Improving the quality of job designImproving the quality of job design

David HolmanDavid HolmanInstitute of Work PsychologyInstitute of Work Psychology

University of SheffieldUniversity of Sheffield

Central Questions in Job Design ResearchCentral Questions in Job Design Research

What is the nature of job design?What is the nature of job design?Job characteristics theory Job characteristics theory –– HackmanHackman & Oldham& OldhamDemandsDemands--Resources models Resources models –– KarasekKarasek & & TheorellTheorell; Bakker; Bakker

How do job designs arise? How do job designs arise? Strategic and institutional influences on job designStrategic and institutional influences on job designSocioSocio--technical theory/Social shaping of technologytechnical theory/Social shaping of technologyJob craftingJob crafting

What is the geography of job design?What is the geography of job design?Where are the high quality jobs?Where are the high quality jobs?Who has them?Who has them?

Central Questions in Job Design ResearchCentral Questions in Job Design Research

How can we intervene to redesign jobs and improve job How can we intervene to redesign jobs and improve job quality?quality?

Efforts to improve job quality must be multiEfforts to improve job quality must be multi--faceted faceted Institutional and organisational effortInstitutional and organisational effort

Study 1: Do institutions have an effect on job design?Study 1: Do institutions have an effect on job design?

Study 2: Participative job redesign interventionStudy 2: Participative job redesign intervention

Study 1: Institutional Effects on Job DesignStudy 1: Institutional Effects on Job Design

Varieties of CapitalismVarieties of Capitalism (Hall & Soskice, 2001)(Hall & Soskice, 2001)

Coordinated economiesCoordinated economies (e.g., Germany, Sweden, Denmark)(e.g., Germany, Sweden, Denmark)•• Collaborative and strategic relationshipsCollaborative and strategic relationships•• TU and employer association coordinate wages and trainingTU and employer association coordinate wages and training•• Strong labour market regulations/employment protectionStrong labour market regulations/employment protection•• Influential trade unionsInfluential trade unions

Liberal economiesLiberal economies (e.g., US, UK, Canada)(e.g., US, UK, Canada)•• Coordination through competitive markets and formal contractsCoordination through competitive markets and formal contracts•• Relatively weak trade unions and employers associationsRelatively weak trade unions and employers associations•• Limited employment protectionLimited employment protection

Industrialising economiesIndustrialising economies (E.g., India, Brazil)(E.g., India, Brazil)•• Pressures to liberalise more intense (Frenkel, 2006)Pressures to liberalise more intense (Frenkel, 2006)

Three mechanismsThree mechanisms•• Skills and TrainingSkills and Training mechanism mechanism –– job skills and job complexity job skills and job complexity •• Trade UnionTrade Union mechanism mechanism –– effectiveness in standard setting, resisting effectiveness in standard setting, resisting

deleterious practicesdeleterious practices•• CulturalCultural mechanism mechanism –– skill governed skill governed vsvs rule governed employment rule governed employment

logics used in decisions (logics used in decisions (FligsteinFligstein & & ByrkejflotByrkejflot, 1996, 1996))

Hypothesis Hypothesis •• Coordinated economies will have better job design than liberal aCoordinated economies will have better job design than liberal and nd

industrialising economiesindustrialising economies

MethodMethod

National workplace surveyNational workplace survey•• 17 countries17 countries•• Liberal: USA, UK, Ireland, Canada, Israel, S. KoreaLiberal: USA, UK, Ireland, Canada, Israel, S. Korea•• Coordinated: Denmark, Germany, Austria, France, Spain, NetherlanCoordinated: Denmark, Germany, Austria, France, Spain, Netherlands, ds,

SwedenSweden•• Industrializing: India, Brazil, Poland, South AfricaIndustrializing: India, Brazil, Poland, South Africa

•• 2,477 call centres (sample sizes, 432,477 call centres (sample sizes, 43--437, M=145)437, M=145)•• 474,941 employees474,941 employees

•• EstablishmentEstablishment--levellevel•• Senior managerSenior manager•• Telephone, post, interview, internetTelephone, post, interview, internet

MethodMethodMultiMulti--level analysislevel analysis•• Organisations levelOrganisations level--11•• Countries levelCountries level--22

LevelLevel--1 measures1 measures•• Work designWork design -- Job discretion & Performance monitoringJob discretion & Performance monitoring

LevelLevel--2 measures2 measures•• Type of economyType of economy –– liberal, coordinated, industrialisingliberal, coordinated, industrialising

Coordinated economies have lower job discretion and high performCoordinated economies have lower job discretion and high performance ance monitoring than liberal economiesmonitoring than liberal economiesIndustrializing economies have lower job discretion than coordinIndustrializing economies have lower job discretion than coordinated, and ated, and higher performance monitoring than coordinated and liberal econohigher performance monitoring than coordinated and liberal economies mies

Job DiscretionJob Discretion(1(1--5)5)

Performance MonitoringPerformance Monitoring(1(1--7)7)

OverallOverall 2.82.8(Low(Low--Moderate)Moderate)

5.15.1(Fortnightly)(Fortnightly)

CoordinatedCoordinated 3.03.0 4.34.3

LiberalLiberal 2.72.7 5.35.3

IndustrializingIndustrializing 2.72.7 6.26.2

Total variance explainedTotal variance explained•• 2% job discretion2% job discretion•• 10% performance monitoring10% performance monitoring•• Institutional factors explain more of total variance in performaInstitutional factors explain more of total variance in performance nce

monitoring than strategic factors (2%)monitoring than strategic factors (2%)

Job quality Job quality –– high discretion and low monitoringhigh discretion and low monitoring

Proportion of OrganizationsProportion of Organizations Proportion of CSRsProportion of CSRs

Level ofLevel ofJob DiscretionJob Discretion

Level of Performance Level of Performance MonitoringMonitoring

Level of Performance Level of Performance MonitoringMonitoring

More than More than fortnightlyfortnightly

Fortnightly Fortnightly --DailyDaily

More than More than fortnightlyfortnightly

FortnightlyFortnightly-- DailyDaily

None/Very littleNone/Very little 1010 2626 55 5858

ModerateModerate 1414 2424 66 2121

A lot/a great dealA lot/a great deal 1111 1111 22 77

ConclusionConclusion

•• Decisions about job design influenced by institutional contextDecisions about job design influenced by institutional context•• Institutions matter Institutions matter –– despite increased globalisationdespite increased globalisation•• Job quality is low in call centresJob quality is low in call centres

•• Further research on mechanisms through which institutions have tFurther research on mechanisms through which institutions have their heir effectseffects

Study 2: Study 2: Job Redesign Intervention:Job Redesign Intervention:

A serendipitous quasiA serendipitous quasi--experimentexperiment

Mediational assumption of job redesign interventionsMediational assumption of job redesign interventions•• Intervention Intervention –– job characteristics job characteristics –– outcomeoutcome•• Intervention Intervention –– job control job control –– wellwell--being (Bond & Bunce, 2001)being (Bond & Bunce, 2001)

Interventions seek multiple changes to job characteristicsInterventions seek multiple changes to job characteristics•• Are multiple changes necessary if change in outcome can be achieAre multiple changes necessary if change in outcome can be achieved ved

by job control alone?by job control alone?

Job Redesign InterventionJob Redesign Intervention

SettingSetting•• Dept of Health Insurance companyDept of Health Insurance company•• 240 employees in 5 sections240 employees in 5 sections•• Administrative work (except call centre section)Administrative work (except call centre section)•• High turnover (23%) and low job satisfactionHigh turnover (23%) and low job satisfaction

Intervention: Scenarios Planning ToolIntervention: Scenarios Planning Tool•• ParticipativeParticipative•• Informative Informative –– learn about job design theorylearn about job design theory•• Envisaging different scenariosEnvisaging different scenarios

1. Assessment and redesign 1. Assessment and redesign –– one day work shopone day work shop•• Each team identifies core tasks and obstaclesEach team identifies core tasks and obstacles•• Current job design scenario rated against job characteristics, iCurrent job design scenario rated against job characteristics, i.e., job .e., job

control, feedback, skill utilization, participation, obstaclescontrol, feedback, skill utilization, participation, obstacles•• And performance and wellAnd performance and well--beingbeing•• Envisage three new scenarios Envisage three new scenarios –– including high performance and wellincluding high performance and well--

being scenariobeing scenario•• Ideas to achieve this new scenarioIdeas to achieve this new scenario

2. Implementation2. Implementation

Teams tasked with implementing changesTeams tasked with implementing changes•• Job control and participationJob control and participation, e.g., adoption of new tasks, training on , e.g., adoption of new tasks, training on

new tasks, access to new parts of IT systems, involvement in desnew tasks, access to new parts of IT systems, involvement in design of ign of a new IT system, team member control of the rota; a new IT system, team member control of the rota;

•• Skill utilizationSkill utilization, e.g., training on new tasks, adoption of new tasks; , e.g., training on new tasks, adoption of new tasks; •• FeedbackFeedback, e.g., clearer specification of performance criteria, more , e.g., clearer specification of performance criteria, more

regular performance feedback; regular performance feedback; •• Removal of task obstaclesRemoval of task obstacles, e.g., visits to teams in other departments to , e.g., visits to teams in other departments to

exchange understanding on job roles, updates to equipment, and exchange understanding on job roles, updates to equipment, and changes to procedures.changes to procedures.

Two team representatives monitor implementationTwo team representatives monitor implementationMeet once/month for 3 monthsMeet once/month for 3 months

Research DesignResearch Design

Stage 1Stage 1 BriefingBriefing Month 1Month 1Stage 2Stage 2 PrePre--questionnairequestionnaire Month 2Month 2Stage 3Stage 3 InterventionIntervention Month 3Month 3Stage 4Stage 4 ImplementationImplementation Months 3Months 3--77Stage 5Stage 5 PostPost--questionnairequestionnaire Month 9Month 9

Research DesignResearch Design

•• Stage 1Stage 1 BriefingBriefing Month 1Month 1•• Stage 2Stage 2 PrePre--questionnairequestionnaire Month 2Month 2•• Stage 3Stage 3 InterventionIntervention Month 3Month 3•• Stage 4Stage 4 ImplementationImplementation Months 3Months 3--77•• Stage 5Stage 5 PostPost--questionnairequestionnaire Month 9Month 9

A serendipitous eventA serendipitous event•• Outsourcing initiative in 2 sectionsOutsourcing initiative in 2 sections•• Removed key tasksRemoved key tasks•• Compromised interventionCompromised intervention

•• Two Groups Two Groups –– Experimental group & NonExperimental group & Non--Equivalent control groupEquivalent control group•• Advantage over traditional control groupsAdvantage over traditional control groups

Sample Sample •• Longitudinal sample n = 119Longitudinal sample n = 119•• 71 Intervention group, 48 Control group71 Intervention group, 48 Control group

MeasuresMeasures

•• WellWell--beingbeing•• Job controlJob control•• FeedbackFeedback•• Skill utilizationSkill utilization•• ParticipationParticipation•• Job controlJob control

Multilevel analysisMultilevel analysis•• Occasions (levelOccasions (level--1) nested within individuals (level1) nested within individuals (level--2)2)

ResultsResultsMeans Means Longitudinal Sample

InterventionMean

Control Mean

Well-Being 1 3.23 3.35

Well-Being 2 3.41* 3.31

Job Control 1 3.24 3.45

Job Control 2 3.51* 3.50

Participation 1 2.34 2.30

Participation 2 2.54* 2.13

Skill Utilization 1 4.76 5.09

Skill Utilization 2 4.96* 4.83

Feedback 1 4.60 4.98

Feedback 2 5.34* 4.78

Task obstacles 1 2.66 2.69

Task obstacles 2 2.58* 2.57

Direct effect of intervention Direct effect of intervention •• Intervention X measurement occasionIntervention X measurement occasion•• Direct effects on wellDirect effects on well--being being •• Direct effects on all job characteristics except job obstaclesDirect effects on all job characteristics except job obstacles

Mediated effectsMediated effects

•• Job characteristics (except obstacles) mediate relationship betwJob characteristics (except obstacles) mediate relationship between een intervention and wellintervention and well--beingbeing

ConclusionConclusion•• Utility of participative approaches to job redesignUtility of participative approaches to job redesign•• Utility of focusing on multiple changesUtility of focusing on multiple changes

•• Job characteristics are a mediating mechanismJob characteristics are a mediating mechanism•• Other mechanisms? Psychological contract, active learningOther mechanisms? Psychological contract, active learning

•• Limitations to job redesignLimitations to job redesignWithin one departmentWithin one departmentTechnologyTechnologyOutsourcing initiativeOutsourcing initiative

Future directions for job design researchFuture directions for job design research

MultiMulti--level theories of job designlevel theories of job design

Group/OrganisationGroup/Organisation

JobJob

Event/enactedEvent/enacted

•• How are job characteristics enacted and created over time?How are job characteristics enacted and created over time?

Managerial job designManagerial job design

Mechanisms of job designMechanisms of job design