Modelling the Relationships Between the ... - Business & Law€¦ · Characteristics Survey (BCS) are used in the analysis. The study applies probit regression modelling on the 2010-11

345AUSTRALIAN JOURNAL OF LABOUR ECONOMICS

Volume 18 • Number 3 • 2015 • pp 345 - 374

Modelling the Relationships Between the Use of STEM* Skills, Collaboration, R&D, and Innovation among Australian Businesses

Franklin Soriano and Ruel Abello,AustralianBureauofStatistics

Abstract This paper investigates the relationship between the use of STEM/Non-STEM skills, collaboration, R&D and innovation, including novelty of innovation, among Australian businesses. The analysis employs standard probit modelling using the 2010-11 and 2011-12 ABS Business Characteristics Survey data. Results show that the use of STEM skills and collaboration in R&D are very strongly associated with an increase in the likelihood of innovating. The combined impact of collaboration in R&D, expenditure in R&D, and the use of STEM skills on the probability of having a ‘new to the world’ type of innovation is also found to be strong.

Keywords:Skills,Collaboration,R&D,Innovation,Australia.

JELClassification:O32,O31,J24,D29

1. Introduction ThereisgrowingrecognitionoftheimportanceofhumancapitalinshapingAustralia’sfutureprosperity.ArecentreportfortheAustralianCouncilofLearnedAcademiesargued that building capacity particularly in the fields of science, technology,engineering and mathematics (STEM) is pivotal to competitiveness in the globaleconomy (Marginson, et al. 2013). Another study showed that an increase in theproportion of workers in skilled occupations is followed by an increase in labour

Address for correspondence:FranklinSoriano,AnalyticalServicesUnit,AustralianBureauofStatistics,ACT,2617Australia.Email:[email protected]: The authors would like to thank the following: Dr Roslyn Prinsley, DrKrisztian Baranyai (Office of the Chief Scientist (OCS), Department of Industry and Science(DIS)), JasonWong,KerryO’Brien,SybilleMcKeown,DrSiu-MingTam (ABS) andDrGregConnolly(DepartmentofEmployment)fortheirvaluablesuggestionsandcomments.Theviewsexpressedinthispaperarethoseoftheauthors,anddonotnecessarilyrepresentthoseoftheABSortheDIS.Wherequotedorused,theyshouldbeattributedclearlytotheauthors.Anyerrorsaretheauthors.*Science,Technology,EngineeringandMathematics.ThispaperisbasedonastudyundertakenfortheOCS,DIS.Anearlierversionofthispaperwaspresentedatthe25thAustralianLabourMarketResearchWorkshop(ALMRW)inNovember2014.©TheCentreforLabourMarketResearch,2015

346AUSTRALIAN JOURNAL OF LABOUR ECONOMICSVOLUME 18 • NUMBER 3 • 2015

productivity,andorganisationofhumancapitalwasimportantindetermininglabourproductivity(Connolly,et al.2012).

Afirm’scompetitivenessisattributedtoR&D,innovationandcollaborativeeffortsbetweenfirms.InAustralia,however,notmanystudieshavelookedintotherelationshipbetweentheseandSTEMemployment.ButintheU.S.,Lieponen(2005)examined the complementarity between employees’ skills and firms’ innovationactivities and found that high technical skills are complementary with R&Dcollaborationandproductorprocessinnovationandthathumancapitalareseenasan enabling factor in profitable innovation.The study however did not specificallyseparateSTEMskills.

TheABS has investigated several related topics in the past, such as skillsshortages, information and communication technology (ICT), flexible workingarrangements,governmentassistance,innovationandproductivity,butnoneofthesespecificallylookedattheeffectsofSTEMskillsandSTEMemployment(see,Wong,et al.2007;BrunkerandO-Fischer,2008;ABS,2008;TodhunterandAbello,2011;Rotaru,DzhumashevaandSoriano,2013;andRotaru,2013).

This paper investigates the relationship between the use of STEM/Non-STEMskills, collaboration,R&Dand innovation, includingnoveltyof innovation,among Australian businesses. Specifically, it examines the association of thefollowingfactors:businesssize,industry,collaborationonR&D,foreignownership,market competition, skills shortage, working arrangement, government assistanceandICTintensity,onthelikelihoodofthebusinessinnovatingifitisusingSTEMorNon-STEMskills.TheimpactofhavingR&DandcollaborationontheprobabilityofachievingaspecifiedhighestdegreeofnoveltybetweenbusinessesusingSTEMorNon-STEMskillsisalsoassessed.

Firm level data from theAustralianBureau ofStatistics’ annualBusinessCharacteristics Survey (BCS) are used in the analysis. The study applies probitregressionmodellingonthe2010-11to2011-12wavesoftheBCS.

Conceptual Framework and Definitions Innovationisaprimarydriverofanation’seconomicgrowth.AsAustraliacontinuesto compete in the global economy,Australian businesses need to be innovative toincrease performance (DIISR, 2009). The association between innovation andeconomicgrowthisparticularlyimportantinaneconomywhichissubjecttobindingconstraints on the rate of growth of its primary inputs. In the current economicenvironment,Australia is experiencing an increasing incidence of such constraintsparticularlyinrelationtoskilledlabourmarkets(ABS,2008).

An educated and skilled workforce is essential for successful innovation.Undernormalcircumstances,innovationincreasesproductivityandcreatesprosperity.However, in reality inmanycases, the linkbetween innovationandproductivity isconvolutedanditmaytakemanyyearsforinnovationstoincreaseproductivity.Therelationshipbetweenskillsandinnovationinthelongterm,cycles.Theskillsoftheworkforceandmanagementdeterminetheinnovationthattakesplace,whichthenhelpdeterminethedemandforskillsinabusiness,whichtheninfluencetheinnovationandsoon(Tether,et al.2005).

347FRANKLIN SORIANO AND RUEL ABELLO

Modell ing the Relationships Between the Use of STEM* Ski l ls , Collaboration, R&D, and Innovation among Australian Businesses

Researchandexperimentaldevelopmentworkleadstoinnovation.VanZon’s(2001)modelhaddemonstratedthelinkbetweenskillsandinnovationthroughR&D.PeoplewithSTEMcapabilities(skills,knowledgeandwaysofthinking)areemployedtodriveR&Dwork.STEMcapabilitiescomeprimarilyfromthosewithformalSTEMqualifications,althoughsomepeopleemployedinoccupationsrequiringSTEMskillsmayhavenon-STEMfieldsastheirqualification.TheliteraturereviewofStanwick(2011) discussed the kind of skills which contribute to innovation. The latter alsoconcluded that a good educational foundation is the key to promoting successfulinnovativepractice.

Figure1belowillustratestherelationshipsbetweentheuseofskilledworkers,R&Dandinnovation.

Figure 1 - Conceptual framework for the analysis of STEM, R&D, and Innovation

Skillsaretypicallyacquiredfromformaleducation,generallifeexperienceandlearningonthejob.However,whenpolicymakerstalkabout‘skills’theygenerallyarereferring to theskillsobtainedfromformaleducation.Oftenqualificationsandeducationlevelsareusedasproxyforskills,for thesimplereasonthat theycanbequantified(Karmel,2012).

STEM qualifications are defined according to the Australian Standard Classification of Education, 2001, as those of Postgraduate degree level, Masterdegreelevel,GraduatediplomaandGraduatecertificatelevel,Bachelordegreelevel,Advanceddiplomalevel,andCertificatesIIandIVlevels–inanyofthefieldsbelow:

• NaturalandPhysicalSciences(NPS)(includingMathematicalSciences)• InformationTechnology(IT)• EngineeringandRelatedTechnologies(ERT)• Agricultural,EnvironmentalandRelatedStudies(AERS)

However,forthisstudy,STEMskillsvariableshavebeenconstructedbasedonthetypeofskillsusedbyabusinessasreportedintheBCS.AbusinessisconsideredtohaveusedSTEMskillsifitreportedusinganyofthefollowingskills:Engineering,ScientificandResearch,ITprofessionals,andITsupporttechnicians.ThesearebasedonsubjectiveresponsesbybusinessestotheBCSquestionaboutthetypesofskillsusedinundertakingcorebusinessactivities.

�S�T�E�M� �s�k�i�l�l�s

�N�o�n�-�S�T�E�M�s�k�i�l�l�s

�S�T�E�M�e�m�p�l�o�y�m�e�n�t �R�&�D �I�n�n�o�v�a�t�i�o�n


Researchandexperimentaldevelopment(R&D)comprisescreativeworkundertakenonasystematicbasisinordertoincreasethestockofknowledge,includingknowledgeofpeople,cultureandsociety,andtheuseofthisstockofknowledgetodevisenewapplications.R&Dcontainsanappreciableelementofnoveltyandscientificortechnologicaluncertainty,orrisk,toovercome.

The R&D performed by business is investigative work that has actual orpotential use in the development of new or enhancedmaterials, products, devices,processes,systemsorservices(seeOECD,2002,pp.30-48forthecompletedefinitionandconventionofR&D).

Inthecurrentinvestigation,twoR&Dvariableshavebeenconstructedandusedinthemodelling.Thefirstoneisbusinessinvolvementinco-operativearrangementsin joint researchanddevelopment (i.e., collaboration inR&D).Tocollaborate is toparticipate in jointprojectswithotherbusinessesororganisations (includingwiderpartsofthebusiness’enterprisegroup).ThesecondR&Dvariable(i.e.,expenditureonR&D)isusedinthemodellingofinnovationnovelty.Thisvariableindicateswhetherornotinnovatingbusinesseshaveinvestedinresearchandexperimentaldevelopment.

ThedefinitionofinnovationfollowstheOsloManualas‘…the implementation of a new and significantly improved product (good or service); or process, a new marketing method, or a new organisational method in business practices, workplace organisation or external relations.’(OECD,2005,p.46)

Abusinessiscalled‘innovation-active’ifitengagedinanyinnovationactivitiesthatwereimplemented,ongoingorabandonedduringaperiod.Abusinessiscalledan ‘innovator’ if it successfully developed and implemented an innovation, whichmayhavetakenmanyyearstocomplete.Thecurrentinvestigationisconductedfor‘innovation-active’businesses.

Four types of innovation are covered in the analysis: product innovation,processinnovation,organisationalormanagerialinnovation,andmarketinginnovation(seeOECD,2005,pp.48-53forthefulldescriptionofthedifferenttypesofinnovation).Notethatabusinesscoulddomorethanonetypeofinnovation.

It is important to note that many of the key variables used in this studyare ‘self-reported’ rather thanobjectivelymeasured.That is, adegreeof subjectivejudgement on the part of the business respondent is inherent in theBCS data; forexample,questions relating to thehighestdegreeof innovationnovelty (new to theworld,newtoAustralia,newtotheindustryornewtothebusiness).Insuchcases,themodelled outcomes should be properly interpreted as relating to the predictedlikelihoodofcertainoutcomes.

AppendixA1givesmoredefinitions,includingthoseofthevariablesusedinthemodels.

2. Data The analysis used data from the ABS Business Characteristics Survey which hasdetailedinformationonthetypesofskillsusedinthebusiness,businessdemographics,innovationactivity,ICTusage,R&Dexpenditure,innovationnovelty,andmanyothervariables relevant to the analysis. The study utilised firm level data forAustralianbusinessescoveredby the2010-11 to2011-12wavesof theBCS.Thedata isbased



ontheANZSIC2006classificationofindustry.ForeachwaveoftheBCS,auniquesamplewas constructed from the responses of the small,medium-sized, large andcomplexbusinesses.(SeeABS(2013b)formoreinformationaboutthesurvey.)

3. Methodology Theanalysisusedstandardprobitmodelling(seeWooldridge,2010)tohelpanswertheresearchquestionsabove.Anumberofmodelshavebeenestimated-threewerebinaryprobitmodelsand threewereorderedprobitmodels.Amodelwas run toestablishtherelationshipbetweentheuseofSTEM/Non-STEMskills,R&DandthedegreeofnoveltyofinnovationamongAustralianinnovatingbusinesses.Weightswerenotusedinthemodellingasthevariablesusedinthesamplestratificationdesign(i.e.,industryandemployment-basedsize)werealreadyincludedasexplanatoryvariables.

ThestudyalsoexaminedtheimpactoftheuseofSTEM/Non-STEMskillsonbusinessinnovationandontheprobabilityofachievingaspecifiedhighestdegreeofnovelty.Themarginaleffects(orpredictedprobabilities)werealsocomputed.

AllmodellingprocedureswereexecutedusingSASandSTATA12.AppendixA2showsthedetailsoftheregressionmodels.

4. Results Cross-tabulations TheBCScoversfourbroadtypesofinnovation(goodsorservices,operationalprocesses,organisational/managerialprocesses,andmarketingmethods)acrossthreeinnovationstatuses(introduced,stillindevelopment,andabandoned).Businesseswereaskedtoindicateiftheyhadintroducedanyneworsignificantlyimprovedtypeofinnovationduringthereferenceperiod(i.e.,yearending30June2011for2010-11andyearending30June2012for2011-12).Businessescouldreportmultipletypesofinnovationand/ormultipleinnovationstatuses.IntheBCS,allbusinesseswerealsoaskedtoreportthetypesofskillsusedinundertakingtheircorebusinessactivitiesduringthereferenceperiod.Again,businessescouldreportmultipletypesofskillsused.

Table1presentstheproportionofAustralianbusinessesthatengagedinanyinnovationactivitiesbyskillsusedinthetworeferenceyears.Thetablealsoshowstheestimatedproportion for innovationactivebusinesses, i.e., those thatundertookany innovative activity irrespective ofwhether the innovationwas introduced, stillindevelopmentorabandoned.Readersmayagainrefer toappendixA1for thefulldescriptionofthefivecategoriesofskillsusedinthetable.


Tab

le 1

- P

rop

ort

ion

of

Inn

ova

tin

g a

nd

No

n-i

nn

ova

tin

g B

usi

ne

sse

s u

sin

g S

TEM

an

d O

the

r Sk

ills,

201

0-1

1 a

nd

201

1-12

Busin

esse

s

Bu

sines

ses

Busin

esse

s tha

t Bu

sines

ses

Busin

esse

s

Busin

esse

s

tha

t tha

t int

rodu

ced

that

with

Busin

esse

s tha

t did

intro

duce

d int

rodu

ced

any n

ew or

int

rodu

ced

innov

ative

wi

th an

y no

t hav

e any

-

an

y new

or

any n

ew or

sig

nifica

ntly

any n

ew or

ac

tivity

inn

ovati

ve

innov

ative

signifi

cantl

y sig

nifica

ntly

impr

oved

sig

nifica

ntly

which

was

ac

tivity

ac

tivity

(non

Es

timate

d im

prov

ed

impr

oved

or

ganis

ation

al/

impr

oved

ab

ando

ned

(inno

vatio

n–

innov

ation

–

numb

er of

go

ods

oper

ation

al ma

nage

rial

mark

eting

or

still

in

activ

e ac

tive

bu

sines

ses

or se

rvice

s pr

oces

ses

proc

esse

s me

thods

de

velop

ment

busin

esse

s) bu

sines

ses)

(A

) (B

) (C

) (D

) (E

) (F

) (G

) (H

)Sk

ills u

sed

’000

%

%

%

%

%

%

%

2011-12

STEM

andN

on-STE

Msk

ills

224

29.1

31.9

37.4

31.8

44.9

66.7

33.3

STEM

skillso

nly

44

26.7

21.0

23.9

22.2

32.8

52.3

47.7

OtherN

on-STE

Msk

illso

nly

387

16.9

13.6

17.2

14.6

19.0

37.5

62.5

Tradesan

dothe

rNon-STE

Msk

ills

47

17.7

18.6

21.5

21.4

30.0

49.4

50.6

Tradeskillsonly

74

10.6

8.5

10.9

9.4

12.1

28.3

71.7

TOTA

L776

20.4

19.1

23.0

19.9

27.3

46.6

53.4

2010-11

STEM

andN

on-STE

Msk

ills

202

28.0

28.7

33.0

28.1

36.6

60.0

40.1

STEM

skillso

nly

50

23.4

19.2

20.0

15.8

24.9

45.5

54.5

OtherN

on-STE

Msk

illso

nly

397

12.9

11.0

12.5

12.6

15.6

29.8

70.2

Tradesan

dothe

rNon-STE

Msk

ills

50

11.1

13.1

17.3

12.2

23.6

35.8

64.2

Tradeskillsonly

65

10.4

11.4

14.2

11.6

9.6

28.8

71.2

TOTA

L764

17.3

16.4

18.9

16.8

21.8

39.1

61.0

Note:Colu

mnAco

ntains

theestima

tednu

mberofAu

stralianbu

sinessesbyskillsused,andisthed

enom

inatorusedinca

lculating

thep

roportionsinc

olumn

sBtoH.C

olumn

sBtoE

conta

inthe

proportiono

fbusine

ssesthatin

troduceda

particulartypeofinnovationb

yskillsused.C

olumn

Fconta

insthep

roportionofbu

sinessesw

ithan

yinnovativeac

tivitywhic

hwa

sabandonedorstillinde

velop

mentbysk

illsu

sed.Co

lumnsGan

dHco

ntainthe

perce

ntagefrequencyd

istrib

utionsforinnovation-activea

ndno

n-innovation-ac

tiveb

usine

ssesin

eachsk

illsu

sedc

ategory.Colu

mnGplusco

lumnH

equals100%

.



Skills used and innovation-active businesses Fromtable1,weobservedthattheproportionofbusinessesthatwereinnovation-activein2011-12was66.7percent,asignificantincreaseof6.7percentagepointsfromthepreviousyear.Forbusinesses that indicatedusingTradeskillsonly,andOthernon-STEMskillsonly, theestimatedproportionofnon-innovation-activebusinesseswassignificantlyhigherthantheproportionofinnovation-activebusinesses(inbothyears).

Skills used and different types of innovation BusinessesthatusedacombinationofSTEMandNon-STEMskillsweresignificantlymore likely toengage inanyoneof the threebroad typesof innovation thanotherbusinesses. There was no significant difference between businesses using STEMskillsonlyandSTEMandNon-STEMskillsintermsoftheproportionengagedingoodsandservicesinnovationin2011-12.Inaddition,businessesusingthesaidskillscategorieswere significantlymore likely to engage in this type of innovation thanotherbusinesses.BusinessesthatreportedusingSTEMandNon-STEMskillsweresignificantlymore likely thanotherbusinesses tohaveabandonedor stillbe in thedevelopmentprocessofinnovationactivities.

Modelling and Impact Analysis Belowaretheselectedkeyfindingscomingfromthedifferentmodels.TheregressionmodellingoutputsareinappendixA3.

For the purpose of this paper, all calculated marginal effects are withreferencetoafirmthatissmallinsize,hasmoderateICTintensity,noskillshortagewithin business, no skill shortage within labour market, no effective competition,100percentAustralianowned,noflexibleworkingarrangements,and receivednogovernmentassistance.

STEM skills and innovation TheuseofSTEMskillsisstronglyassociatedwiththelikelihoodofinnovation.Thepredictedprobabilityofbeinganinnovatorrisesfrom24.1percentforanon-STEMskillsuserto38.3percentforaSTEMskillsuserin2011-12.

Other factors associated with innovating businesses AlsopositivelyandsignificantlyassociatedwithinnovationarehigherICTintensity,having minimal to strong degree of market competition, having flexible workingarrangements,having>0percent to50percent foreignownership, lackingskilledstaff,andreceivinggovernmentfinancialassistance.

Cooperative arrangements or collaboration in R&D HavingcooperativearrangementsorcollaborationinR&Disalsofoundtobestronglyassociatedwithinnovation.ThecombinedimpactofcollaborationinR&DanduseofSTEMskillsonthelikelihoodofinnovationisverystrong.CollaboratingbusinesseswhichuseSTEMskillshavea48.3percent(in2010-11)and53.0percent(in2011-12) chances of innovating. A business which does not engage in any cooperative


arrangement in R&D aswell as not using any STEM skills has lower chances ofinnovating(i.e.,19.2percentin2010-11and22.9percentin2011-12).

Business size Whereresultsaresignificant,businesssizeisstronglyassociatedwithanincreaseinthepredictedprobabilityofinnovationifthebusinessusesSTEMskills.Inaddition,generally for all business sizes, the model predicts that a reference firm in theManufacturingindustrythatcollaborateshasmorechancesofinnovatingthanafirmthatdoesnotengageincollaboration.Specifically,for2011-12,themodelpredictsthatamicro(1-4employees)businesshasa56.8percentchanceofinnovatingifitusesSTEMskills,comparedwith60.5percentforsmall(5-19employees),57.9percentformedium(20-199employees),and52.7forlarge(200+employees)businesses.

Use of STEM skills across industries TheprobabilityofinnovationisrelativelyhigherforabusinessthatusesSTEMskillscomparedwith a business that does not use STEM skills, in all of the industries.Relativetothemanufacturingindustry,financialandinsuranceservices(in2010-11only),businessesinretailtradeandwholesaletradeindustrieshaverelativelyhigherlikelihoodofinnovation.

Types of innovation TheuseofSTEMskillsandcollaborationinR&Darestronglyassociatedwithhigherlikelihoodofinnovatinginalltypesofinnovation.In2011-12,forgoodsandservicesinnovation, collaborating businesses which use STEM skills have 32.1 per centprobabilityofinnovatingcomparedwith22.3percentforbusinessesthatdonotuseSTEMskills.Foroperational processes innovation, collaboratingbusinesseswhichuse STEM skills have 25.9 per cent likelihood of innovating comparedwith 15.8percent forbusinesses thatdonotuseSTEMskills.Fororganisational/managerialprocessesinnovation,collaboratingbusinessesthatuseSTEMskillshave31.8percentlikelihoodofinnovationcomparedwith20.4percentforbusinessesnotusingSTEMskills. Formarketingmethods innovation, collaborating businesses that useSTEMskillshavea15.6percent likelihoodof innovationcomparedwith9.5percent forbusinessesnotusingSTEMskills.Theresultsweresimilarfor2010-11.

Degree of novelty in innovation InnovatingbusinessesthatuseSTEMskillsaresignificantlymorelikelytoachieveahigherdegreeofnoveltyofinnovationthaninnovatingbusinessesthatdonotuseSTEM skills. On average, a small non-collaborating Australian-owned innovatingbusinessinmanufacturingthatisnotengagedinR&D,butisusingSTEMskills,is63.6percentmorelikelytoachieveahighestdegreeofnoveltyof‘newtotheworld’thanafirmnotusingSTEMskills;or39.5percentmorelikelytoachieveahighestdegreeofnoveltyof‘newtoAustralia’thanafirmnotusingSTEMskills.

Having expenditure inR&D is significantly associatedwith an increase inthepredictedprobabilityofachievinga ‘new to theworld’ innovation.Themodel



predictsthatasmallnon-collaboratingAustralianownedinnovatingbusinessintheManufacturingindustrythatisinvestinginR&Dhasan11.2percentprobabilityofhaving a ‘new to theworld’ innovation,comparedwith3.6percent fora similarbusinessthatisnotengagedinR&D.

Having cooperative arrangements in R&D is associated with statisticallysignificant but relatively modest increases in the probability of higher degrees ofinnovation novelty among innovating businesses. This probability is higher if thebusiness uses STEM skills. Innovating businesses having collaboration in R&D,expenditureinR&D,andusingSTEMskillsare15percentmorelikelytoachieve‘newtotheworld’innovation,comparedwith10.7percentforsimilarbusinessesthatdonotuseSTEMskills.

5. Conclusion ThispaperinvestigatedtherelationshipbetweentheuseofSTEM/Non-STEMskills,collaboration, R&D and innovation among Australian businesses. It explored theassociation between use of skills and innovation for each type of innovation (i.e.,goods and services, operational processes, organisational/managerial processes,marketingmethods).Italsoexaminedtherelationshipbetweenthedegreeofnoveltyof innovationachievedbyinnovatingAustralianbusinesses,whether theyareusingSTEM skills or not. Other relevant business characteristics such as business size,industryofoperation,skillshortages,degreeofmarketcompetition,degreeofforeignownership, ICT intensity, flexibleworking arrangements, and governmentfinancialassistancewerealsotakenintoaccountalongsidecollaborationandR&D.

ThedataavailablefromtheABSBCShaveproventobeadequatetomodeltheaboverelationshipsusingstandardprobitmodellingprocedures.

The probit regression models found that the use of STEM skills andcollaborationinR&Dareverystronglyassociatedwithanincreaseinthelikelihoodofbeinganinnovatingbusiness.Inmostcases,thefollowingvariables:ICTintensity,market competition, lacking skilled staff, flexible working arrangements andgovernmentfinancialassistancearefoundtobesignificantlyassociatedwithahigherlikelihoodofbeingan innovator.Thepredictedprobability for achievingahighestdegreeofinnovationnoveltyishigherforaninnovatingbusinesswhichisusingSTEMskillsandhasinvestedinR&DthanforaninnovatingbusinessnotusingSTEMskillsandhasnotinvestedinR&D.

PossiblefutureworkthatwouldfurtherenhancethecurrentinvestigationistoconsiderlabourproductivityandlinkitwithinnovationanduseofSTEM/Non-STEMskills.TheAustralianBureauofStatisticsiscurrentlyintheprocessofcreatingan‘ExpandedAnalyticalBusinessLongitudinalDatabase(EABLD)’thatcouldprovidearichdatasourceformicrolevellabourproductivityanalysisandpaneldatamodelling.


AppendicesA1 Definitions of Variables Thissectiondescribesthecompilationofthevariablesusedintheanalysis.

Innovation Thescopeofinnovativeactivity,asmeasuredbytheBCS,followstheOsloManual(OECD,2005)andcoversfourbroadtypesofinnovation:

• Goods or services–Anygoodorserviceorcombinationofthesewhichisnewtoabusiness(orsignificantlyimproved).Itscharacteristicsorintendedusesdiffersignificantlyfromthosepreviouslyproduced/offered.

• Operational processes–Neworsignificantlyimprovedmethodsofproducingordeliveringgoodsorservicesofabusiness(includingsignificantchangeintechniques,equipmentand/orsoftware).

• Organisational/managerial processes – New or significantly improvedstrategies, structures or routines of a business which aim to improveperformance.

• Marketing methods – New or significantly improved design, packaging orsalesmethodsaimedtoincreasetheappealofgoodsorservicesofabusinessortoenternewmarkets.

Therearethreestatusesofinnovation,namely:• Introduced or implemented – the business successfully introduced orimplemented an innovation during the reference period (although theinnovationdoesnotneedtohavebeencommerciallysuccessful);

• Still in development – the business was in the process of developing,introducingor implementingan innovationduring the referenceperiodbutworkontheinnovationwasstillinprogressattheendoftheperiod;and,

• Abandoned–thebusinessabandonedthedevelopmentand/orintroductionofaninnovationduringthereferenceperiod(i.e.,workontheinnovationceasedwithoutfullintroductionoccurring).

Abusinessiscalled‘innovation-active’ifitengagedinanyinnovationactivitiesthatwereimplemented,stillindevelopmentorabandonedduringtheperiod.NotethatintheBCS,businessescouldreportmorethanonetypeofinnovation.

TheBCSalsoaskedthedegreeofnoveltyofinnovationachievedbyinnovatingAustralianbusinesses.Thedegreeofnoveltyarecategorisedas:

• Innovationisnewtotheworld;• InnovationisnewtoAustraliabutnotnewtotheworld;• InnovationisnewtotheindustrywithinAustraliabutnotnewtotheworldorAustralia;and,

• Innovationisnewtothebusinessonly.Thetablebelowdescribesthedifferentinnovation(dependent)variablesusedformodelling.



Description Range of ValuesInnovation(binary) 0/1dummyFirmengaged/notengagedinanytypesofinnovation(i.e.,overallmeasureofinnovation)Innovation(binary)–forparticulartypeofinnovation 0/1dummyFirmengaged/notengagedinthistypeofinnovation,say• Goodsandservices• Operationalprocesses• Organisational/Managerialprocesses• MarketingmethodsInnovationDiversity(categorical) 0to4• Noinnovationactivityatall• Exactly1typeofinnovation• Exactly2typesofinnovation• Exactly3typesofinnovation• Exactly4typesofinnovationInnovationNovelty(categorical) 3to0• Innovationisnewtotheworld• InnovationisnewtoAustraliabutnotnewtotheworld• InnovationisnewtotheindustrywithinAustraliabut notnewtotheworldorAustralia• Innovationisnewtothebusinessonly

STEM Skills Two forms of STEM/Non-STEM skills variables have been constructed, a binaryand a categorical. The categorical variablewas designed to refine and capture theassociationoftheusethedifferenttypesofskills(e.g.,STEM;Non-STEM;Trade;andOtherNon-STEM)onbusinessinnovation.

Description Range of ValuesSTEMSkills(binary) 0/1dummyFirmused/notusedanyofthefollowingtypesofSTEMskills• Engineering• ScientificandResearch• ITprofessionals• ITsupporttechniciansSTEM/Non-STEMSkills(categorical) 0/1dummy• FirmdiduseanySTEMskillsandNon-STEMskills (eachcategory) (i.e.,TradeorOtherNon-STEM–Transport,plantand machineryoperation;Marketing;Projectmanagement; Businessmanagement;andFinancial)• FirmdiduseSTEMskillsonly• FirmdiduseTradeskillsonly• FirmdiduseTradeandOtherNon-STEMskillsonly• FirmdiduseOtherNon-STEMskillsonly


Collaboration and R&D ThefollowingcollaborationandR&Dindicatorswerealsocompiled.

Description Range of ValuesCollaborationinR&D(binary) 0/1dummyBusinesswasinvolvedinco-operativearrangementforjointresearchanddevelopment(R&D)ExpenditureonR&D(binary) 0/1dummyBusinessreportedhavingexpenditureonresearchandexperimentaldevelopmenteitheracquiredfromotherfirmsorperformedbythebusinessesitself,forinnovation

Selected key business characteristics Theotherkeybusinesscharacteristicsemployedinthemodellingaredescribedbelow.Theselectionof thekeybusinesscharacteristicshasbeenmainlybasedonthetworecent researchpublicationsofABSon innovation.SeeRotaru (2013);andRotaru,et al. (2013) for more information about the justification for their selection. SkillshortageindicatorshavebeenaddedfollowingtheABS(2008)papersubmissiontotheInnovationReview.

Description Range of ValuesNumberofemployees(businesssize) 0/1dummy• 1-4Employees (eachcategory)• 5-19Employees• 20-199Employees• 200+EmployeesDegreeofcompetitioninthemarket 0/1dummy• Noeffectivecompetition(0competitor) (eachcategory)• Minimal(1-2competitors)• Moderatetostrong(3ormorecompetitors)Degreeofforeignownership 0/1dummy• 100percentAustralianowned (eachcategory)• >0percentto50percentforeignowned• >50percentforeignownedIndustrydivision 0/1dummy(ANZSIC2006) (eachcategory)• Agriculture,ForestryandFishing• Mining• Manufacturing• Electricity,water,gasandwasteservices• Construction• Wholesale• RetailTrade• Accommodationandfoodservice• Transport,postalandwarehousing• Information,mediaandtelecommunications• Financialandinsuranceservices• Rental,hiringandrealestateservices• Professional,scientificandtechnicalservices• Administrativeandsupportservices



Description Range of Values• Healthcareandsocialassistance• Artsandrecreationservices• OtherservicesICTintensity 0/1dummy• Mostintense (eachcategory) Business had broadband connection, web presence, and places or receives orders via the internet or web• High Business had broadband connection, web presence, but does not receive orders via the internet or web• Moderate Business had broadband connection, but has no web presence• Low Business does not use broadband connectionFlexibleWorkingArrangement(binary) 0/1dummyFirmofferedthefollowingworkingarrangementstotheiremployees:• Flexibleworkinghours• Flexibleleave• Jobsharing• WorkingfromhomeGovernmentfinancialassistance(binary) 0/1dummyFirmreceived/notreceivedanyformofassistance(i.e.,grants,on-goingfunding;taxconcession;subsidies;rebates;andothergovernmentfinancialassistance)Skillsshortage(binary) 0/1dummy• Firmreportedthatlackofskilledpersonswithinthebusiness (each) significantlyhamperstheirinnovation• Firmreportedthatlackofskilledpersonswithinthelabour marketsignificantlyhamperstheirinnovation

A2 Methodology The following methodological approaches have been used to answer the researchquestions.

Probit regression Toaddresstheresearchquestion:• What is the relationship between the use of STEM/Non-STEM skills, collaboration

in R&D and innovation among Australian businesses?Fivemodels(i.e.,Models1-5)havebeenspecifiedandestimatedusingthestandardprobitmodellingprocedure.

Model 1. Binary probit with binary STEM skillThefirstmodelisgivenbyInnovation = Binary Probit(STEM skills,X2)where:Innovation–Binaryvariabletakingthevalue1ifthebusinesswasinnovativeand0otherwiseSTEM skills – Binary


X2 stands for the vector of other variables included in themodel. These variablesincludedthefollowing:

• Businesssize• Industryofoperation• CooperativeR&D• MarketCompetition• ForeignOwnership• FlexibleWorkingArrangement• GovernmentFinancialAssistance• ICTintensity• Skillshortage.

Note that inorder tospecify thebinaryprobitmodelwecan follow the traditionalapproachofusingthelatentvariabley*

itodeterminethebinaryvariable,yi,abinaryvariableindicatingwhetherafirminnovated(i.e.,yi =1)ornot(i.e.,yi =0)thelatentvariableisgivenby

y*i=X1,i b+ei∀i =1,…,N

where ei is the random error term (which is assumed to have a standard normaldistribution), N stands for total number of businesses, X1,i is a (1 x k) vector ofconditioningvariablesforbusinessi(i.e.X2includingtheSTEMskills),andbisavectorofparameterscorrespondingtothekconditioningvariables.However,asy*

iisunobservedinpractice,weusedtheobserveddichotomousvariableyiwhichindicatesthesignofthelatentvariable,y*

i.Notethatyiisgivenbyyi=

Followingthisapproach,thebinaryprobitmodelisspecifiedby

P(yi=1|X1)=P(y*i>0|X1)=Φ(X1b)

whereΦ(.)isthestandardnormalcumulativedistributionfunction.Notethatthevaluesoftheparametervectorbwereestimatedusingmaximumlikelihoodestimation(MLE).

Model 2. Binary probit with categorical STEM/Non-STEM skillsThesecondmodelisthesameasthefirstmodelbutwithcategoricalskillsvariable.Thecategoricalvariable isdesigned tocapture theassociationof theuse theothertypesofskills(i.e.TradeandNon-STEM)oninnovation.Theskillsvariableisdefinedasfollows:

• FirmdiduseanySTEMskillsandNon-STEMskills(i.e.,TradeorOtherNon-STEM–Transport,plantandmachineryoperation;Marketing;Projectmanagement;Businessmanagement;andFinancial)

• FirmdiduseSTEMskillsonly• FirmdiduseTradeskillonly• FirmdiduseOtherNon-STEMskillsonly• FirmdiduseTradeandOtherNon-STEMskillsonly

1, ify*i>0

0,otherwise.



Model 3. Binary probit for a particular type of innovationThe thirdmodel isavariationofmodels1and2with thedependentvariable (i.e.,innovation)takingavalueof1ifthefirmperformedaparticulartypeofinnovation(e.g.,goodsandservices)and0otherwise.ThismodelisrunforeachofthefourtypesofinnovationwitheitherabinarySTEMskillsorcategoricalskillsvariables.

Model 4. Ordered probit with binary STEM skillsThefourthmodelisanorderedprobitgivenby:Innovation = Ordered Probit (STEM skills, X2 ).

In this case, all covariates are the same as in Model 1 but Innovation iscategorical (or polychotomous) having the following subcategories: no innovation,exactly 1 type of innovation, exactly 2 types of innovation, exactly 3 types ofinnovation,andexactlyfourtypesofinnovation.

The derivation of thismodel is not too different from the previous binaryprobitmodelderivation.Thedifferencenowconsists inthefact that theinnovationvariableyihasfivevalues,0 if there isno innovation,1 if thebusinessengaged inexactly one type of innovation, 2 if the business engaged in exactly two types ofinnovation,3ifthebusinessengagedinexactlythreetypesofinnovation,and4ifthebusinessengagedinexactlyfourtypesofinnovation.

Byconsideringthelatentvariabley*iandthethresholdparameters(g1,g2,…,

g4),yicanbedeterminedas

0,ify*i≤g1

1,ifg1<y*i≤g2

yi = 2,ifg2<y*i≤g3

3,ifg3<y*i≤g4

4,ify*i>g4

Themodelisthenspecifiedby:

P(yi=0|X1)=P(y*i≤g4|X1)=1−Φ(X1,i b−g1)

P(yi=1|X1)=P(g1<y*i≤g2|X1)=Φ(X1,i b−g1)−Φ(X1,i b−g2)



P(yi=4|X1)=P(y*i>g4|X1)=Φ(X1,i b−g4)

where,asbefore,Φ(.)standsforthestandardnormalcumulativedistributionfunction.OnceagaintheparameterswereestimatedusingMLE.


Model 5. Ordered probit with categorical STEM/Non-STEM skillsThefifth innovationmodel is the same as the fourthmodel butwith a categoricalskillsvariable.Inthiscase,allcovariatesarethesameasinModel2.Toaddresstheresearchquestion:• What is the relationship between the use of STEM/Non-STEM skills, collaboration

inR&D, R&D expenditure and the degree of novelty of innovation among Australian innovating businesses?

another model has been specified and estimated again using the standard probitmodellingprocedure.Themodelisrunforasampleofinnovatingfirmsonly.

Model 6. Ordered probit on innovation novelty with binary collaborationThesixthinnovationmodelisanorderedprobitgivenby:Degree of Innovation Novelty = Ordered Probit (Skills, Collaboration in R&D, Expenditure on R&D, X3). where:DegreeofInnovationNoveltyiscategorical(orpolychotomous)havingthefollowingsubcategories:

• Innovationisnewtothebusinessonly• InnovationisnewtotheindustrywithinAustraliabutnotnewtotheworldorAustralia

• InnovationisnewtoAustraliabutnotnewtotheworld• Innovationisnewtotheworld

Skills–EitherabinarySTEMskillsorcategoricalskillsvariablesX3standsforthevectorofothercovariatesincludedinthemodel.Theseare:

• Businesssize• Industryofoperation• ForeignOwnership.

TheestimationprocedurefortheabovemodelfollowsthatofModel4.

Impact Analysis Themodellingconductedhereareallcross-sectionalinnatureandassuchitisnotpossible to establish the existence or direction of ‘causality’ between the variousconditioning (business characteristics) variables and innovation.While theremaybebroadlyor particularlyprior views in relation to causality, the current analysiscan establish only statistical association between the conditioning variables andinnovation.

However,theimpactofusingSTEM/Non-STEMskillsonthelikelihoodofbusinessinnovatingcanbeinvestigated.Also,theimpactofusingSTEM/Non-STEMskillsontheprobabilityofengaginginaparticulartypeofinnovationcanalsobemeasured.Hence,tocomplementtheinterpretationoftheregressionresults,theall-else-equalincrementalimpactsofselectedconditioningvariablesarealsocalculated.Thesearereferred toas‘marginal’effectseventhoughtheconditioningvariableswereinalmostallcasesdiscrete.Forbinaryvariables(e.g.,useofSTEMskillvs.notuseofSTEMskill),the‘marginal’effectwastheincrementtotheprobabilityarising



fromvaryingthevariablefrom0to1.Whileforcategoricalvariables,theeffectwastheincrementtotheprobabilityfrommovingfromthereferencevaluetothevalueinquestion.

TheDITR(2006,pp.32-34;and2007,p.40-41)studiesprovideanillustrationof the methodology behind the estimation of this ‘marginal’ effect (i.e. impact onprobability).

A3 Selected Regression Results Table A1 - Results of the Probit (binary) Model for Innovation with STEM Skills as Binary

Variable 2010-11 2011-12Intercept -0.548*** -0.081STEMskills Not used STEM skills UsedSTEMskills 0.440*** 0.405***Skillsshortagewithinbusinesshamperinginnovation No skills shortage Haveskillsshortage 0.316*** 0.368***Skillsshortagewithinlabourmarkethamperinginnovation No skills shortage Haveskillsshortage 0.147*** 0.296***ICTIntensity ICT_intensity=1 (Most intense) ICT_intensity=2(High) -0.318*** -0.372*** ICT_intensity=3(Moderate) -0.666*** -0.741*** ICT_intensity=4(Low) -0.824*** -1.055***Numberofemployees 1-4employees -0.101** -0.095 5-19 employees 20-199employees -0.010 -0.067 200+employees -0.122** -0.197***Marketcompetition No effective competition Minimal 0.333*** 0.310*** Moderatetostrong 0.367*** 0.281***Foreignownership 100% Australian owned Foreignownership>0%to50% 0.226** 0.236** Foreignownership>50% 0.056 0.047CollaborationinR&D JointR&D(co-operative)agreement 0.387*** 0.413*** No joint R&D (co-operative) agreementFlexibleWorkingArrangement Haveflexibleworkingarrangements 0.378*** 0.296*** No flexible working arrangementGovernmentfinancialassistance Receivedgovernmentassistance 0.152*** 0.189*** Not received government assistanceIndustry Manufacturing Accomodationandfoodservices -0.067 -0.205** Administrativeandsupportservices -0.105 -0.103 Agriculture,forestryandfishing -0.048 -0.212*


Table A1 - Results of the Probit (binary) Model for Innovation with STEM Skills as Binary (continued)

Variable 2010-11 2011-12 Artsandrecreationservices -0.042 -0.020 Construction -0.181** -0.267*** Electricity,water,gasandwasteservices 0.051 -0.152 Financialandinsuranceservices 0.167* -0.140 Healthcareandsocialassistance -0.023 -0.079 Information,mediaandtelecommunications 0.035 -0.079 Mining -0.471*** -0.408*** Otherservices 0.028 -0.141 Professional,scientificandtechnicalservices -0.087 -0.234*** Rental,hiringandrealestateservices -0.183** -0.185 RetailTrade 0.192** 0.062 Transport,postalandwarehousing -0.229*** -0.216** Wholesaletrade 0.106 0.010Numberofobservations(n) 7548 5554AIC 8765.5 6219.3LogLikelihood -4349.7 -3076.6

Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.

Table A2 - Results of the Probit (binary) Model for Innovation with STEM/Non-STEM Skills as Categorical

Variable 2010-11 2011-12Intercept -0.558*** -0.074STEMskills Other Non-STEM skills only STEMandNon-STEMskills 0.470*** 0.430*** STEMskillsonly 0.354*** 0.229** Tradeskillsonly 0.030 -0.123 TradeandotherNon-STEMskillsonly 0.105 0.129Skillsshortagewithinbusinesshamperinginnovation No skills shortage Haveskillsshortage 0.314*** 0.366***Skillsshortagewithinlabourmarkethamperinginnovation No skills shortage Haveskillsshortage 0.141*** 0.293***ICTIntensity ICT_intensity=1 (Most intense) ICT_intensity=2(High) -0.316*** -0.367*** ICT_intensity=3(Moderate) -0.666*** -0.733*** ICT_intensity=4(Low) -0.822*** -1.051***Numberofemployees 1-4employees -0.099** -0.095 5-19 employees 20-199employees -0.013 -0.073 200+employees -0.131** -0.212***Marketcompetition No effective competition Minimal 0.331*** 0.304*** Moderatetostrong 0.363*** 0.276***



Table A2 - Results of the Probit (binary) Model for Innovation with STEM/Non-STEM Skills as Categorical (continued)

Variable 2010-11 2011-12Foreignownership 100% Australian owned Foreignownership>0%to50% 0.227** 0.238** Foreignownership>50% 0.058 0.045CollaborationinR&D JointR&D(co-operative)agreement 0.385*** 0.409*** No joint R&D (co-operative) agreementFlexibleWorkingArrangement Haveflexibleworkingarrangements 0.373*** 0.292*** No flexible working arrangementGovernmentfinancialassistance Receivedgovernmentassistance 0.147*** 0.186*** Not received government assistanceIndustry Manufacturing Accomodationandfoodservices -0.062 -0.205** Administrativeandsupportservices -0.095 -0.099 Agriculture,forestryandfishing -0.050 -0.230* Artsandrecreationservices -0.033 -0.023 Construction -0.191** -0.265*** Electricity,water,gasandwasteservices 0.049 -0.159 Financialandinsuranceservices 0.173* -0.143 Healthcareandsocialassistance -0.006 -0.072 Information,mediaandtelecommunications 0.048 -0.062 Mining -0.472*** -0.414*** Otherservices 0.026 -0.127 Professional,scientificandtechnicalservices -0.072 -0.221** Rental,hiringandrealestateservices -0.177** -0.189 RetailTrade 0.199** 0.058 Transport,postalandwarehousing -0.227*** -0.224** Wholesaletrade 0.112 0.007Numberofobservations(n) 7548 5554AIC 8767.1 6216.8LogLikelihood -4347.6 -3072.4

Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.


Tab

le A

3 -

Re

sult

s o

f th

e P

rob

it (

bin

ary

) M

od

el f

or

the

Dif

fere

nt

Typ

es

of

Inn

ova

tio

n w

ith

STE

M S

kills

as

Bin

ary

Go

ods a

nd Se

rvice

s Op

erati

onal

Proc

esse

s Or

ganis

ation

al/M

anag

erial

M

arke

ting M

ethod

s Va

riable

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

Intercept

-0.97

6***

-0.714***

-1.008*

**

-0.57

1***

-1.210*

**

-0.81

6***

-1.325*

**

-0.92

3***

STEM

skills

No

t use

d STE

M sk

ills

UsedST

EMsk

ills

0.379**

*0.2

96**

*0.3

83**

*0.3

55**

*0.3

61**

*0.3

54**

*0.2

95**

*0.2

98**

*Skillsshortagewithinbusin

essh

amperin

ginn

ovation

No

skill

s sho

rtage

Ha

vesk

illsshortage

0.202**

*0.3

07**

*0.2

85**

*0.3

00**

*0.3

49**

*0.3

94**

*0.2

87**

*0.2

37**

*Skillsshortagewithinlab

ourm

arket

hamp

eringinnovation

No

skill

s sho

rtage

Ha

vesk

illsshortage

0.132**

*0.1

46**

0.1

49**

*0.2

39**

*0.2

02**

*0.2

20**

*0.1

40**

*0.1

60**

ICTInten

sity

IC

T_int

ensit

y=1 (

Mos

t inte

nse)

ICT_intensity=2(High)

-0.25

7***

-0.25

3***

-0.27

2***

-0.32

4***

-0.21

1***

-0.24

7***

-0.19

9***

-0.35

6***

ICT_intensity=3(Modera

te)

-0.46

5***

-0.54

0***

-0.47

5***

-0.53

6***

-0.43

1***

-0.51

0***

-0.76

6***

-0.84

3***

ICT_intensity=4(Low)

-0.45

6***

-0.64

0***

-0.63

4***

-0.96

5***

-0.67

8***

-0.87

4***

-0.85

2***

-1.033*

**Nu

mberofem

ployees

1-4e

mploy

ees

-0.03

20.0

55

-0.08

0

-0.18

7***

-0.13

5***

-0.22

0***

0.095*

0.027

5-

19 em

ploye

es

20-19

9emp

loyees

-0.13

3**

-0.14

2*

0.115**

-0.04

00.1

01**

-0.04

4-0.05

9-0.21

4***

200+em

ployees

-0.23

2***

-0.24

9***

0.123**

0.0

73

0.041

-0.01

7

-0.29

3***

-0.34

5***

Mark

etcomp

etition

No

effec

tive c

ompe

tition

Minima

l0.3

42**

*0.5

03**

*0.1

82**

0.1

69**

0.1

76**

0.1

84**

0.4

03**

*0.2

29**

Modera

tetostrong

0.309**

*0.4

22**

*0.1

87**

*0.1

59**

0.2

20**

*0.1

77**

*0.4

70**

*0.3

91**

*Foreignow

nersh

ip

100%

Aus

tralia

n own

ed

Foreignow

nersh

ip>0%

to50

%

0.244**

*0.1

17

0.102

0.079

0.110

0.241**

*0.1

67*

0.173*

Foreignow

nersh

ip>50

%

0.142**

0.1

93**

*0.0

59

-0.02

70.0

97*

0.011

-0.00

5-0.00

2



Tab

le A

3 -

Re

sult

s o

f th

e P

rob

it (

bin

ary

) M

od

el f

or

the

Dif

fere

nt

Typ

es

of

Inn

ova

tio

n w

ith

STE

M S

kills

as

Bin

ary

(c

on

tin

ue

d)

Go

ods a

nd Se

rvice

s Op

erati

onal

Proc

esse

s Or

ganis

ation

al/M

anag

erial

M

arke

ting M

ethod

s Va

riable

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

Collaboration

inR&D

JointR&D

(co-opera

tive)agree

ment

0.401**

*0.3

30**

*0.3

54**

*0.3

29**

*0.3

87**

*0.3

99**

*0.2

59**

*0.2

45**

*

No jo

int R

&D (c

o-op

erati

ve) a

gree

ment

FlexibleW

orkin

gArra

ngem

ent

Ha

vefle

xiblewo

rking

arran

geme

nts

0.299**

*0.1

87**

*0.3

03**

*0.2

46**

*0.3

70**

*0.3

70**

*0.3

23**

*0.2

24**

*

No fle

xible

work

ing ar

rang

emen

t

Go

vernme

ntfinancia

lassista

nce

Receive

dgovern

mentassistan

ce

0.146**

*0.1

18**

*0.1

40**

*0.1

11**

0.1

65**

*0.1

28**

*0.1

59**

*0.0

72

No

t rec

eived

gove

rnme

nt as

sistan

ce

Industry

M

anufa

cturin

g

Accomo

dationa

ndfoodservice

s-0.171*

*-0.24

9***

-0.21

9***

-0.45

5***

-0.00

5-0.27

2***

0.290**

*0.3

05**

*

Adminis

trativea

ndsu

pportserv

ices

-0.36

8***

-0.42

4***

-0.08

2-0.26

5**

0.110

0.045

0.105

0.158*

Ag

riculture,for

estry

andfi

shing

-0.43

0***

-0.83

1***

0.017

-0.14

4

-0.03

40.0

39

0.011

-0.02

9

Artsandrecrea

tions

ervice

s-0.20

8***

-0.32

2***

-0.18

9**

-0.31

1***

0.009

-0.26

8**

0.316**

*0.4

14**

*

Constru

ction

-0.56

0***

-0.72

3***

-0.19

5**

-0.27

5***

0.089

0.037

-0.30

2***

-0.17

9**

Electricity,wa

ter,gasan

dwasteservice

s-0.47

5***

-0.718***

0.143

-0.09

20.1

06

-0.12

1-0.14

4-0.17

3

Financialan

dinsuranceservice

s-0.08

0-0.30

5**

0.273**

*-0.07

90.2

79**

*0.0

57*

0.245**

0.0

99

He

althc

arean

dsocialassistan

ce

-0.27

4***

-0.31

9***

-0.11

2-0.24

0***

0.085

0.137*

0.024

0.167**

Infor

mation,me

diaan

dtele

comm

unica

tions0.1

42*

-0.06

5

-0.05

8-0.24

9**

-0.09

7-0.161

0.183**

0.0

98

Mining

-0.90

3***

-1.088***

-0.27

2***

-0.34

9***

-0.23

1***

-0.22

8**

-0.79

3***

-0.73

8***

Otherserv

ices

-0.20

4**

-0.36

8***

-0.12

5-0.36

1***

-0.02

30.0

41

0.241**

*-0.02

0

Profe

ssional,scien

tifica

ndtechnic

alser

vices

-0.28

1***

-0.39

9***

-0.118

-0.33

2***

0.099

-0.06

0-0.01

0-0.07

5

Renta

l,hiringan

drealesta

teservice

s-0.41

6***

-0.57

9***

-0.141

-0.22

5*

0.124

0.108

0.108

0.235**

RetailTrad

e0.0

83

-0.08

7

-0.03

6-0.20

8**

0.125

-0.13

30.3

87**

*0.3

98**

*

Transport,p

ostalan

dware

housing

-0.54

0***

-0.51

3***

0.006

-0.09

80.0

87

-0.07

1-0.23

5***

-0.11

3

Wholesale

trade

0.120

-0.07

1

-0.00

9-0.12

30.1

12

-0.02

40.2

47**

*0.2

80**

*Nu

mberofobservations(n)

7548

5554

7548

5554

7548

5554

7548

5554

AIC

7989.8

6240.7

8490.3

6438.4

8486.9

6483.7

7946.7

6220.6

LogL

ikelihood

-3961.9

-3087.3

-4212.2

-31

86.2

-4210.4

-32

08.9

-3940.3

-3077.3

No

te:**

*,**an

d*de

notesig

nificancea

tthe1%,5%

and1

0%levels,respectively.


Tab

le A

4 -

Re

sult

s o

f th

e P

rob

it (

bin

ary

) M

od

el f

or

the

Dif

fere

nt

Typ

es

of

Inn

ova

tio

n w

ith

STE

M/N

on

-STE

M S

kills

as

Ca

teg

oric

al

Go

ods a

nd Se

rvice

s Op

erati

onal

Proc

esse

s Or

ganis

ation

al/M

anag

erial

M

arke

ting M

ethod

s Va

riable

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

Intercept

-0.95

9***

-0.68

7***

-1.010*

**

-0.54

2***

-1.227*

**

-0.79

9***

-1.345*

**

-0.92

3***

STEM

skills

Ot

her N

on-S

TEM

skill

s only

STEM

andN

on-STE

Msk

ills

0.390**

*0.2

81**

*0.4

11**

*0.3

61**

*0.4

19**

*0.3

66**

*0.3

53**

*0.3

22**

*

STEM

skillso

nly

0.278**

*0.2

16**

0.2

50**

*0.1

21

0.121

0.145

0.056

0.173*

Tradeskillsonly

-0.18

4*

-0.25

2**

-0.04

4-0.25

7**

0.050

-0.16

6

0.079

-0.10

8

Tradea

ndotherN

on-STE

Msk

illso

nly

0.104

-0.01

8

0.104

0.031

0.155*

0.045

0.135

0.160

Skillsshortagewithinbusin

essh

amperin

ginn

ovation

No

skill

s sho

rtage

Ha

vesk

illsshortage

0.202**

*0.3

06**

*0.2

84**

*0.2

97**

*0.3

46**

*0.3

92**

*0.2

83**

*0.2

35**

*Skillsshortagewithinlab

ourm

arket

hamp

eringinnovation

No sk

ills s

horta

ge

Ha

vesk

illsshortage

0.131**

*0.1

50**

0.1

46**

*0.2

43**

*0.1

94**

*0.2

22**

*0.1

32**

*0.1

58**

*ICTInten

sity

IC

T_int

ensit

y=1 (

Mos

t inte

nse)

ICT_intensity=2(High)

-0.25

4***

-0.25

0***

-0.26

9***

-0.31

8***

-0.20

6***

-0.24

2***

-0.19

6***

-0.35

3***

ICT_intensity=3(Modera

te)

-0.46

2***

-0.53

5***

-0.47

2***

-0.52

8***

-0.42

8***

-0.50

3***

-0.76

5***

-0.83

7***

ICT_intensity=4(Low)

-0.45

6***

-0.63

5***

-0.63

4***

-0.96

4***

-0.67

8***

-0.86

8***

-0.85

1***

-1.029*

**Nu

mberofem

ployees

1-4e

mploy

ees

-0.03

10.0

55

-0.07

8*

-0.18

7**

-0.13

2***

-0.22

0***

0.100*

0.027

5-1

9emp

loyees

20-19

9emp

loyees

-0.13

9***

-0.14

5*

0.111**

-0.04

90.0

95*

-0.05

2-0.06

4-0.21

8***

200+em

ployees

-0.24

4***

-0.25

5***

0.112**

0.0

56

0.023

-0.03

3-0.31

1***

-0.35

5***

Mark

etcomp

etition

No

effec

tive c

ompe

tition

Minima

l0.3

41**

*0.5

02**

*0.1

80**

0.1

65**

0.1

73**

0.1

81**

0.4

01**

*0.2

26**

Modera

tetostrong

0.308**

*0.4

23**

*0.1

84**

*0.1

57**

0.2

15**

*0.1

74**

*0.4

67**

*0.3

87**

*Foreignow

nersh

ip

100%

Aus

tralia

n own

ed

Foreignow

nersh

ip>0%

to50

%

0.241**

*0.1

16

0.100

0.079

0.106

0.241**

*0.1

61*

0.174*

Foreignow

nersh

ip>50

%

0.140**

0.1

88**

*0.0

60

-0.03

10.0

99*

0.008

-0.00

4-0.00

3



Tab

le A

4 -

Re

sult

s o

f th

e P

rob

it (

bin

ary

) M

od

el f

or

the

Dif

fere

nt

Typ

es

of

Inn

ova

tio

n w

ith

STE

M/N

on

-STE

M S

kills

as

Ca

teg

oric

al (

co

nti

nu

ed

)

Go

ods a

nd Se

rvice

s Op

erati

onal

Proc

esse

s Or

ganis

ation

al/M

anag

erial

M

arke

ting M

ethod

s Va

riable

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

20

10-11

20

11-12

Collaboration

inR&D

JointR&D

(co-opera

tive)agree

ment

0.398**

*0.3

31**

*0.3

50**

*0.3

27**

*0.3

81**

*0.3

96**

*0.2

52**

*0.2

42**

*

No jo

int R

&D (c

o-op

erati

ve) a

gree

ment

FlexibleW

orkin

gArra

ngem

ent

Ha

vefle

xiblewo

rking

arran

geme

nts

0.295**

*0.1

84**

*0.2

99**

*0.2

41**

*0.3

66**

*0.3

67**

*0.3

18**

*0.2

19**

*

No fle

xible

work

ing ar

rang

emen

t

Go

vernme

ntfinancia

lassista

nce

Receive

dgovern

mentassistan

ce

0.145**

*0.1

20**

*0.1

35**

*0.1

11**

0.1

54**

*0.1

27**

*0.1

48**

*0.0

69

No

t rec

eived

gove

rnme

nt as

sistan

ce

Industry

Manufactur

ing

Ac

como

dationa

ndfoodservice

s-0.17

8**

-0.26

0***

-0.21

7***

-0.46

2***

0.005

-0.27

6***

0.299**

*0.3

07**

*

Adminis

trativea

ndsu

pportserv

ices

-0.37

3***

-0.42

9***

-0.07

5-0.26

4***

0.131*

0.048

0.125

0.162*

Ag

riculture,for

estry

andfi

shing

-0.44

7***

-0.84

7***

0.009

-0.16

6

-0.03

80.0

23

0.009

-0.04

2

Artsandrecrea

tions

ervice

s-0.22

0**

-0.33

4***

-0.18

8**

-0.32

3***

0.021

-0.27

6**

0.330**

*0.4

12**

*

Constru

ction

-0.56

4***

-0.70

0***

-0.20

1**

-0.26

0***

0.074

0.047

-0.31

8***

-0.18

2**

Electricity,wa

ter,gasan

dwasteservice

s-0.49

2***

-0.72

6***

0.134

-0.10

10.1

00

-0.12

7-0.15

0-0.17

7

Financialan

dinsuranceservice

s-0.08

8-0.31

7***

0.274**

*-0.08

90.2

89**

*0.0

52

0.255**

*0.1

00*

He

althc

arean

dsocialassistan

ce

-0.27

6***

-0.32

9***

-0.10

0-0.24

2***

0.116

0.140*

0.054

0.175**

Infor

mation,me

diaan

dtele

comm

unica

tions0

.140

-0.06

8

-0.04

9-0.23

3**

-0.07

3-0.14

2

0.208**

0.1

14

Mining

-0.91

7***

-1.093***

-0.27

8***

-0.35

6***

-0.23

4***

-0.23

2**

-0.79

4***

-0.74

5***

Otherserv

ices

-0.19

2**

-0.33

5***

-0.12

1-0.33

6***

-0.02

50.0

62

0.238**

-0.01

3

Profe

ssional,scien

tifica

ndtechnic

alser

vices

-0.28

0***

-0.40

1***

-0.10

3

-0.31

9***

0.131*

-0.04

40.0

21

-0.06

3

Renta

l,hiringan

drealesta

teservice

s-0.42

7***

-0.59

0***

-0.14

2

-0.23

5**

0.134

0.101

0.119

0.234**

RetailTrad

e0.0

78

-0.09

6

-0.03

2-0.21

5**

0.138*

-0.13

60.4

01**

*0.3

97**

*

Transport,p

ostalan

dware

housing

-0.55

3***

-0.52

5***

0.002

-0.11

30.0

89

-0.07

9-0.23

1**

-0.116

Wholesale

trade

0.112

-0.07

3

-0.00

7-0.12

40.1

23

-0.02

40.2

58**

*0.2

80**

*Nu

mberofobservations(n)

7548

5554

7548

5554

7548

5554

7548

5554

AIC

7987.8

6241.1

8489.7

6432.3

8473.8

6481.3

7935.5

6220.4

LogL

ikelihood

-3957.9

-3084.6

-4208.9

-31

80.1

-4200.9

-32

04.6

-3931.7

-3074.2

No

te: **

*,**an

d*de

notesig

nificancea

tthe1%,5%

and1

0%levels,respectively.


Table A5 - Results of the Ordered Probit Model for Innovation with STEM Skills as Binary

Variable 2010-11 2011-12Intercept 1typeofinnovation -0.533*** -0.050 2typesofinnovation -0.997*** -0.587*** 3typesofinnovation -1.508*** -1.129*** 4typesofinnovation -2.056*** -1.694***STEMskills Not used STEM skills UsedSTEMskills 0.414*** 0.382***Skillsshortagewithinbusinesshamperinginnovation No skills shortage Haveskillsshortage 0.312*** 0.350***Skillsshortagewithinlabourmarkethamperinginnovation No skills shortage Haveskillsshortage 0.174*** 0.214***ICTIntensity ICT_intensity=1 (Most intense) ICT_intensity=2(High) -0.274*** -0.345*** ICT_intensity=3(Moderate) -0.619*** -0.708*** ICT_intensity=4(Low) -0.762*** -1.019***Numberofemployees 1-4employees -0.064 -0.103** 5-19 employees 20-199employees 0.006 -0.118* 200+employees -0.107** -0.163***Marketcompetition No effective competition Minimal 0.304*** 0.298*** Moderatetostrong 0.329*** 0.317***Foreignownership 100% Australian owned Foreignownership>0%to50% 0.177** 0.176** Foreignownership>50% 0.075* 0.054CollaborationinR&D JointR&D(co-operative)agreement 0.380*** 0.366*** No joint R&D (co-operative) agreement FlexibleWorkingArrangement Haveflexibleworkingarrangements 0.369*** 0.290*** No flexible working arrangement Governmentfinancialassistance Receivedgovernmentassistance 0.165*** 0.137*** Not received government assistance Industry Manufacturing Accomodationandfoodservices -0.038 -0.188** Administrativeandsupportservices -0.083 -0.134* Agriculture,forestryandfishing -0.112 -0.273*** Artsandrecreationservices -0.022 -0.115 Construction -0.251*** -0.321*** Electricity,water,gasandwasteservices -0.075 -0.288*** Financialandinsuranceservices 0.188** -0.071 Healthcareandsocialassistance -0.075 -0.068 Information,mediaandtelecommunications 0.047 -0.104 Mining -0.564*** -0.603***



Table A5 - Results of the Ordered Probit Model for Innovation with STEM Skills as Binary (continued)

Variable 2010-11 2011-12 Otherservices -0.020 -0.191* Professional,scientificandtechnicalservices -0.095 -0.250*** Rental,hiringandrealestateservices -0.113 -0.146 RetailTrade 0.160** 0.000 Transport,postalandwarehousing -0.199*** -0.226*** Wholesaletrade 0.141** 0.019Numberofobservations(n) 7548 5554AIC 19589.8 15250.6LogLikelihood -9758.9 -7589.3 Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.

Table A6 - Results of the Ordered Probit Model for Innovation with STEM/Non-STEM Skills as Categorical

Variable 2010-11 2011-12Intercept 1typeofinnovation -0.540*** -0.035 2typesofinnovation -1.005*** -0.572*** 3typesofinnovation -1.517*** -1.115*** 4typesofinnovation -2.067*** -1.680***STEMskills Other Non-STEM skills only STEMandNon-STEMskills 0.457*** 0.396*** STEMskillsonly 0.245*** 0.194** Tradeskillsonly -0.007 -0.197** TradeandotherNon-STEMskillsonly 0.143** 0.094Skillsshortagewithinbusinesshamperinginnovation No skills shortage Haveskillsshortage 0.310*** 0.348***Skillsshortagewithinlabourmarkethamperinginnovation No skills shortage Haveskillsshortage 0.169*** 0.216***ICTIntensity ICT_intensity=1 (Most intense) ICT_intensity=2(High) -0.271*** -0.340*** ICT_intensity=3(Moderate) -0.617*** -0.700*** ICT_intensity=4(Low) -0.760*** -1.014***Numberofemployees 1-4employees -0.062 -0.103** 5-19 employees 20-199employees 0.0005 -0.125** 200+employees -0.122*** -0.177***Marketcompetition No effective competition Minimal 0.301*** 0.294*** Moderatetostrong 0.325*** 0.314***Foreignownership 100% Australian owned Foreignownership>0%to50% 0.175** 0.177** Foreignownership>50% 0.077* 0.051


Table A6 - Results of the Ordered Probit Model for Innovation with STEM/Non-STEM Skills as Categorical (continued)

Variable 2010-11 2011-12CollaborationinR&D JointR&D(co-operative)agreement 0.376*** 0.365*** No joint R&D (co-operative) agreement FlexibleWorkingArrangement Haveflexibleworkingarrangements 0.363*** 0.286*** No flexible working arrangement Governmentfinancialassistance Receivedgovernmentassistance 0.158*** 0.135*** Not received government assistance Industry Manufacturing Accomodationandfoodservices -0.032 -0.192** Administrativeandsupportservices -0.071 -0.132* Agriculture,forestryandfishing -0.117 -0.291*** Artsandrecreationservices -0.015 -0.121 Construction -0.262*** -0.309*** Electricity,water,gasandwasteservices -0.083 -0.295*** Financialandinsuranceservices 0.193** -0.077 Healthcareandsocialassistance -0.055 -0.065 Information,mediaandtelecommunications 0.064 -0.087 Mining -0.570*** -0.610*** Otherservices -0.017 -0.168* Professional,scientificandtechnicalservices -0.074 -0.237*** Rental,hiringandrealestateservices -0.108 -0.152* RetailTrade 0.168** -0.005 Transport,postalandwarehousing -0.200*** -0.235*** Wholesaletrade 0.147** 0.020Numberofobservations(n) 7548 5554AIC 19579.9 15242.5LogLikelihood -9750.9 -7582.2 Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.



Table A7 - Results of the Ordered Probit Model for Innovation Novelty with STEM Skills as Binary, 2010-11

Variable 2010-11Intercept Newtotheindustry -1.107*** NewtoAustralia -1.495*** Newtotheworld -1.991***STEMskills Not used STEM skills UsedSTEMskills 0.208***Numberofemployees 1-4employees 0.018 5-19 employees 20-199employees -0.100 200+employees -0.170**Foreignownership 100% Australian owned Foreignownership>0%to50% 0.133 Foreignownership>50% 0.310***CollaborationinR&D JointR&D(co-operative)agreement 0.181*** No joint R&D (co-operative) agreement ExpenditureonR&D HaveexpenditureonR&D 0.586*** No expenditure on R&D Industry Manufacturing Accomodationandfoodservices -0.356*** Administrativeandsupportservices -0.411*** Agriculture,forestryandfishing -0.337** Artsandrecreationservices -0.152 Construction -0.459*** Electricity,water,gasandwasteservices -0.254 Financialandinsuranceservices -0.287** Healthcareandsocialassistance -0.280** Information,mediaandtelecommunications 0.078 Mining -0.679*** Otherservices -0.264* Professional,scientificandtechnicalservices -0.174 Rental,hiringandrealestateservices -0.240 RetailTrade -0.151 Transport,postalandwarehousing -0.266** Wholesaletrade -0.006Numberofobservations(n) 3554AIC 4189.5LogLikelihood -2067.7 Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.


Table A8 - Results of the Ordered Probit Model for Innovation Novelty with STEM/Non-STEM Skills as Categorical, 2010-11

Variable 2010-11Intercept Newtotheindustry -1.048*** NewtoAustralia -1.437*** Newtotheworld -1.934***STEMskills Other Non-STEM skills only STEMandNon-STEMskills 0.168** STEMskillsonly 0.204* Tradeskillsonly -0.584** TradeandotherNon-STEMskillsonly 0.007Numberofemployees 1-4employees 0.013 5-19 employees 20-199employees -0.105 200+employees -0.173**Foreignownership 100% Australian owned Foreignownership>0%to50% 0.134 Foreignownership>50% 0.305***CollaborationinR&D JointR&D(co-operative)agreement 0.179*** No joint R&D (co-operative) agreement ExpenditureonR&D HaveexpenditureonR&D 0.583*** NoexpenditureonR&D Industry Manufacturing Accomodationandfoodservices -0.373*** Administrativeandsupportservices -0.434*** Agriculture,forestryandfishing -0.358** Artsandrecreationservices -0.179 Construction -0.450*** Electricity,water,gasandwasteservices -0.275 Financialandinsuranceservices -0.308** Healthcareandsocialassistance -0.302** Information,mediaandtelecommunications 0.054 Mining -0.699*** Otherservices -0.235 Professional,scientificandtechnicalservices -0.193* Rental,hiringandrealestateservices -0.267* RetailTrade -0.172 Transport,postalandwarehousing -0.288** Wholesaletrade -0.024Numberofobservations(n) 3554AIC 4188.258LogLikelihood -2064.1289 Note:***,**and*denotesignificanceatthe1%,5%and10%levels,respectively.



References AustralianBureauofStatistics(2001),‘AustralianStandardClassificationofEducation

(ASCED),2001’,ABSCat.No.1272.0,Canberra.Australian Bureau of Statistics (2008), ‘Key Aspects of Innovation in Australian

Businesses:Micro-DataAnalysisofthe2003and2005InnovationSurveys’,ABSSubmissiontotheReviewoftheNationalInnovationSystem,Canberra.

Australian Bureau of Statistics (2013a), ‘Summary of IT Use and Innovation inAustralianBusiness,2011-12’,ABSCat.No.8166.0,Canberra.

Australian Bureau of Statistics (2013b), ‘Selected Characteristics of AustralianBusiness,2011-12’,ABSCat.No.8167.0,Canberra.

AustralianBureauofStatistics(2013c),‘Microdata:BusinessLongitudinalDatabase,Australia2006-07to2010-11’,ABSCat.No.8168.0.55.001,Canberra.

Brunker, D. and Orzechowska-Fischer, E. (2008), ‘An Analysis of Developmentsin Skills Shortages that Hampered Innovation using the 2003 and 2005AustralianInnovationSurveys’,Paperpresentedto theJuly2008ESAMinWellington,ABS,Canberra.

Connolly, G., Trott, D. and Li, Y. (2012), ‘Workplace Agreements and OtherDeterminantsofLabourProductivity’,Paperpresentedtothe23rdAustralianLabourMarketResearchWorkshop,SwanValleyOasisResort,HenleyBrook,Perth,WA,29-30November.

Department of Industry Innovation, Science, Research and Tertiary Education(DIISRTE) (2012), ‘Australian Innovation System Report’, DIISRTE,CommonwealthofAustralia,Canberra.

DepartmentofInnovation,Industry,ScienceandResearch(DIISR)(2009),‘PoweringIdeas: An innovation Agenda for the 21st Century, Commonwealth ofAustralia’,Canberra,ISBN9780642725844.

Departmentof IndustryTourismandResources (DITR) (2006), ‘CollaborationandOtherFactors InfluencingInnovationNovelty inAustralianBusinesses:AnEconometricAnalysis’,DITR,CommonwealthofAustralia,Canberra.

Departmentof IndustryTourismandResources (DITR) (2007a), ‘AspectsofSkillsShortagesandInnovationinAustralianBusinesses:AnAnalysisofthe2003InnovationSurveyData’,DITR,Canberra.

Department of Industry Tourism and Resources (DITR) (2007b), ‘Patterns ofInnovationinAustralianManufacturing2003:AnAnalysisofDatafromthe2003SurveyofInnovationinAustralianBusiness’,DITR,Canberra.

Greene,W.H.(2012),Econometric Analysis,7thedition,UpperSaddleRiver:PrenticeHall.

Karmel,T.(2012),‘SkillsandtheProductivityChallenge’,FeatureArticleinChapter3. Skills and Innovation of the 2012Australian InnovationSystemReport,DIISRTE,Canberra.

Lieponen, A. (2005), ‘Skills and Innovation’, International Journal of Industrial Organisation,2(3).

Marginson, S., Tytler, R., Freeman, B. and Roberts, K. (2013), ‘STEM: Countrycomparisons. Report for the Australian Council of Learned Academies’,http://www.acola.org.au(accessedon6November2013).


OrganizationforEconomicCooperationandDevelopment(2002),Frascati Manual: Proposed Standard Practice for Surveys on Research and Experimental Development,6thedition,OECD,Paris.

OrganizationforEconomicCooperationandDevelopment(2005),The Measurement of Scientific and Technological Activities: Guidelines for Collecting and Interpreting Innovation Data: Oslo Manual, 3rd edition, prepared bythe Working Party of the National Experts on Scientific and TechnologyIndicators,OECD,Paris.

Rotaru,C.(2013),‘DiscreteChoicePanelDataModellingUsingtheABSBusinessLongitudinal Database’, Methodology Research Papers, ABS Cat. No.1352.0.55.139,ABS,Canberra.

Rotaru,C.,Dzhumasheva,S.andSoriano,F.(2013),‘PropensityScoreMatching:AnApplication using theABSBusinessCharacteristics Survey’,Methodology Research Papers,ABSCat.No.1352.0.55.128,ABS,Canberra.

Stanwick,J.(2011),‘Innovation:ItsLinkswithProductivityandSkillDevelopment–Ataglance’,NationalCentreforVocationalEducationResearch,DepartmentofEducation,EmploymentandWorkplaceRelation,Adelaide.

Tether,B.,Mina,A.,Consoli,D.andGagliardiD. (2005), ‘ALiteratureReviewonSkills and Innovation. How Does Successful Innovation Impact on theDemandforSkillsandHowDoSkillsDriveInnovation?’,ACRICReportfortheDepartmentofTradeandIndustry,Manchester.

Todhunter,J.andAbello,R.(2011),‘BusinessInnovationandtheUseofInformationandCommunicationsTechnology’,Methodology Research Papers,ABSCat.No.1351.0.55.033,ABS,Canberra.

Van Zon, A. (2001), ‘A Simple Endogenous Growth Model with AsymmetricEmployment Opportunities by Skills’, Maastricht Economic ResearchInstitute on Innovation and Technology (MERIT) – Infonomics ResearchMemorandumseries2001-09,Netherlands.

Wong,M.,Page,D.,Abello,R. andPang,K.P. (2007), ‘Explorations of Innovationand Business Performances using Linked Firm-Level Data’, Methodology Research Papers,ABSCat.No.1351.0.55.020,ABS,Canberra.

Wooldridge,J.M.(2010),Econometric Analysis of Cross Section and Panel Data,2ndedition,MITPress,Cambridge.

Documents

Modelling the Relationships Between the ... - Business & Law€¦ · Characteristics Survey (BCS) are used in the analysis. The study applies probit regression modelling on the 2010-11