A Con olled Expe imen o E alua e he E ec s
o Mind ulness in So wa e Enginee ing
Bea iz Be ná dez
bea @us.es
Amado Du án
amado @us.es
José A. Pa ejo
japa [email protected]
An onio Ruiz–Co és
a [email protected]
Depa men o Compu e Languages and Sys ems
Uni e si y o Se ille, Spain
ABSTRACT
Con ex . Many epo s suppo he ac ha some psycho–social
aspec s o so wa e enginee s a e key ac o s o he quali y o he
so wa e de elopmen p ocess and i s esul ing p oduc s. Based on
he expe ience o some o he au ho s a e mo e han a yea o
p ac ising mind ulness—a medi a ion echnique aimed o inc ease
clea ness o mind and awa eness—we guessed ha i could be in-
e es ing o empi ically e alua e whe he mind ulness a ec s posi-
i ely no only he beha iou bu also he p o essional pe o mance
o so wa e enginee s.
Goal. In his pape , we p esen a quasi–expe imen ca ied ou a
he Uni e si y o Se ille o e alua e whe he So wa e Enginee -
ing & In o ma ion Sys ems s uden s enhance hei concep ual mod-
elling skills a e he con inued daily p ac ice o mind ulness du ing
ou weeks.
Me hod. S uden s we e di ided in o wo g oups: one g oup p ac-
ised mind ulness, and he o he — he con ol g oup—we e ained
in public speaking. In o de o s udy he possible cause–and–e ec
ela ionship, e ec i eness ( he a e o model elemen s co ec ly
iden i ied) and e iciency ( he numbe o model elemen s co ec ly
iden i ied pe uni o ime) o he s uden s de eloping concep ual
modelling exe cises we e measu ed be o e and a e aking he
mind- ulness and public speaking sessions.
Resul s. The expe imen esul s ha e e ealed ha he s uden s who
p ac ised mind ulness ha e become mo e e icien in de elop-ing
concep ual models han hose who a ended he public speaking
sessions. Wi h espec o e ec i eness, some enhancemen ha e
been obse ed, al hough no as signi ican as in he case o e i-
ciency.
Conclusions. This ising end in e ec i eness sugges s ha he
numbe o sessions could ha e been insu icien and ha a longe
pe iod o sessions could ha e also enhanced e ec i eness signi i-
can ly.
Keywo ds
Mind ulness, Quasi–expe imen , Concep ual Modelling
1. INTRODUCTION
So wa e quali y emains as one o he mos impo an challenges
in So wa e Enginee ing (SE). Bo h so wa e indus y and academia
ha e made eal e o s o imp o e so wa e quali y and a e awa e
ha he main issues a e no only echnical, bu also o ganiza ional
and psycho–social [15].
Since 1996, he S andish G oup has been analysing c i ical ac-
o s in so wa e de elopmen , published as he well–known CHAOS
epo s [31]. Al hough some ac o s ha e o do wi h p ojec man-
agemen and echnology, many o hem a e s ongly social such
as use in ol emen , execu i e managemen suppo , clea s a e-
men o equi emen s, and ha d–wo king, ocused s a . In addi-
ion, emo ional ma u i y—de ined as he skill o be sel –awa e, so-
cially awa e, sel –managed and o manage ela ionships—has e-
cen ly become one o he op– en c i ical ac o s in he las edi-
ion o he epo [32]. Because o i s impo ance, he S andish
G oup has de eloped a p og am o help so wa e de elopmen o -
ganiza ions iden i y ouble a eas and imp o e he emo ional ma u-
i y o hei employees [33]. In addi ion, ecen ly published wo ks
[1] ha e posi i ely ela ed some aspec s o he pe sonali y o so -
wa e enginee s such as ex o e sion, ag eeableness, openness o
new ideas, and conscien iousness, wi h he abili y o in eg a e in o
a eam and also on aspec s o code quali y.
I hus seems easonable o educa e so wa e enginee s in some
psycho–social skills ha p omo e he smoo h unning o he so -
wa e de elopmen p ocess. A possible app oach o achie ing his
goal is mind ulness [4], a medi a i e p ac ice ha has in luence on
some aspec s o pe sonali y which a e ele an o he p oblem a
hand. Fo example, some wo ks ha e al eady ecommended he
p ac ice o mind ulness o p omo e he ideal wo k en i onmen in
agile de elopmen eams, i.e. daily mee ings, coope a ion wi h cus-
ome s and use s (C&Us), and a co dial a mosphe e be ween he
membe s o he de elopmen eam [36, 18].
1.1 Psycho–social aspec s in SE
Psycho–social aspec s in SE ha e a e y special in luence in
he Requi emen s Enginee ing (RE) phase, since in e ac ion wi h
C&Us is mo e c i ical o he p ojec success han in any o he
phase [7]. I is essen ial o pu onesel in he shoes o C&Us in
o de o unde s and hei posi ion and he needs o be sa is ied by
he so wa e sys em o be de eloped. In he book abou mind ul-
ness [17], Lange includes a quo e o Analysis Te minable and In-
e minable by Sigmund F eud (1937) ha , ela i ely speaking, can
be applied o he beginning o so wa e p ojec s when so wa e en-
ginee s ha e o unde s and he p oblem domain o he o ganisa ion
o which he so wa e is de eloped:
We know ha he i s s ep owa ds he in ellec ual
mas e y o he wo ld in which we li e is he disco e y
o gene al p inciples, ules and laws ha which b ing
o de in o chaos. By such men al ope a ions we sim-
pli y he wo ld o phenomena bu we canno a oid al-
si ying i in doing so, specially when we a e dealing
wi h p ocesses o de elopmen and change.
Mo e o en han no , so wa e de elope s a e no expe s in he
p oblem domain and he alsi ying phenomenon desc ibed by F eud
is likely o occu . The e o e, so wa e enginee s—o mo e speci -
ically equi emen s enginee s—should de elop he skills o unde -
s anding he p oblem domain as i ac ually is, being open–minded,
a oiding excessi e simpli ica ion, and ocusing all hei a en ion
on elici ing use s needs. Fu he mo e, in ou expe ience in SE ed-
uca ion, we ha e obse ed ha mos o ou s uden s a e e y in-
e es ed in compu e – ela ed echnology and p og amming bu a e
ha dly mo i a ed o pe o m no so echnical asks like concep ual
modelling, le alone mee ing wi h C&Us o elici ing o alida -
ing equi emen s—pa ially due o an in o e ed pe sonali y e y
common in so wa e enginee s [6]. They should be p o ided wi h
echniques o make hem eel mo e com o able in hose s ages o
he so wa e p ocess when psycho–social skills a e, a leas , as el-
e an as he echnical ones.
The aim o his pape is o p esen an expe imen al s udy in he
Uni e si y o Se ille in o de o analyse whe he he p ac ice o
echniques like mind ulness imp o e he SE p ocess, speci ically
he RE phase. Conc e ely, we ha e ca ied ou a quasi–expe imen
in which we ha e e alua ed whe he he p ac ice o mind ulness in
SE s uden s imp o es: (i) he quali y o concep ual models hey de-
elop; and (ii) hei concep ual modelling pe o mance. The con i -
ma ion o he hypo heses, by means a amily o expe imen s, would
lead us o conside he possibili y o in oducing his discipline as a
c oss– aining o u u e so wa e enginee s and, in he long e m,
o de elop mind ulness–based pe sonal g ow h p og ams o SE o -
ganiza ions.
The es o he pape is o ganized as ollows. In sec ion 2, mind-
ulness is b ie ly desc ibed and he bene i s o i s p ac ice in a ious
con ex s a e discussed. In Sec ion 3, he goals o he expe imen ,
subjec s, ma e ial, and expe imen al design a e o mally desc ibed;
also, a iables a e iden i ied and he expe imen al hypo heses a e
enuncia ed. In Sec ion 4, he expe imen execu ion is p esen ed. In
Sec ion 5, he desc ip i e s a is ics a e p esen ed oge he wi h da a
p epa a ion and hypo heses es . In Sec ion 6, he expe imen al e-
sul s, hei implica ions in he ield, and he h ea s o expe imen al
alidi y a e analysed. The gene aliza ion o he ob ained esul s is
also discussed in his sec ion. In he inal sec ion, we discuss he
conclusions and he u u e wo k.
2. THE PRACTICE OF MINDFULNESS
Mind ulness is an English wo d used o ansla e he Pali wo d
sa i, which means ha ing awa eness, a en ion, and emembe ing.
Mind ulness can be simply de ined as “ he uni e sal and basic hu-
man abili y o be awa e o he con en s o he mind momen –by–
momen ” [29]. The e m mind ulness is usually used wi h wo
complemen a y meanings. On he one hand, he p ac ice o o mal
S ep Desc ip ion
1 Si wi h you back s aigh in a elaxed way.
2 I he e is limi in ime, se an ala m.
3 Re iew you body ying o dis end ensions.
4 Pe o m h ee slow, deep b ea hs.
5 Begin o b ea he no mally, unp e en ious, ocusing on
b ea hing.
6 I hough comes, you expe ience i and gen ly le i go.
7 Enjoy, i i comes, a ew seconds o spaciousness and
openness in mind.
8 Back slowly o con ac wi h ex e nal eali y.
Table 1: Recipe o o mal mind ulness p ac ice (simpli ied)
mind ulness, in which a pe son—o a g oup o people—d aw away
o a quie place o a while ( en minu es a leas ) o medi a e on
b ea h. Du ing he medi a ion session, he pe son’s in en ion is o
keep he mind calmed using b ea hing as suppo . B ea hing is he
usual medi a ion suppo because o i s una oidabili y. O he com-
mon op ions a e ocusing on an objec using he sense o sigh , o
in su ounding sounds using he sense o hea ing. A o mal mind-
ulness session ecipe, based on [22, 29], is summa ized in Table
1. On he o he hand, in o mal mind ulness consis s o ans e ing
his s a e o consciousness o e e yday ac i i ies.
The p ac ice o o mal mind ulness leads o a high deg ee o
awa eness in daily li e, i.e. a g ea e abili y o s ay in he p esen
momen a he han ehashing he pas o imagining he u u e. This
is due o he abili y o obse e and abandon hough s wi hou iden-
i ying ou sel es wi h hem. Mind ulness helps us see he eali y
a ound us clea ly and sol e p oblems mo e e icien ly, since i e-
duces men al wande ing while pe o ming asks.
A a neu ological le el, hese e ec s o mind ulness a e explained
by some changes in b ain ac i i y. The p e on al co ex o he hu-
man b ain ha e e ol ed o de elop he abili y o sol e p oblems.
Fo me ly, hose p oblems used o ha e a physical na u e, like mak-
ing ools o weapons o haun ing. Today, he na u e o p oblems
is e y he e ogeneous, including no only abs ac bu also social
p oblems like helping o he people o each a consensus [26]. How-
e e , a hype ac i i y o he p e on al co ex has he undesi ed e -
ec s o umina ion and wande ing ha , pa adoxically, p e en us
om ha ing a clea ision o eali y and sol ing p oblems p op-
e ly. No ice ha he inc ease o his hype ac i i y is one o he
consequences o ou cu en ela ionship wi h echnology, i.e. he
ubiqui y o In e ne –connec ed de ices and he con inuous in e -
up ions hey gene a e. Neu oscien is s ha e demons a ed ha a
con inued p ac ice o mind ulness educe p e on al co ex hype -
ac i i y and inc eases he ac i i y o o he a eas o he b ain which
a e ac i e when conc e e asks a e pe o med [29].
2.1 E idences o mind ulness bene i s
The e a e many psychological s udies epo ing he bene i s o
mind ulness. Fo example, in [25] a quali a i e s udy examined
he in luence o mind ulness in a 15–week cou se wi h g adua e
s uden s. Pa icipan s epo ed an inc ease o hei men al cla i y,
o ganiza ion, awa eness, and accep ance o emo ions and pe sonal
issues. In [19], a con olled expe imen based on G adua e Reco d
Examina ions (GRE) and assessing e bal, quan i a i e and analy -
ical skills o measu e eading comp ehension, concen a ion, le el
o mind wande ing, and wo king memo y is desc ibed. The ou -
comes showed a g ea imp o emen in he g oup o people ha
a ended 15 mind ulness sessions. In [23], neu o–psychologis s
s udied he e ec s o mind ulness in 136 he e ogeneous pa ien s
showing ha , a e wo mon hs o daily 20–minu e p ac ice, a sig-
ni ican pe cen age expe ienced be e pe sonal well–being in e ms
o men al cla i y, equanimi y, wisdom and sel –compassion based
on s anda d heal h su eys (ques ionnai es). Mind ulness has also
been success ully used as he apeu ic ea men o indi iduals p one
o anxie y, s essed people, e c. [9, 12, 28]. In [8], a ca e ul li e a-
u e e iew compiles he main bene i s o mind ulness a a pe sonal
le el: sel –con ol, sel – ega d, equanimi y, sel –awa eness, sel –
insigh , in ui ion, egula ion emo ion, and well—being in gene al.
2.2 Mind ulness in he wo kplace
Wi h espec o social ela ions, he main epo ed bene i s o
mind ulness a e ela ed o empa hy, asse i eness, emo ion egu-
la ion, dec eased eac i i y, inc eased esponse lexibili y, coun-
selling skills, and emo ional in elligence in gene al [8]. Bene i s
o mind ulness in labou ela ions, especially in s ess ul wo king
a eas like heal h o eaching, ha e also been epo ed [21].
In he con ex o he so wa e indus y in Silicon Valley, he p ac-
ice o mind ulness is os e ed a guing imp o emen s in Employee
Rela ionship Managemen — eac ing less emo ionally, o exam-
ple. I has also been obse ed ha mind ulness imp o es mem-
o y and execu i e unc ions, and inc eases he abili y o concen-
a e on as –changing s imuli [27]. Pa icula ly in Google, engi-
nee Chade–Men Tan [35] is de eloping a aining p og am based
on mind ulness, using i o unde s and hei co–wo ke s’ mo i a-
ions, o enhance hei c ea i i y and p oduc i i y, and o cul i a e
emo ional in elligence.
In he Managemen Science and O ganisa ional Lea ning ields,
he e a e some p oposals o he so–called o ganisa ional mind ul-
ness, which de ines a se o p inciples like including de e ence o
expe ise, p eoccupa ion wi h ailu e, eluc ance o simpli y in e -
p e a ions, sensi i i y o ope a ions, and commi men o esilience
[37]. Those p inciples ha e a posi i e in luence in he capabili y
o change di ec ion, eac in a lexible manne , imp o e eliabil-
i y in con ol sys ems o inc ease inno a ion in in o ma ion ech-
nology [24, 34]. When applied o RE, o ganisa ional mind ulness
can help change w ong beha iou pa e ns. Fo example, in [24]
a case s udy is p esen ed in which RE asks a e de ined, asks in-
e iciencies a e iden i ied, and hei causes a e analysed using he
a o emen ioned o ganisa ional mind ulness p inciples. Using he
esul s o he analysis, he RE asks a e ede ined and, as a esul ,
ine iciencies dec ease and become unde con ol.
3. EXPERIMENT PLANNING
Acco ding o he SE expe imen al axonomy p esen ed in [30],
he expe imen desc ibed in his pape belongs o he Requi emen s
Analysis and Documen Comp ehension ca ego ies since i s main
objec i e is o asce ain i he p ac ice o mind ulness imp o es he
capabili y o s uden s in he pe o mance o a conc e e RE ask:
analysing an in e iew ansc ip and de eloping he co esponding
concep ual model. He ea e , he expe imen al p ocess is desc ibed
using he guidelines p oposed by Jedli schka e al. [13] wi h mino
changes.
3.1 Goals
Based on he ecommenda ions p oposed in [38] abou using he
GQM empla e in expe imen a ion in SE, he goal de ini ion o he
expe imen can be summa ized as:
Analyse he p ac ice o mind ulness
o he pu pose o e alua ing i s e ec s
wi h espec o he pe o mance o s uden s de eloping concep-
ual models, and he quali y o he de eloped models
om he poin o iew o he expe imen e s
in he con ex o sophomo e s uden s o he deg ee on So wa e
Enginee ing a he Uni e si y o Se ille.
3.2 Pa icipan s
Since we aim o s udy he e ec s o mind ulness on ea ly s ages
o he so wa e p ocess, subjec s a e sophomo es en olled in he In-
oduc ion o So wa e Enginee ing and In o ma ion Sys ems (ISEIS)
annual cou se. Speci ically, 32 s uden s (30 male, 2 emale) pa ic-
ipa ed in he expe imen du ing he i s semes e o he 2013–2014
academic yea . In o de o inc ease he mo i a ion, he s uden s
we e o e ed ex a poin s o aking pa in he expe imen .
The subjec s we e di ided in o wo g oups— he mind ulness g oup
and he public speaking g oup—depending on he mo i a ions and
p e e ences exp essed by he s uden s in an in e es ques ionnai e.
Consequen ly, g oups a e unbalanced and he s udy is he e o e a
quasi–expe imen , due o he absence o andom assignmen . Al-
hough his assignmen mechanism has se e al disad an ages [11,
14], he expe imen e s conside ed indi iduals’ mo i a ion essen ial
o a oid mo ali y and ensu e ha he ea men is ac ually applied,
some hing c ucial o he alidi y o he expe imen .
3.3 Expe imen al ma e ial
The ma e ial used in he conduc ion o he expe imen (a ailable
a h p://www.lsi.us.es/~bea /ExpMa e /mind ulness)
comp ises o he ollowing i ems:
•A ques ionnai e wi h h ee sec ions: (i) ques ions ega ding
he impo ance o quali y in so wa e, and he ela ionship
be ween he quali y o he so wa e p oduc and ha o he
p ocess leading o i s c ea ion; (ii) ques ions abou he in e -
es on pa icipa ing in he esea ch; and (iii) ques ions abou
he choice o mind ulness o public speaking.
•Two exe cises o concep ual modelling. The s a emen s o
he exe cises we e p esen ed as an in e iew ansc ip , i.e.
ques ions and answe s (Q&A) be ween a equi emen s engi-
nee and a cus ome abou he p oblem domain and he ex-
pec ed sys em beha iou .
The i s exe cise is abou an hypo he ical in o ma ion sys em
o ERASMUS g an s managemen , whe eas he second one is
abou an hypo he ical in o ma ion sys em o End o Deg ee
P ojec s (EODP) managemen . Bo h p oblem domains we e
chosen because o s uden s’ amilia i y and in e es , since all
o hem a e po en ial candida es o an ERASMUS g an and
ha e o de elop an EODP.
3.4 Tasks
Du ing he expe imen , he subjec s pe o med he ollowing asks
(see Table 2): (i) illing ou he in e es ques ionnai e a e a b ie
in oduc ion o he ongoing esea ch and mind ulness; (ii) doing he
i s concep ual modelling exe cise (ERASMUS); (iii) a ending he
sessions on he c oss– aining opics o hei in e es , along wi h
he ISEIS lessons, du ing ou weeks, ou days a week; (i ) doing
he second concep ual modelling exe cise (EODP).
3.5 Va iables and pa ame e s
The independen a iables o ac o s ha could ha e had an im-
pac on he esul s a e:
In oduc o y p esen a ion & Ques ionnai e o in e es (30min)
++
Mind ulness g oup Public Speaking g oup
Semina abou mind ulness
(1h)
Semina abou public
speaking (1h)
ERASMUS concep ual modelling exe cise (2h)
4 imes/week ⇥
4 weeks ⇥10
minu e mind ul-
ness sessions
ISEIS scheduled
lessons
4 imes/week ⇥
4 weeks ⇥10
minu e public
speaking sessions
EODP concep ual modelling exe cise (1.5h)
Table 2: Expe imen schedule
C oss–T aining ( aining in complemen a y sessions): he main
independen a iable o he expe imen is he speci ic opic
augh du ing a se o complimen a y sessions pe o med du -
ing ou weeks. This ac o has wo le els, he ea men is
mind ulness and he con ol condi ion is public speaking, i.e.
he sessions on public speaking a e used as an placebo, hus
he expe imen e s implici ly assume hey ha e no impac on
he esponse a iables. Acco ding o [13], his a iable is
conside ed as a esou ce, since i enhances he necessa y
skills o he pe o mance o pe sonal and p o essional oles
o he s uden s and i is in a nominal scale.
P e/Pos –Exe cise (exe cises done be o e and a e c oss– aining
sessions): his a iable ep esen s he concep ual modelling
exe cises ha s uden s ha e done. I has wo le els: ERAS-
MUS and EODP.
Since s uden s a e also lea ning SE du ing hese 4 weeks, an
imp o emen in bo h g oups should be obse ed. Howe e ,
we in end o analyse whe he he imp o emen o s uden s in
he mind ulness g oup is signi ican ly di e en om hose in
public speaking o no . Acco ding o [13], his is a ma e ial
a iable and i is measu ed in nominal scale.
In o de o de ine dependen a iables, an indi ec measu e called
seman ic exp essi eness (SEMEX) was used. The seman ic exp es-
si eness measu es o wha ex en —compa ed o a canonical con-
cep ual model de eloped by he expe imen e s— he concep ual mod-
els de eloped by expe imen al subjec s a e able o ep esen all he
p oblem domain scena ios desc ibed in he in e iew ansc ip s.
Model elemen s a e conside ed as co ec ly iden i ied i he e ex-
is iden ical o seman ically equi alen elemen s in he canonical
model. The o mula o compu ing his in e nal a iable, measu ed
in a a io scale, is he ollowing:
SEMEX =CLASSOK CLASSKO
2+ASSOCOK +ATTROK
whe e CLASSOK is he numbe o classes co ec ly iden i ied; CLASSKO
is he numbe o classes inco ec ly iden i ied; and ASSOCOK and
ATTROK he numbe o co ec ly iden i ied associa ions and a ibu es
espec i ely.
Taking SEMEX in o accoun , he ollowing dependen a iables
we e conside ed:
E ec i eness: he pe cen age o seman ic exp essi eness achie ed
by a subjec measu ed in a io scale:
EFFECTIVENESS =SEMEX
CLASSC+ASSOCC+ATTRC
whe e CLASSC,ASSOCC,ATTRCa e he numbe o classes,
associa ions, and a ibu es in he canonical solu ion p e i-
ously ag eed by he expe imen e s, i.e. hei sum is he o al
seman ic exp essi eness.
E iciency: his a iable, ollowing he ecommenda ions in [16],
is de ined as he a io be ween he achie ed seman ic exp es-
si eness and he ime spen by a subjec in doing a concep ual
modelling exe cise:
EFFICIENCY =SEMEX
TIME
whe e TIME is he ime spen in doing he concep ual mod-
elling exe cise measu ed in minu es.
Following [14], some pa ame e s we e iden i ied: he lack o ex-
pe ience o he subjec s in eal p ojec s, he simila size o he con-
cep ual modelling exe cises, he p oblem domain o he concep ual
modelling exe cises ela ed o academic opics, and he numbe o
sessions on each c oss– aining opic.
3.6 Hypo heses
Two g oups o hypo heses, one o each dependen a iable, we e
es ed in he expe imen .
E ec i eness hypo heses
H0,1: he e is no di e ence in he e ec i eness o subjec s in he
ERASMUS and he EODP exe cises, i.e. nei he he eaching
o ISEIS no he c oss– aining sessions ha e a signi ican e -
ec on he e ec i eness o subjec s. // H1,1:¬H0,1
H0,2: he e is no di e ence in e ec i eness be ween he subjec s
p ac ising mind ulness and hose subjec s a ending public
speaking sessions— he expe imen e s assumed ha he di -
e ences obse ed be ween he e ec i eness o he ERAS-
MUS and he EODP exe cises a e due o he eaching o ISEIS
exclusi ely). // H1,2:¬H0,2
E iciency hypo heses
H0,3: he e is no di e ence in he e iciency o subjec s in he
ERASMUS and he EODP exe cises, i.e. nei he he eaching
o ISEIS no he c oss– aining sessions ha e a signi ican e -
ec on he e iciency o subjec s. // H1,3:¬H0,3
H0,4: he e is no di e ence in e iciency be ween he subjec s
p ac ising mind ulness and hose subjec s a ending public
speaking sessions— he expe imen e s assumed ha he di -
e ences obse ed be ween he e iciency o he ERASMUS
and he EODP exe cises a e due o he eaching o ISEIS ex-
clusi ely). // H1,4:¬H0,4
3.7 Design
In SE, expe imen a ion is usually aimed a es ing he goodness
o me hods, ools, o echnologies, so he usual expe imen al de-
signs a e simila o hose p esen ed in [14]. When epea ed mea-
su es a e aken, he eason is usually ei he imp o ing he p ecision
by making compa isons wi hin ma ched pai s o he expe imen
ma e ial [38], o maximizing da a collec ion [2].
Ne e heless, o he expe imen p esen ed in his pape , which
includes pe sonal de elopmen aspec s in o empi ical SE, i seems
easonable o use a design ha allows he measu e o he depen-
den a iables be o e and a e he p ac ice o mind ulness. Con-
side ing he hypo heses, he expe imen al design should ake in o
accoun wo ac o s wi h wo le els: he c oss– aining ecei ed
C oss–T aining
Mind ulness Public
Speaking
P e/Pos P e–Exe cise (ERASMUS) G1G2
Exe cise Pos –Exe cise (EODP) G1G2
Table 3: Expe imen al design — 2⇥2 mixed ac o ial
(a be ween–subjec s ac o ) and he concep ual modelling exe cises
(a wi hin–subjec s ac o ). Since dedica ing wo ISEIS classes o
he p e and pos concep ual modelling exe cises was a o dable,
a quasi–expe imen wi h he 2⇥2 mixed ac o ial design om [5]
was selec ed (see Table 3). In his design, commonly used in he
ields o psychology and medicine when he e olu ion o pa ien s
unde a gi en ea men (o he apy) a e a ce ain amoun o ime
needs o be s udied [19, 21, 25], each subjec is assigned o one
single ea men (C oss–T aining le el) and wo epea ed measu es
on he esponse a iables a e aken.
4. EXECUTION
Fi s o all, he ongoing esea ch and a b ie in oduc ion o mind-
ulness we e p esen ed o 87 s uden s du ing hei o dina y class
schedule. A e ha —du ing he same class— he in e es ques-
ionnai e desc ibed in Sec ion 3.3 was gi en o he s uden s, who
illed i ou manually. Once he answe s we e analysed by he ex-
pe imen e s, he s uden s we e di ided in o wo g oups acco ding
o hei p e e ences, excep 12 s uden s who showed no in e es in
pa icipa ing and we e he e o e kep aside om he expe imen .
P io o he day o he expe imen execu ion, each g oup was
gi en an in oduc o y semina abou he co esponding opic, i.e.
mind ulness o G1and public speaking o G2. The semina s ook
abou one hou and we e deli e ed ou o o dina y class schedule.
S uden s’ a endance was con olled by a sign–in shee .
Nex , he wo g oups o s uden s did he p e–exe cise (ERASMUS)
he same day du ing class schedule. The aim o he exe cise was
eading an in e iew ansc ip ca e ully and de eloping a concep-
ual model iden i ying classes, associa ions and a ibu es.
Then, du ing ou weeks, each g oup o s uden s a ended com-
plemen a y sessions o he co esponding opic ou days a week.
The sessions ook place always in he same condi ions, i.e. same
class oom and same hou o each g oup.
Ul ima ely, he wo g oups o s uden s did he pos –exe cise (EODP)
he same day du ing class schedule. Each g oup did he exe cise in
di e en class ooms, always moni o ed by one expe imen e . In he
G1g oup, he ini ial 15 minu es we e dedica ed o a guided o mal
medi a ion in g oup, simila o hose pe o med du ing he p e ious
ou weeks. A e he 15–minu e medi a ion, he s uden s did he
exe cise indi idually. Du ing he exe cise, he wo expe imen e s
moni o ing he exe cise answe ed ques ions om he s uden s ha
we e ela ed o he no a ion only (UML), ying o a oid any dis-
cussion abou he in e p e a ion o he in e iew ansc ip in o de
o a oid any in luence on he ou come a iables.
Once he s uden s’ pa icipa ion was inished, bo h exe cises we e
blindly checked and he ob ained da a was inpu in o Excel and hen
impo ed in o SPSS 22. Fo each subjec , he ollowing da a we e
collec ed: answe s o he in e es ques ionnai e, numbe o sessions
in he c oss– aining, and seman ic exp essi eness, e ec i eness,
ime, and e iciency o bo h exe cises.
Table 4 shows he numbe o s uden s who pa icipa ed in each
phase o he expe imen . The in e es column shows he numbe o
subjec s who showed in e es in mind ulness and public speaking
espec i ely. The a endance column shows he a e age numbe
In e es P e A end.
(mean)
Pos Sample
Numbe G138 18 23 17 16
o subjec s G237 24 10 18 16
Table 4: Numbe o subjec s du ing execu ion
o subjec s who a ended he c oss– aining sessions du ing ou
weeks. The p e and pos columns show he numbe o subjec s
who did he p e and pos –exe cises. The sample column indica es
he inal numbe o subjec s who whe e inally conside ed du ing
analysis and e alua ion o he expe imen . The di e ence be ween
he sample size and he numbe o pa icipa ing subjec s is mo i-
a ed by he ollowing de ia ions:
•In he G1g oup, 5 subjec s a ended he mind ulness ses-
sions wi hou ha ing done he p e–exe cise. Only 1 ou o
he 17 subjec s who did he pos –exe cise did no do he p e–
exe cise. As a esul , only he 16 subjec s who did bo h exe -
cises we e included in he sample.
•Applying he same c i e ia in he G2g oup, only 16 subjec s
did bo h exe cises and we e included in he sample.
5. ANALYSIS AND EVALUATION
A summa y o he desc ip i e s a is ics o he expe imen in dis-
played in Table 5, showing he mean and s anda d de ia ion o he
e ec i eness and e iciency o he p e and pos exe cises. Figu es
1 and 2 depic he dispe sion o he same dependen a iables.
Mean (µ)S d. De . ()
P e–Ex. Pos –Ex. P e–Ex. Pos –Ex.
E ec i e– G10.395 0.673 0.183 0.131
ness G20.416 0.611 0.084 0.092
E iciency G10.172 0.502 0.075 0.177
G20.188 0.365 0.042 0.108
Table 5: Desc ip i e s a is ics o he sample wi h 32 subjec s
Wi h espec o he po en ial ou lie s depic ed in Figu es 1 and
2, he low e ec i eness o subjec 8 was sc u inized inding ha ,
al hough some ex ual no es we e de eloped by he subjec , hey
could no be conside ed as a alid concep ual model and he e o e
no alid concep ual elemen could be iden i ied, i.e. SEMEX =0in
he p e–exe cise, so he subjec was le ou o he da a analysis.
The e y high e iciency o subjec s 10 and 11 was also in es i-
ga ed, bu no o he cause han a high pe o mance was ound. As a
consequence, hey we e conside ed as genuine alues since hey
showed an ou s anding pe o mance be o e and a e he c oss–
aining sessions. Rega ding subjec 31, al hough he showed a high
e iciency only in he p e–exe cise, no eason was ound o dis-
ca d him. Howe e , he s a is ical es s o he hypo hesis shown in
Sec ions 5.1 and 5.2 ha e been pe o med on bo h samples, i.e. in-
cluding and excluding subjec s 10 and 11, and he esul s ega ding
hypo heses accep ance o ejec ion we e no a ec ed when using
↵=0.01.
Then, due o he lack o andom assignmen o g oups, he di e -
ences be ween G1and G2we e examined be o e s a ing he c oss–
aining sessions. In o de o check he hypo hesis o simila i y o
g oups in e ms o he alues o he dependen a iables [5], one–
way analysis o a iance (ANOVA) was conduc ed on each o he

Figu e 1: Box plo o e ec i eness Figu e 2: Box plo o e iciency
ou come measu es. Since he null hypo heses we e no ejec ed
(e ec i eness(p– alue) = 0,896 and e iciency(p– alue) = 0,816),
he e is no e idence o signi ican di e ences be ween g oups, which
mi iga es o some ex en he p oblems and isks caused by he lack
o andom assignmen [5].
5.1 Tes ing e ec i eness hypo heses
The pu pose o hese hypo heses is o de e mine how he inde-
penden a iables ha e in luenced on e ec i eness, i.e.:
•Whe he he ISEIS lessons ha e in luenced on e ec i eness.
•Whe he he in e ac ion be ween he c oss– aining and he
ISEIS lessons has in luenced on e ec i eness.
The means o e ec i eness ob ained o each le el o he inde-
penden a iables a e shown in igu e 3. Fo bo h c oss– aining
le els, he mean o e ec i eness a e ea men is highe han be-
o e. This is no su p ising, since s uden s’ knowledge o SE has
been imp o ed due o he ISEIS lessons, ha ing an impac on e -
ec i eness. I is also no iceable ha he line showing he imp o e-
men o he mind ulness g oup has a s eepe slope han he line o
he public speaking g oup.
To es he e ec i eness hypo heses, i was i s e alua ed whe he
he da a ollowed a no mal dis ibu ion o no (see Shapi o–Wilk
esul s in able 6).
Gi en ha all he g oups we e no mal a he le el o 0.05, a
mixed–model ANOVA was conduc ed, wi h C oss–T aining as a
be ween–subjec s ac o and P e/Pos –Exe cise as a wi hin–subjec s
ac o . In able 7 he esul s ob ained by means o ANOVA o e ec-
i eness a e shown, in which he las column ep esen s he pa ial
Dependen G oup Shapi o–Wilk
a iable signi icance le el
E ec i eness G1(p e) 0.830
G2(p e) 0.709
G1(pos ) 0.377
G2(pos ) 0.405
Table 6: Shapi o–Wilk no mali y es esul s o e ec i eness
e a–squa ed (⌘2
p) as an indica o o he e ec size o g oups. The
i s ow o he able shows he da a o he P e/Pos –Exe cise in-
dependen a iable o he expe imen ; he second ow shows he
da a o he in e ac ion be ween he a o emen ioned a iable and
C oss–T aining; inally, he e o is shown in he las ow. The anal-
ysis summa ised in able 7 e ealed a signi ican e ec (↵=0.01)
o P e/Pos –Exe cise, i.e. bo h g oups ha e a s a is ically be e
e ec i eness a e aining han be o e aining, measu ed by he
pe cen age o seman ic exp essi eness achie ed. Howe e , he in-
e ac ion be ween P e/Pos –Exe cise and C oss–T aining is no sig-
ni ican a he 0.01 le el. This means ha , al hough he mean o
e ec i eness a ied signi ican ly o bo h exe cises, he e ec o
he C oss–T aining is no necessa ily linked o hese di e ences.
The e o e, he null hypo hesis H01 is ejec ed a he ↵=0.01 sig-
ni icance le el, bu he e is no enough e idence as o ejec H02 a
such signi icance le el. Howe e , his same analysis o e ec i e-
Figu e 3: P o ile plo o mean o e ec i eness
Sou ce o a ia ion Type III Sum
o Squa es
Deg ees o
F eedom
Mean Squa e F– a io Signi icance ⌘2
p
P e/Pos –Exe cise 0.860 1 0.860 90.253 0.000 0.757
P e/Pos –Exe cise * C oss– aining 0.026 1 0.026 2.713 0.110 0.086
E o (P e/Pos –Exe cise) 0.276 29 0.010
Table 7: ANOVA o e ec i eness
ness was epea ed emo ing ou lie s 10 and 11. The p– alue o he
in e ac ion be ween independen a iables in his case was 0.048,
i.e. he e ec i eness o s uden s a ending ISEIS and p ac ising
mind ulness is be e ha hose ha jus a ended ISEIS and public
speaking sessions. Acco ding o his esul , H02 could be ejec ed
a he ↵=0.05 signi icance le el. We do no claim he exis ence
o a signi ican di e ence on he e ec i eness due o mind ulness,
since he p– alue is almos in he limi o s a is ical signi icance and
he emo al o ou lie s 10 and 11 could be objec ed. Howe e , he
esul s aise he suspicion abou he exis ence o such a di e ence
in p ac ice. Fu he mo e, since he numbe o mind ulness sessions
could be oo low as o show a signi ican e ec a ↵=0.01, we
conside ha i is wo hy o eplica e he expe imen in o de o dis-
ce n i a cause–and–e ec ela ionship ac ually exis s be ween he
p ac ice o mind ulness and he imp o emen o e ec i eness.
5.2 Tes ing e iciency hypo heses
The pu pose o hese hypo heses is o de e mine how he inde-
penden a iables ha e in luenced on e iciency, i.e.:
•Whe he he ISEIS lessons ha e in luenced on e iciency.
•Whe he he in e ac ion be ween he c oss– aining and he
ISEIS lessons ha e in luenced on e iciency.
The means o e iciency ob ained o each le el o he indepen-
den a iables a e shown in igu e 4. As o e ec i eness, he mean
o e iciency a e ea men is highe han be o e o bo h c oss–
aining le els. In o de o es he e iciency hypo heses, i was
i s ly e alua ed whe he he da a ollowed a no mal dis ibu ion o
no (see Shapi o–Wilk esul s in able 8).
Figu e 4: P o ile plo o mean o e iciency
Dependen G oup Shapi o–Wilk
a iable signi icance le el
E iciency G1(p e) 0.275
G2(p e) 0.560
G1(pos ) 0.039
G2(pos ) 0.243
Table 8: Shapi o–Wilk no mali y es esul s o e iciency
Al hough he da a o G1(pos ) was no no mal a he le el o
0.05, i was decided o ca y ou a mixed–model ANOVA in a sim-
ila way as wi h he e ec i eness hypo heses es s, since mos o
he da a was no mal and he s a is ical es is obus [10]. The e-
sul s ob ained by means o ANOVA o e ec i eness a e shown in
Table 9. This analysis e ealed a signi ican e ec (↵=0.01)
o P e/Pos –Exe cise, i.e. bo h g oups had a s a is ically be e
e iciency a e aining han be o e aining, as measu ed by he
seman ic exp essi eness achie ed pe uni o ime. A his signi i-
cance le el (↵=0.01), he in e ac ion be ween P e/Pos –Exe cise
and C oss–T aining is also signi ican . This means ha , al hough
he mean e iciency a ied signi ican ly o bo h exe cises in bo h
g oups, he p ac ice o mind ulness is linked o a highe imp o e-
men in he e iciency. The e o e, he null hypo heses H03 and H04
a e ejec ed a he ↵=0.01 signi icance le el. In a e y simi-
la way o e ec i eness, we epea ed he analysis o he e iciency
wi hou aking in o accoun ou lie s 10 and 11. In his case, he e-
sul s we e essen ially he same, i.e. he p– alue o he in e ac ion
o independen a iables was 0.001, hus H04 was ejec ed a he
↵=0.01 signi icance le el.
6. DISCUSSION
The main inding o he expe imen desc ibed in his pape is ha
he p ac ice o mind ulness make SE s uden s capable o achie ing
simila esul s in less ime— hey become mo e e icien —in he
concep ual modelling o a p oblem domain desc ibed by an in e -
iew ansc ip . Compa ed o his, he obse ed mind ulness e ec
on e ec i eness has been ela i ely small and no enough e idences
exis o ejec he null hypo hesis. This could be because (i) ei he
he e is no casual ela ionship be ween mind ulness and s uden s’
e ec i eness in concep ual modelling; o (ii) because he numbe
o mind ulness sessions was insu icien o p oduce he expec ed
e ec . We hink he la e is mo e plausible since mos epo s in-
dica e ha is a e wo mon hs o mind ulness p ac ice as a he apy
when hei e ec s begin o be pe cei ed. Ne e heless, he main
expe imen ha ha e in luenced ou wo k [19], used only 15 ses-
sions o mind ulness as a way o enhancing some cogni i e skills
bu no as a he apy. Fo his eason, we hink ha is necessa y o
adjus he numbe o mind ulness sessions in u u e eplica ions o
he expe imen . Rega ding he ype o ask chosen o he expe i-
men , we hink ha concep ual modelling was a be e choice han
eading comp ehension [30]. The eason is ha mind ulness is sup-
posed o inc ease abs ac ion as a esul o a be e men al cla i y,
and his is an essen ial skill o concep ual modelling.
Sou ce o a ia ion Type III Sum
o Squa es
Deg ees o
F eedom
Mean Squa e F– a io Signi icance ⌘2
p
P e/Pos –Exe cise 1.047 1 1.047 247.515 0.000 0.895
P e/Pos –Exe cise * C oss–T aining 0.072 1 0.072 17.001 0.000 0.370
E o (P e/Pos –Exe cise) 0.123 29 0.004
Table 9: ANOVA o e iciency
Wi h espec o he possibili ies o eplica ing his expe imen
in o he uni e si ies, he main p oblem is inding a pe son com-
mi ed o guide he mind ulness sessions. O he au ho s like [25]
use CDs wi h eco ded medi a ion guides o indi idual use, bu
we hink ha young s uden s a e no likely o keep on p ac ising
mind ulness indi idually wi hou he eedback o a g oup. O he
possibili ies, like p ac ising mind ulness in g oup bu dis ibu i ely
ia ideocon e encing, ha e no been explo ed ye .
6.1 Th ea s o alidi y
In [38], a ho ough compila ion o h ea s o he alidi y o he
esul s o empi ical s udies is p esen ed. In his sec ion, he h ea s
ela ed o he conclusion,in e nal,cons uc , and ex e nal alidi-
ies a e analysed and he ac ions pe o med o educe hem a e de-
sc ibed.
Conclusion alidi y: his alidi y is conce ned wi h he s a is ical
ela ionship be ween he ea men and he ou come. In he
expe imen desc ibed in his pape , he main h ea o conclu-
sion alidi y was he small size o he sample. Ne e heless,
al hough small, he sample size was accep able o he s a is-
ical es s applied [14], and all he assump ions o each es
we e e i ied be o e hei applica ion. In addi ion, hose sub-
jec s who missed one o he exe cises we e excluded om
he sample, in o de o apply he es s p ope ly.
In e nal alidi y: his alidi y is conce ned wi h he ea men —
and no o he uncon olled o unmeasu ed ac o —being he
cause o he ou come. The ollowing h ea s desc ibed in [38]
we e iden i ied:
His o y: since bo h g oups pe o med he expe imen asks
simul aneously, wi hou any signi ican inciden , his
h ea was neu alised.
Ma u a ion: bo h g oups ma u a ed simul aneously wi h e-
spec o hei knowledge in SE, since hey all a ended
he same numbe o ISEIS sessions be ween he p e and
pos exe cises. The e o e, his h ea was also neu alised.
Mo ali y: in o de o educe his h ea , he s uden s we e
o e ed ex a poin s o pa icipa ing in he expe imen .
Fu he mo e, no subjec s abandoned he mind ulness
sessions due o edium o lack o in e es .
Selec ion & assignmen bias: he main h ea s o he in e -
nal alidi y o he expe imen we e selec ion and as-
signmen bias, since i was a quasi–expe imen . How-
e e , he e a e no e idences o signi ican di e ences
in he g oups p io o ea men , as e i ied by means
o a one–way ANOVA es on he measu es o he p e–
exe cise dependen a iables (see Sec ion 5 o de ails).
On he o he hand, mo i a ion was also simila in bo h
g oups because he subjec s pa icipa ed in he expe i-
men on hei own choice, hey eely chose he c oss–
aining opic, and he p e and pos exe cises we e ele-
an o hei ma ks in ISEIS.
Howe e , gi en he shown implica ion o s uden s in
his expe imen , we plan o pe o m andom assignmen
in u u e eplica ions.
Cons uc alidi y: his alidi y is conce ned wi h he ela ion be-
ween heo y and obse a ion. Since he expe imen includes
wo c oss– aining opics as ea men s, and wo exe cises,
he mono–ope a ion bias was educed because o he cause
cons uc being comple ely ep esen ed. The mono–me hod
bias was also educed by he analysis o he e ec i eness and
e iciency dependen a iables.
Ex e nal alidi y: he g ea e he ex e nal alidi y, he mo e he
esul s o an empi ical s udy can be gene alized o cu en
SE p ac ice. Two h ea s we e iden i ied which can limi such
gene aliza ion:
Ma e ials used: he size o he in e iew ansc ip s migh
no be ep esen a i e o indus ial p oblems, bu i was
app op ia e o he a ailable ime o he p e and pos
exe cises. Howe e , we hink ha he in ellec ual p o-
cesses applied du ing concep ual modelling—and po-
en ially imp o ed by mind ulness—a e basically he
same ega dless o he size o he p oblem a hand.
Subjec s: since he asks pe o med du ing he p e and pos
exe cises did no equi e high le els o indus ial expe-
ience, using s uden s as subjec s ins ead o SE p o es-
sionals could be conside ed as app op ia e [20]. Mo e-
o e , s uden s a e he nex gene a ion o p o essionals,
so hey a e close o he popula ion unde s udy [16].
7. CONCLUSIONS AND FUTURE WORK
In his pape , he esul s o an expe imen o analyse he e ec s
o mind ulness on he de elopmen o concep ual models a e p e-
sen ed. P e iously, he main psycho–social ac o s ha in luence
he SE p ocess we e e iewed, showing he connec ions wi h he
alleged bene i s o mind ulness.
In ou opinion, concep ual modelling is a di icul ask o ou
s uden s. We hink ha he p o en bene i s o mind ulness, i.e.
men al cla i y, eading comp ehension, concen a ion, e c., could
be a ele an suppo o his ask. The expe imen esul s e ealed
ha s uden s who p ac ised mind ulness ob ained simila esul s in
less ime han he o he s, i.e. a signi ican imp o emen in e i-
ciency was obse ed. Howe e , he obse ed imp o emen in e -
ec i eness, al hough impo an , was no s a is ically signi ican .
I he esul s o his expe imen we e con i med by eplica ion,
he imp o emen in e iciency would be qui e in e es ing o SE
o ganiza ions. A key ac o o he adop ion o mind ulness in in-
dus y is pe suasion, i.e. o con ince employees o he goodness o
i s p ac ice. Once con inced, i is necessa y o acili a e con inui y,
an app op ia e a mosphe e in he o ganisa ion, so wa e guides o
medi a ion (a mobile app o example), p o iding ollow–up mee -
ings, e c.
A u u e wo k in he sho e m is eplica ing he expe imen
inc easing he numbe o c oss– aining sessions and using an-
dom assignmen . Ano he op ion is conduc ing a dis ibu ed quasi–
expe imen , in which an expe imen is un ac oss mul iple si es,
o ganised as a single s udy a he han as a se o eplica ions [3].
Then, changing he expe imen al objec , i would be possible o
s udy he e ec s o mind ulness on he quali y and pe o mance o
he subjec s du ing o he SE ac i i ies, such as elici a ion and ne-
go ia ion o equi emen s, o esolu ion o echnological p oblems.
Fu he mo e, we would like o examine whe he mind ulness im-
p o es eam–wo k among in o e ed so wa e enginee s.
Once he s eng hs o he p ac ice o mind ulness ha e been iden-
i ied, we would like o pe o m case s udies in eal companies in
o de o de elop a mind ulness–based pe sonal g ow h p og am o
i s u he adop ion.
8. ACKNOWLEDGMENTS
The au ho s would like o hank he ISEIS s uden s who pa ici-
pa ed in he expe imen o hei excellen a i ude. We would also
like o hank D . M. Gene o, om he Uni e si y o Cas illa–La
Mancha, o being always willing o answe ou expe imen al un-
ce ain ies. We also hank he s a o he Escuela Técnica Supe io
de Ingenie ía In o má ica o he Uni e si y o Se ille o a ang-
ing any hing we needed in he class ooms o he mind ulness and
public speaking sessions. Finally, we hank E. Pé ez In an es and
J. Mu ga o guiding some o he mind ulness sessions.
This wo k was pa ially suppo ed by he EU Commission (FEDER),
he Spanish and Andalusian R&D&I p og amme g an s TAPAS
(TIN2012–32273), COPAS (P12–TIC–1867), and THEOS (TIC–
5906).
9. REFERENCES
[1] S. T. Acuña, M. Gómez, and N. Ju is o. How do pe sonali y,
eam p ocesses and ask cha ac e is ics ela e o job
sa is ac ion and so wa e quali y? In o ma ion and So wa e
Technology, 51(3):627–639, 2009.
[2] L. B iand, Y. Labiche, and R. Mad azo-Ri e a. An
expe imen al e alua ion o he impac o sys em sequence
diag ams and sys em ope a ion con ac s on he quali y o he
domain model. In Empi ical So wa e Enginee ing and
Measu emen (ESEM), 2011 In e na ional Symposium on,
pages 157–166. IEEE, 2011.
[3] D. Budgen, B. Ki chenham, S. Cha e s, S. Gibbs,
A. Poh hong, J. Keung, and P. B e e on. Lessons om
conduc ing a dis ibu ed quasi-expe imen . In Empi ical
So wa e Enginee ing and Measu emen , 2013 ACM/IEEE
In e na ional Symposium on, pages 143–152. IEEE, 2013.
[4] B. S. Bu le and P. H. G ay. Reliabili y, mind ulness, and
in o ma ion sys ems. MIS Qua e ly, 30(2):211–224, 2006.
[5] D. T. Campbell and S. Julian. Expe imen al and
Quasi–Expe imen al Designs o Resea ch. Wadswo h,
Uni ed S a es, 1963.
[6] L. F. Cap e z. Pe sonali y ypes in so wa e enginee ing.
In e na ional Jou nal o Human–Compu e S udies,
58(2):207–214, 2003.
[7] A. Da is. 201 p inciples o so wa e de elopmen .
McG aw–Hil, New Yo k, 1995.
[8] D. M. Da is and J. A. Hayes. Wha a e he bene i s o
mind ulness? a p ac ice e iew o psycho he apy- ela ed
esea ch. Psycho he apy, 48(2):198, 2011.
[9] C. K. Ge me , R. D. Siegel, and P. R. Fun on. Mind ulness
and Psycho he apy. The Guil o d P ess, New Yo k, 2013.
[10] G. Glass, P. D. Peckham, and J. R. Sande s. Consequences o
ailu e o mee assump ions unde lying he ixed e ec s
analyses o a iance and co a iance. Re iew o Educa ional
Resea ch, 42(1):237–288, 1972.
[11] J. A. Gline , G. A. Mo gan, and N. L. Leech. Resea ch
me hods in applied se ings: an in eg a ed app oach o
design and analysis (second edi ion). Tylo and F ancis
G oup, 2009.
[12] P. G ossman, L. Niemann, S. Schmid , and H. Walach.
Mind ulness–based s ess educ ion and heal h bene i s: A
me a–analysis. Jou nal o Psychosoma ic Resea ch,
57(1):35–43, 2004.
[13] A. Jedli schka, M. Ciolkowski, and P. D. Guide o Ad anced
Empi ical So wa e Enginee ing, chap e Repo ing
expe imen s in So wa e Enginee ing, pages 201–228.
Sp inge Ve lag, 2008. Eds. Fo es Shull and Janice Singe
and Dag I.K. Sjøbe g.
[14] N. Ju is o and A. M. Mo eno. Basics o So wa e
Enginee ing Expe imen a ion. Kluwe Academic Publishe s,
2001.
[15] T. Kilamo, I. Hammouda, and M. Cha i. Teaching
collabo a i e so wa e de elopmen : A case s udy. In 34 h
In e na ional Con e ence on So wa e Enginee ing (ICSE)
P oceedings, pages 1165–1174. ACM, June 2012.
[16] B. A. Ki chenham, S. L. P leege , D. Hoaglin., K. E. Emam,
and J. Rosenbe g. P elimina y Guidelines o Empi ical
Resea ch in So wa e Enginee ing. IEEE T ansac ions on
So wa e Enginee ing, 28(8):721–734, Augus 2002.
[17] E. J. Lange . Mind ulness. Addison-Wesley, Massachuse s,
1989.
[18] S. Ma ook and K. Kau z. Mind ulness and agile so wa e
de elopmen . In 19 h Aus alasian Con e ence on
In o ma ion Sys ems (ACIS) P oceedings, pages 638–647.
Associa ion o In o ma ion Sys ems/Uni e si y o
Can e bu y, Decembe 2008.
[19] M. D. M azek, M. S. F anklin, D. T. Phillips, B. Bai d, and
J. W. Schoole . Mind ulness aining imp o es wo king
memo y capaci y and g e pe o mance while educing mind
wande ing. Psychological Science, 24(5):776–781, May
2013.
[20] A. A. Po e , L. G. Vo a, and V. R. Basili. Building
Knowledge h ough Families o Expe imen s. IEEE
T ansac ions on So wa e Enginee ing, 25(4):456–473, July
1999.
[21] P. A. Poulin, C. S. Mackenzie, G. Soloway, and
E. Ka ayolas. Mind ulness aining as an e idenced-based
app oach o educing s ess and p omo ing well-being among
human se ices p o essionals. In e na ional Jou nal o
Heal h P omo ion and Educa ion, 46(2):72–80, 2008.
[22] A. Puddicombe. Ge some headspace. Hodde and
S ough on, USA, 2011.
[23] D. Riebel, J. G eeson, G. B aina d, and S. Rosenzweig.
Mind ulness–based s ess educ ion and heal h– ela ed
quali y o li e in a he e ogeneous pa ien popula ion.
Je e son My na B ind Cen e o In eg a ion Medicine
Facul y, 2(1):1–20, July 2001.
[24] D. Sammon, T. Nagle, and J. McA oy. Analysing ISD
pe o mance using na a i e ne wo ks, ou ines and
mind ulness. In o ma ion and So wa e Technology, 2014.
[25] M. B. Schu e, J. Ch is ophe , and S. Ch is ophe . Mind–body
medicine and he a o sel -ca e: Teaching mind ulness o
counseling s uden s h ough yoga, medi a ion, and qigong.