al e hei p io i ies. The alues and
easons o employees and he di e ences among hem mus
be aken in o accoun in esea ch o HRM and pe o mance.
Resea ch on wo k o ce di e si y has ocused on wo main
aspec s: demog aphic di e si y and human capi al di e si y.
O ganiza ional demog aphy di e si y esea che s ocus mainly
on cha ac e is ics ha a e isible such as age, o gende , o on
a ibu es ela ed o wo k, such as unc ional expe ience. The
heo y mos equen ly ollowed by di e si y esea che s
s a es ha a ia ions in he composi ion o wo k g oups o
eams a ec s g oup p ocesses, and his p ocess has an impac
on pe o mance (Williams & O’Reilly, 1998). This app oach
elies on he knowledge-based iew and he decision-making
pe spec i e, also called Cogni i e Resou ce Theo y (Woeh ,
A ciniega & Poling, 2013); i de ends ha di e se alues in
employees will con ibu e o a be e i m pe o mance.
Posi i e e ec s on pe o mance a e achie ed when
o ganiza ions adap hei managemen p ac ices o he
cha ac e is ics o hei wo k o ce, hus imp o ing
managemen and pe o mance (Benschop, 2001). Howe e ,
based on he li e a u e e iewed, in ou s udy, we ha e
ocused ou a en ion on human capi al di e si y (educa ion
le el, unc ional specializa ion, leng h o se ice and enu e).
We use hese ac o s o es ablish he mode a ing e ec o
employee di e si y on he ela ionship be ween HR policies
ocusing on commi men and o ganiza ional pe o mance. The
main eason as we men ioned in he in oduc ion is ha we
ha e no ound any esea ch s udy ha has used hese ac o s
as mode a o a iables.
The e o e, in Figu e 1, we summa ize he heo e ical
backg ound and esea ch model o he li e a u e o be
analyzed.
Figu e 1 - Resea ch model
2.4 Hypo heses
2.4.1 Educa ion le el
The concep o educa ion le el e e s o o mal educa ion,
including p ima y, seconda y and highe educa ion. This is one
o he indica o s ha has been mos equen ly used o
measu e human capi al based on he academic le el
comple ed. We can expec educa ional di e si y o be
associa ed posi i ely o g oup pe o mance, al hough his
impac may no be as di ec as i could be ini ially expec ed,
since i may indica e dis inc educa ion le els o di e en
educa ional expe iences (Williams & O’Reilly, 1998). Some
esea ch s udies sugges ha educa ional he e ogenei y is
associa ed o alle company g ow h a es and o s a egic
ini ia i es. Dong eng (2013) inds and e i ies ha highe -
educa ion di e si y b ing on highe con lic le els as well as
be e eam pe o mance, since he g ea e he educa ional
di e ences, he easie i will be o encou age use ul discussion
leading o an imp o emen o employee pe o mance.
Educa ional di e ences a ec he way in which in o ma ion
and i s sou ces a e accessed, pe cei ed and analyzed; his will
in luence decision-making, and he way indi iduals pe cei e
he eali y a ound hem. The e o e, i is likely ha he g ea e
he educa ional he e ogenei y, he mo e di e en he
pe cep ions o employees and, as a consequence, he mo e
di e en hei con ibu ion o wo k g oups will be. This
ci cums ance may ha e an impac on he ela ionship be ween
HRMp and OP.
The e o e, we p opose he ollowing hypo hesis:
H1: Educa ion le el di e si y will mode a e he ela ionship
be ween HRM p ac ices and OP.
2.4.2 Func ional specializa ion
The e m specializa ion e e s o he wo k backg ound o an
indi idual in an o ganiza ion, such as inances, ope a ions,
sales… F om his pe spec i e i is conside ed ha he di ision
o labo in an o ganiza ion leads o a ce ain g oupings o
asks, and he e o e, o knowledge associa ed o he unc ional
depa men s in which employees ha e gained hei
expe ience. Thus, specializa ion le el p o ides us wi h a use ul
and easily-accessible indica o o he expe ience gained by
indi iduals, and consequen ly, o he eam in which hey
collabo a e (Bunde son, 2003). The ela ionship be ween
HRM p ac ices
Selec ion, T aining, E alua ion, Flexible emune a ion
Job design, Bidi ec ional communica ion, Job s abili y
Equali y and Job quali y
O ganiza ional
pe o mance
Educa ion Specializa ion
Leng h o enu e
Leng h o se ice
H1 H2
H4
H3
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
184
indi iduals wi h di e en unc ional backg ounds in an
o ganiza ion will impac on hei decisions, in ol emen , e c.
especially in he case o mul i unc ional eams. Some
labo a o y and ield s udies ha e ound ha unc ional
di e si y has a posi i e impac on employee pe o mance
(Williams and O’Reilly, 1998). Consequen ly, om a ce ain
le el o specializa ion on, con lic le els among wo k eam
membe s may inc ease, hinde ing in e nal communica ion and
coo dina ion among i s membe s (Jehn and Bez uko a, 2004).
Al hough esea ch s udies ha e shown he p esence o e ec s
be ween unc ional specializa ion di e si y and pe o mance,
some esul s may be con adic o y.
The e o e, we p opose he ollowing hypo hesis:
H2: Func ional specializa ion di e si y will mode a e he
ela ionship be ween HRM p ac ices and OP.
2.4.3 Leng h o se ice and enu e di e si y
This concep e e s o he ime an employee has been wo king
in an o ganiza ion, and i has been used in p e ious esea ch
s udies (Pe e i and Neg o, 2007). Leng h o se ice has been
associa ed o how amilia employees a e wi h he policies,
p ocedu es and o he ci cums ance o he posi ions, eams
and depa men s o he o ganiza ions o which hey wo k o
ha e wo ked. Membe s ha e go di e en iews o hei
o ganiza ion and o i s p ocesses. Employees wi h sho e
imes o se ice a e mo e lexible, bu hey end o be c i ical
owa ds exis ing p ocesses, and his may gene a e con lic s
among employees. Addi ionally, i is no in equen ha hey
a e s e eo yped and/o excluded by mo e e e an employees
(Pe e i and Neg o, 2007).
Lampel, Shamsie & Lan (2006) say ha he success o his ype
o di e si y will depend on he combina ion o alues sha ed
by he o ganiza ion. Whe e he e is li le di e ence in
employee leng h o se ice, he e a e ewe in e up ions in
communica ion and ewe ask con lic s, as a esul o he
p esence o codes, ules and alues ha a e commonly known
o wo k eam membe s. High dispa i y among membe s has
been ela ed o low le els o g oup cohesion, g ea e use o
o mal communica ion, as well as highe c ea i i y le els in
esea ch ac i i ies.
Some esea ch s udies sugges ha leng h o se ice di e si y
has a posi i e impac on pe o mance (Ha ison e al., 2002).
This a gumen has been suppo ed by Williams and O’Reilly
(1998), who a gue ha he e will be a posi i e ela ion,
p o ided ha o ganiza ions ake in o accoun any nega i e
e ec s esul ing om di e si y. They also sugges ha a
g ea e di e si y in enu e may ha e an impac on he
ou comes expec ed om he di e en ac ions and HRM
p ac ices. Howe e , hey do no s a e whe he leng h o
se ice a he company and enu e ha e he same e ec s.
F om his pe spec i e, we p opose he ollowing hypo hesis:
H3: Leng h o se ice di e si y will mode a e he
ela ionship be ween HRM p ac ices and OP.
H4: Tenu e di e si y will mode a e he ela ionship
be ween HRM p ac ices and OP.
3. Me hodology
3.1 Popula ion and sample
The ques ionnai e me hodology was adop ed in esea ch s udy
on human esou ces managemen . Conce ning he sample, he
ques ionnai es ha e been answe ed by HR manage s, HR
gene al manage s and gene al manage s.
Al hough he gene al s anda d in HR esea ch is o use one
single in o man in each company due o he di icul y o
ob aining mul iple in o man s, in ac , we ecognize he
possibili y o single-me hod bias in ou da a. Consequen ly, we
inc ease he con idence in ou da a by: (1) unde aking a ac o
analysis which showed he absence o a single gene al ac o o
accoun o mos o he co a iances in ou a iables, indica ing
he absence o common me hod a iance p oblems o ou da a;
(2) pe sonally iden i ying quali ied esponden s (e.g. HRM
manage s). P e ious li e a u e has shown ha he iews o a
single bu well-quali ied in o man may be e cap u e a i m’s
beha io as compa ed o he iews o se e al esponden s in he
case o o ganiza ions whe e ele an decisions a e o en highly
cen alized (T igue o-Sánchez e al., 2012).
The selec ed people in each company we e con ac ed by
elephone. They we e explained he impo ance o aking pa
in he s udy and also i s use ulness. Whe e equi ed, we make
he commi men o send hem he ou comes o ou esea ch
s udy. They we e also assu ed ha all in o ma ion would be
ea ed in a con iden ial, comp ehensi e and anonymous
manne . Finally, we highligh ed he impo ance o any
sugges ions he in e iewees may p opose o us, and ou
g a i ude o hei pa icipa ion. All o hese aspec s we e
emphasized in he in oduc o y le e which was subsequen ly
sen along wi h he ques ionnai e and a p epaid en elope o
e u ning upon comple ion.
The companies we e selec ed om he SABI da abase (Ibe ian
Balances Analysis Sys em). Among he egis e ed companies,
1300 had be ween 100 and 2000 employees, acco ding o he
da a egis e ed in 2007, and 1169 had been es ablished be o e
2003. The esul ing popula ion was 902 wi h a well-balanced
ep esen a ion o all p oduc i e sec o s. A o al o 103
ques ionnai es we e e u ned in di e en o ms: ia e-mail,
pos al mail, and pe sonal in e iews a hei o ganiza ions. This
ep esen s a esponse o 11.42%. Wi h ega d o he na u e o
companies, 93% o hem belong o he p i a e sec o and he
emaining, o he public sec o , in o he wo ds; hey belong o
he Spanish go e nmen . The ollowing ables (Table 1 and 2)
summa ize he desc ip i e s a is ics o he uni analysis.
Table 1 - Technical da a o he su ey
Popula ion size
902 Fi ms
Uni e se
HR di ec o and CEO.
Place o S udy
Andalusia (Spain)
Sampling me hod
Su ey: ia e-mail, pos al su ey,
and pe sonal in e iews
Sample size
102 su eys (excluding hose ha
we e illegible and/o incomple e.)
Time pe iod
June 2009 o Oc obe 2009
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
185
Table 2 - S a is ics-cha ac e is ics o he sample ( i ms and in o man s)
Fi m cha ac e is ics
Yea s o ope a ion
(Mean) - 20.90
Numbe o employees
(Mean) - 218.03
To al Asse s (las h ee yea s) in Eu os
(€) - 37 051 984
Business Owne ship
(P i a e) 93%
Numbe o i ms
102
In o man cha ac e is ics
HR depa men manage s
68%
Gene al manage s - 34%
Gende
Male - 66%
Female - 34%
3.2 Measu emen o a iables
A li e a u e e iew was ca ied ou in o de o ob ain a eliable
measu emen scale, using ools b oadly alida ed and
con as ed in p e ious esea ch s udies. The i ems adap ed
om he English li e a u e we e ansla ed om English in o
Spanish by wo na i e Spanish speake s who a e amilia wi h
he HRM e minology in o de o p e en any ambigui y in
measu emen scales.
3.2.1 O ganiza ional pe o mance
To measu e subjec i e, o pe cei ed, o ganiza ional
pe o mance, a educed e sion o he measu emen scale
de eloped by Bon is e al. (2002) o e alua e o ganiza ions as a
whole was used. This scale comp ises only ou i ems. One
example o he i ems used is: ou o ganiza ion is well-
espec ed wi hin he indus y. C onbach´s Alpha (CA)
coe icien o he scale was 0.878, i s Composi e Reliabili y
(CR) = 0.865 and i s A e age Va iance Ex ac ed (AVE) = 0.617.
3.2.2 HRM p ac ices
The a iables used o measu e HRM p ac ices a e suppo ed by
he exis ing li e a u e. We ha e used he measu ing i ems
de eloped by T igue o-Sánchez e al. (2012), con i med o he
Spanish case, wi h a 1-7, 38-i em Like - ype scale. As we
expec ed all he i ems in he scale loaded on a single ac o using
P incipal Componen Analysis. These i ems had a CA o a 0.960
and a CR = 0.951 and an AVE= 0.654. In he scale, 1 indica es
high con ol and low commi men o he o ganiza ion, and 7
indica es high commi men and low con ol. Fo his a iable,
aspec s such as pe sonnel selec ion, aining, e alua ion, wage
lexibili y, job design, communica ion le el, job s abili y, equal
oppo uni ies and he quali y o human esou ces managemen
ha e been measu ed.
3.2.3 Di e si y ac o s
Cons uc s such as Le el o Educa ion (4 i ems), Le el o
Specializa ion (5 i ems), Leng h o Se ice wi hin he
o ganiza ion (2 i ems) and Tenu e (2 i ems) we e e alua ed
(see T igue o-Sánchez e al., 2011).
The ou ac o s o di e si y we e inco po a ed in o he model
by means o Blau’s index o he e ogenei y, which is b oadly
accep ed in social and beha io al sciences. The nex
exp ession (Fo mula 1) shows Blau’s index (1977).
2
1i
PI
Fo mula 1
(I) = Blau’s index
(P) Is he p opo ion o people in each ca ego y s udied and
(i) The numbe o ca ego ies obse ed
[I mus be men ioned ha we we e asking o he
app oxima e pe cen age in each ca ego y].
In o he o ge a pa simonious model, once ha he Blau’s
index was ob ained o each ac o o di e si y, hey we e
ans o med in o 7-poin Like scale whe e: 1 = low
he e ogenei y, 7 = high he e ogenei y.
4. Da a analysis and esul s
Fi s , we analyzed ac o s uc u e and eliabili y o all scales using
PASW S a is ics 18 so wa e (IBM SPSS So wa e, 2009). Fac o
analysis and scale eliabili y p ocedu es wi hin he PASW S a is ics
18 so wa e we e ca ied ou o assess he dimensionali y and
in e nal homogenei y o he scales. The explo a o y ac o analysis
(p incipal axis ac o ing wi h a imax o a ion) iden i ied HRM
p ac ices and pe cei ed o ganiza ional pe o mance as ac o s
wi h good ac o loadings, which was consis en wi h he model
we p oposed. As desc ibed in he “C i e ion a iables” and
“P edic o a iables” sec ions, he eliabili ies o all he cons uc s
exceeded Nunnally’s (1978) ecommended le el o 0.70. Second,
ollowing Dawson’s (2013) ecommenda ions, we c ea ed he
mean-cen e ing (i.e., sub ac ing he mean om he alue o he
o iginal a iable so ha i has a mean o 0) o he HRM p ac ices
and o ganiza ional pe o mance cons uc s. This p ocedu e
ensu es ha he (uns anda dized) eg ession coe icien s o he
main e ec s can be in e p e ed di ec ly in e ms o he o iginal
a iables. I is impo an o men ion ha he e a e o he
al e na i e p ocesses. Howe e , as Dawson (2013) s a es, such
p ocedu es lead o iden ical esul s. Finally, we ha e ca ied ou
he mode a ed eg ession analyses using an online esou ce
(h p://www.a hayes.com/index.h ml).
Di e si y in he leng h o se ice is posi i ely co ela ed o
enu e (p=0.690**), which could be unde s ood ha when
o ganiza ions coexis employees, wi h di e en an iques,
di e si ies also exis in he leng h o he jobs (see Table 3).
Table 3 - Desc ip i e s a is ics and bi a ia e co ela ions o all a iables
No e: *. Co ela ion is signi ican a he le el 0.05 (bila e al). **. Co ela ion is signi ican a he le el 0.01 (bila e al). **p<0.01; *p<0.05; SD=
S anda d De ia ion; N= numbe o companies.
Va iables
N
Mean
SD
1
2
3
4
5
6
1. Educa ion
102
6.070
1.286
1
2. Specializa ion
102
3.270
1.549
-0.004
1
3. Leng h o se ice
102
3.240
1.678
0.171
-0.227*
1
4. Tenu e
102
3.580
1.601
0.059
-0.240*
0.690**
1
5. O ganiza ional pe o mance
102
5.460
0.870
-0.006
0.150
0.018
.0134
1
6. HRM-p ac ices
102
4.194
1.309
0.029
0.287**
-0.165
-0.117
0.579**
1
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
186
Howe e , he e is a nega i e co ela ion be ween unc ional
di e si y and enu e (p=-0.240*), and also be ween unc ional
di e si y and leng h o se ice di e si y (p=-0.227*). This
means ha in companies wi h ewe unc ional a eas in hei
s uc u e he e is a end o highe employee mobili y
be ween he di e en a eas o unc ions, which could enhance
hei poly alence. Mo eo e , we could deduce ha as
companies ha e mo e poly alen s uc u es hey also ha e
mo e lexible HR s uc u es in place, in e ms o bo h
unc ional mobili y and en olmen a es, esul ing a mo e
balanced wo k o ce in e ms o new employees and
pe manen employees. Educa ion le el di e si y does no
co ela e o any o he o he human capi al a iables.
Rega ding he dependen a iable, ou da a show ha he e is
no signi ican co ela ion o he o he a iables in he s udy,
excep o HRMp (p=0.579**) which s eng hens ou
econome ic model and con i ms he esul s ound in he
li e a u e e iew. Mo eo e , HRMp seem o be co ela ed o
unc ional di e si y (p=0.287**). This indica es he impo ance
o he unc ional composi ion o he wo king g oups in HR
policy and managemen .
The mode a ing e ec s o he a iables unde s udy ha e been
shown o human capi al di e si y a iables: mode a ing
e ec s o educa ion (B=0.28**; RA=4.25), unc ional
specializa ion (B=-0.26***; RA=1.13) and leng h o se ice
(B=0.63***; RA=5.62), whe eas enu e has shown no
mode a ing e ec (b=0.52; RA=0.00).
5. Discussion, conclusions and implica ions
Among he ou human capi al a iables po en ially
mode a ing he “HRM p ac ices-OP” ela ion, only one o hem
has no achie ed enough s a is ical signi icance o be
con i med as such ( enu e). This would be in line wi h Pelled’s
(1996) posi ion, since he decla es ha human capi al
indica o s ha e a s ong ela ion wi h pe o mance.
Model 1 (Table 4) con i ms hypo hesis H1 since i shows ha
educa ion di e si y is a condi ioning ac o o he “HRM
p ac ices-OP” ela ion; indeed, i we analyze o e all esul s his
is one o he a iables mos s ongly condi ioning such ela ion
(β=0.28**; RA=4.25), and i does so di ec ly; i.e he mo e
he e ogeneous educa ion le els a e he be e ac o s o ge a
g ea o ganiza ional pe o mance which is achie ed when HR
policies ocusing on commi men . This is in line wi h he
indings o Wiese ma and Ban el (1992) showing ha g ea e
educa ion he e ogenei y is associa ed o highe co po a e
g ow h a es.
Table 4 - Resul s o mode a ed eg ession analyses o Human capi al di e si y and ou come a iables
Model 1
Educa ion
Model 2
Specializa ion
Model 3
Leng h o Se ice
Model 4
Tenu e
Cons an
5.8**
5.31**
4.17**
3.92
Independen a iables
0.28**
-0.26**
0.63**
0.52
HRM-p ac ices
0.37**
-0.50**
0.44**
0.41
Independen a iables x HRM-p ac ices
0.44**
0.75**
-0.25**
-0.31
F
17.26**
21.87**
23.95**
21.71
∆F
1.55
10.79
12.72
3.778
R
0.59
0.63
0.65
0.632
R2
0.35
0.40
0.423
0.399
∆R2
0.01
0.07
0.075
0.023
Region o accep ance (RA)
4.25 (yes)
1.13 (yes)
5.62(yes)
0.00 (NS)
E ec s epo ed a e s anda dized B; *p. 0.05; **p.0.01. 1. RA: The Johnson-Neyman p ocedu e is used o ind a “REGION” o S a is ical Signi icance. NS: no-
signi icance
Model 2 indica es ha he mode a ing a iable wo k
specializa ion condi ions he “HRM p ac ices-OP” ela ion
al hough wi h lowe in ensi y han he a iable educa ion (β=-
0.26***RA= 1.13). In his case, ou indings sugges ha he
mo e homogeneous unc ional specializa ion le els a e, he
be e esul will be achie ed by commi men - ocused HR
p ac ices, which con i ms he hypo hesis H2. In his case, ou
indings a e in line wi h hose o Jehn and Bez uko a (2004),
who conside ha beyond a ce ain le el o specializa ion he
in ensi y o con lic s be ween wo k eam membe s may
inc ease, making in e nal communica ion and coo dina ion
among indi iduals mo e complex. Thus, be e esul s a e
achie ed wi h lowe le els o unc ional specializa ion,
some hing ha could be explained by he cu en demand o
o ganiza ions wi h a high poly alence.
Acco ding o he esul s shown in Model 3, he mos impo an
di e si y a iable in e ms o condi ioning he “HRM p ac ices-
OP” ela ion is leng h o se ice (β=0.63; RA=5.62), which leads
us o accep hypo hesis H3. Thus, di e si y in leng h o se ice
wi h he company p o ides he bes ou comes in he applica ion
o HR p ac ices ocusing on commi men . This may be due, as
Williams and O’Reilly (1998) ha e sugges ed, o he ac ha his
he e ogenei y p o ides an in o ma ion di e si y ha enhances
g oup pe o mance, o ha in he companies unde s udy,
he e ogenei y in leng h o se ice is no associa ed o emo ional
con lic s, since he e is a se o alues sha ed by all membe s o
he o ganiza ion (Lampel e al., 2006).
Finally, he esul s o Model 4 on di e si y in enu e lead us o
ejec hypo hesis H4 since i s alues a e ou o he accep a ion
egion o a model wi h a mode a ing e ec (β=0.52; RA=0.00).
This allows us o conclude ha di e ences in leng h o enu e
among employees a e no a de e minan ac o in
o ganiza ional pe o mance as a as implemen a ion o
commi men - ocused HR p ac ices is conce ned.
In conclusion, we ha e ound ha he indica o s ela ed o
human capi al p o ide a s ong mode a ing e ec in he
T igue o-Sánchez, R., Peña-Vinces, J. C. & Sánchez-Apellániz, M. (2016). Tou ism & Managemen S udies, 12(1), 181-187
187
ela ion be ween HR p ac ices and o ganiza ional
pe o mance, as al eady an icipa ed by Woeh e al. (2003)
and in line wi h Pelled (1996); hese indica o s a e unc ional
specializa ion, leng h o se ice and educa ion.
The esul s o his s udy ha e a numbe o impo an
implica ions o u u e p ac ice. The bes esul s in he
applica ion o human- esou ce policies ha ocus on
commi men a e achie ed when he e is a a ie y in leng h o
se ice. Companies mus conside ha a la ge di e si y in he
specializa ion le el o hei employees may make i di icul o
adap o changes and hinde o ganiza ional pe o mance. As
pe cei ed by hei manage s, a highe di e si y in leng h o
se ice wi h he company and a highe di e si y in educa ion
le els will con ibu e o business success since hey can ely on
a wo k o ce commi ed o an o ganiza ion in which hey can
de elop a p o essional ca ee .
A limi a ion o he s udy is ha we mus be cau ious wi h i s
indings since he s udy has ocused on a speci ic egion wi h
i s own pa icula cul u e. Likewise, i would be in e es ing o
dwell u he in o each o he HR p ac ices in o de o know
mo e accu a ely hei ac ual con ibu ion o he model.
Re e ences
Al-Jaba i, M. (2013). Fac o s a ec ing human esou ce p ac ices in a
sample o di e si ied Pales inian o ganiza ions. In José An ónio C.
San os, Filipa Pe digão & Paulo Águas (Eds.) Book o P oceedings –
Tou ism and Managemen S udies In e na ional Con e ence Alga e
2012, ol.2 (pp. 587-593). Fa o: Uni e si y o he Alga e.
Ba ena-Ma ínez, J., López-Fe nández, M. & Rome o-Fe nández, P.M.
(2013). Towa ds he seeking o HRM policies wi h a socially
esponsible o ien a ion: a compa a i e analysis be ween Ibex-35 i ms
and Fo une’s op 50 mos admi ed companies. In José An ónio C.
San os, Filipa Pe digão & Paulo Águas (Eds.) Book o P oceedings –
Tou ism and Managemen S udies In e na ional Con e ence Alga e
2012, ol.2 (pp. 488-501). Fa o: Uni e si y o he Alga e.
Becke , B. (1964). Human Capi al: A heo e ical and empi ical analysis
wi h special e e ence o educa ion. New Yo k: Na ional Bu eau o
Economic Resea ch.
Benschop, Y.W. (2001). P ide, p ejudice and pe o mance: ela ions
be ween HRM, di e si y and pe o mance. In e na ional Jou nal o
Human Resou ce Managemen , 12(7), 1166-1181.
Blau, P.M. (1977). Inequali y and He e ogenei y: A P imi i e Theo y o
Social S uc u e. New Yo k: F ee P ess.
Bon is, N., C ossan, M.M., & Hulland, J. (2002). Managing an
o ganiza ional lea ning sys em by aligning s ocks and lows. Jou nal o
Managemen S udies, 39(4), 437-469.
Bowen, D.E., & Os o , C. (2004). Unde s anding HRM- i m
pe o mance linkages: The ole o he ‘s engh ’ o he HRM sys ems.
Academy o Managemen Jou nal, 29, 203-221.
Bunde son, J.S. (2003). Team membe unc ional backg ound and
in ol emen in managemen eams: Di ec e ec s and he mode a ing ole
o powe cen aliza ion. Academy o Managemen Jou nal, 46(4), 458-474.
C opanzano, R., & Mi chell, M.S. (2005). Social exchange heo y: an
in e disciplina y e iew. Jou nal o Managemen , 31, 874-900.
Dawson, JF. (2013). Mode a ion in Managemen Resea ch: Wha , Why,
When, and How. Jou nal o Business and Psychology. doi
10.1007/s10869-013-9308-7
Den Ha og, D.N., Boon, C.; Ve bu g, R.M. & C oon, M.A. (2013). HRM,
Communica ion, Sa is ac ion, and Pe cei ed Pe o mace; A C oss-le el
es . Jou nal o Managemen , 39, 1637-1665.
Dong eng, S. (2013). Resea ch on di e si y, con lic s and pe o mance
in c ea i e eam managemen . A ican Jou nal o Business
Managemen , 7(39), 4139-4148.
Gues , D.E. (1997). Human Resou ce Managemen and Pe o mance: a
Re iew and Resea ch agenda. The In e na ional Jou nal o Human
Resou ce Managemen , 8(3), 263-276.
Gues , D.E. (2011). Human esou ce managemen and pe o mance:
s ill sea ching o some answe s. Human Resou ce Managemen
Jou nal, 21(1), 3-13.
Ha ison, D.A., P ice, K.H., Ga in, J.H., & Flo ey, A.T. (2002). Time,
eams, and ask pe o mance: Changing e ec s o su ace –and deep-
le el di e si y on g oup unc ioning. Academy o Managemen Jou nal,
45(5), 1029-1045.
Jackson, S.E., Joshi, A., & E ha d , N.L. (2003). Recen esea ch on eam
and o ganiza ional di e si y: SWOT analysis and implica ions. Jou nal
o Managemen , 29(6), 801-830.
Jehn, K.A., & Bez uko a, K. (2004). A ield s udy o g oup di e si y:
Wo kg oup con ex and pe o mance. Jou nal o O ganiza ional
Beha io , 25(6), 703-729.
Lampel, J., Shamsie, J., & Lan , T.K. (2006). Towa d a deepe
unde s anding o cul u al indus ies. The Business o Cul u e, 3-14.
Ne es, C., Gal ão, A. & Pe ei a, F. (2013). Guidelines in Human
Resou ces Managemen . In José An ónio C. San os, Filipa Pe digão &
Paulo Águas (Eds.) Book o P oceedings – Tou ism and Managemen
S udies In e na ional Con e ence Alga e 2012, ol.2 (pp. 420-429).
Fa o: Uni e si y o he Alga e.
Nunnally, J. (1978), Psychome ic Theo y, 2nd ed., McG aw-Hill, New
Yo k, NY.
Os o , C., & Bowen, D.E. (2000). Mo ing HR o a highe le el: HR
p ac ices and o ganiza ional e ec i eness. In Klein, K.J., Kozlowski,
S.W.J. (Eds.) Mul ile el heo y, esea ch, and me hods in o ganiza ions:
ounda ions, ex ensions, and new di ec ions (pp. 211-266). San
F ancisco: Jossey-Bass.
Pelled, L. (1996). Demog aphic di e si y, con lic , and wo k g oup
ou comes: an in e ening p ocess heo y. O ganiza ion Science, 7(6),
615-631.
Pe e i, F. & Neg o, G. (2007). Mixing gen es and ma ching people: a
s udy in inno a ion and eam composi ion in Hollywood.
T igue o-Sánchez, R., Peña-Vinces, J.C., & Sánchez-Apellániz, M. (2011).
HRM in Spain i s di e si y and he ole o o ganiza ional cul u e: an
empi ical s udy. Eu opean Jou nal o Social Sciences, 26(3), 389-407.
T igue o-Sánchez, R., Peña-Vinces, J.C., González-Rendón, M., &
Sánchez-Apellániz, M. (2012). Human Resou ce Managemen P ac ices
aimed o seek he commi men o employees on he inancial and non-
inancial (subjec i e) pe o mance in Spanish Fi ms: An empi ical
con ibu ion. Jou nal o Economics, Finance and Adminis a i e,
17(32), 17-30.
Van Knippenbe g, D., & Schippe s, M.C. (2007). Wo k g oup di e si y.
Annual Re iew o Psychology, 58, 515-541.
Wal on, R. (1985). F om con ol o commi men in he wo kplace.
Ha a d Business Re iew, 63(2), 77-84.
Williams, K.Y. & O'Reilly III, C.A. (1998). Demog aphy and di e si y in
o ganiza ions: A e iew o 40 yea s o esea ch. In B. S aw & R. Su on
(eds.) Resea ch in O ganiza ional Beha io , 20, G eenwich, CT: JAI
P ess 1998, 77-140.
Woeh , D.J., A ciniega, L.M., & Poling, T.L. (2013). Explo ing he e ec s
o alue di e si y on eam e ec i eness. Jou nal o Business
Psychology, 28, 197-121.
W igh , P.M., & Nishii, L.H. (2006). S a egic HRM and o ganiza ional
beha iou : in eg a ing mul iple le els o analysis, CAHRS Wo king
pape se ies, 05, h p://il .co neli.edu/CAHRS.
A icle his o y:
Submi ed: 20.06.2015
Recei ed in e ised o m: 18.12.2015
Accep ed: 19.12.2015
