Short- and long-term consequencesof age in work teams

An empirical exploration of ageing teams

Franz J. GellertHanze University of Applied Sciences, Groningen, The Netherlands, and

Ben S. KuipersErasmus University of Rotterdam, Rotterdam, The Netherlands

Abstract

Purpose – The purpose of this paper is to explore the effects of age in work teams on short-term teamconsequences, such as satisfaction, involvement, mutual learning, decision making and feedback, andlong-term team consequences, such as quality, sick leave and burnout, and to consider theirimplications for team management and human resource management (HRM) policies in team-basedorganizations facing an ageing work force.

Design/methodology/approach – The study elaborates on the framework of Milliken and Martins,further examining the effects of both average age and age differences. The authors collected objectivedata as well as data through questionnaires among 150 work teams with more than 1,500 white-collarand blue-collar workers from an automotive company in Sweden. With these data the authorsconducted correlation and step-by-step hierarchical regression analyses.

Findings – The analyses showed significant positive effects of average age on both short-term andlong-term consequences. No significant effects of age differences were found.

Research limitations/implications – Conducting a longitudinal study in an automotive companyin Sweden resulted in monocultural findings. The use of a sample from one organization may limit thegeneralization of our findings. Future research should pay more attention to effects of age in teams,compared to individual age effects in organizations and to explore more advanced models that help tounderstand the dynamic processes of age in teams.

Practical implications – The results have implications for management of teams and HRM policyin organizations relating to recruitment, early retirement, training developments and teamcomposition in general.

Originality/value – The paper suggests positive effects of age in work teams and contributes to theliterature about the ageing workforce working in teams.

Keywords Older workers, Team working, Sweden

Paper type Research paper

Work teams are used in many organizations to conduct complex tasks and to improvecompany performance (O’Connell et al., 2002). More elder workers will join work teamsin the future and most descriptions of work-team demographics emphasize the fact thatthe average age of a work team is increasing (Jackson et al., 1995). The importance ofknowledge about the increasing age in work teams is supported by, for example,Schalk et al. (2007) who stated in a position paper: “with respect to the different levelson which performance can be assessed, the team level deserves specific mentioning inrelation to the age factor. Not only is the average age important here, but also the agecomposition of teams.”

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1362-0436.htm

CDI13,2

132

Career Development InternationalVol. 13 No. 2, 2008pp. 132-149q Emerald Group Publishing Limited1362-0436DOI 10.1108/13620430810860549

Organizations envisage difficulties in managing work teams with an increasingnumber of elder employees because elder workers are often stereotyped as lessflexible, less adaptable to new technology, and with less motivation for learning,training and well-being (De Lange et al., 2006). Although the number of researchstudies about age and its effects on team processes and team performance (Sternsand Miklos, 1995; Rhodes, 1983; Waldman and Avolio, 1986; Elfenbein and O’Reilly,2005) has increased in the past, most of these studies focus on age as an easilydetectable attribute of individuals and is researched in combination with otherindividual demographic attributes (O’Reilly et al., 1999; Sterns and Miklos, 1995;Milliken and Martins, 1996; Jackson et al., 2003; Tsui and O’Reilly, 1989). Theresearch that has been published on age effects fails to have a look at the directeffects of average age and age differences on both processes and outcomes in workteams at the team level.

There are several reasons to research such direct effects in work teams:. First, if we turn from age effects at the individual level to a study of age effects in

work teams, little knowledge is available about the dynamics and performanceon the work team level.

. Second, there have been calls to expand research on age effects in work teams, asvery little empirical evidence exists for age effects at the team level as opposed tothe effects on individual level (Milliken and Martins, 1996; Jackson et al., 1995;Horwitz, 2005; Schalk et al., 2007).

. Third, because of the increasing age in work teams, poor knowledge about thedirect effects of age on team processes and team performance may be reducingthe ability of those responsible for team development to take care of elderworkers.

Derived from these reasons, the central question we want to address is: How doaverage age and age differences in work teams directly affect both the team’s processesand performance?

We conducted our study to explore these effects and to increase our knowledgeof the topic using data from a longitudinal study in an automotive company inSweden, involving employees from a large variety of work teams (see Kuipers,2005). Moreover, we have further elaborated a model developed by Milliken andMartins (1996) and expanded their theory about age effects. We think our studywill not only contribute useful findings to theory in social behavior andorganizational behavior, but also will contribute to team management and HRMdealing with ageing workforces in various industries by providing them with newinsights.

First we will describe and define team characteristics related to age and workteams, followed by considering the theory and elaborating the model with factorsof both team processes and team performance in their relationships to age. Ourmethod and measures are described in the next chapter, followed by the resultsand findings revealing the effects of average age and age differences on workteam level. We will finish our paper with practical implications for teammanagement and HRM policies in companies and with future directions for thescientific field as well.

Consequences ofage in work

teams

133

Age and work teamsAgeAge or ageing can be portrayed as a multi-dimensional process that is not easy tocapture within one single definition and includes changes in functioning over time(Schalk et al., 2007). In the literature we find definitions of age as chronological age,functional age, psychological age, organizational age and life span age (De Lange et al.,2006; Sterns and Miklos, 1995; Sterns and Doverspike, 1989). Schalk et al. (2007) statethat functional age is a medical construct in which the actual number of years livedrather than people’s state of health is considered as the determinant for beingconsidered of a certain age. In this paper we follow the definition of the functionalapproach.

Work teamsPast research on work teams reveals various definitions of these teams. For example,Sundstrom et al. (2000) made a distinction between “production groups”, consisting offront line employees who repeatedly produce a tangible output, and “service groups”who cooperate to conduct repeated transactions with customers (see also Kuipers,2005). A more general definition of work teams can be found in Cohen and Baily (1997):“work teams are units that perform the day-to-day functions of organizations. Hybridforms of work teams such as self managing, semi-autonomous, and empowered workteams are frequently employed in organizations”. Along with the study of work teams,there has been an increase in the scholarly attention paid to the impact of teamheterogeneity and homogeneity on team processes and outcomes (Horwitz, 2005;Milliken and Martins, 1996; Williams and O’Reilly, 1998). We define work teams in ourpaper as teams who perform the day-to-day functions and self manage their tasks to ahigh degree.

Theories regarding the relationship between age and teamworkRather than presenting a comprehensive review of all relevant literature, we highlightsome exemplary studies in our paper to address the relationship between age andteamwork processes and performance. A research done by Remery et al. (2003) among1,000 organizations in the Dutch labor market concluded that ageing work teams arenot a topic of special interest. Considering an ageing workforce, and the increasingattention of HRM on such relevant issues, it is necessary to gain a greaterunderstanding of how age affects both work team processes and performance inteam-based organizations. Burke and Ng (2006) found that the role of HRM inorganizations has to be changed regarding the upcoming challenges created by ageingemployees and consequently HRM has to develop policies to deal adequately with elderemployees.

We can summarize the theories about work teams and their relationship to age byusing two main streams of thinking or concepts (Horwitz, 2005).

First, the similarity-attraction paradigm whose researchers, partial to this theory,found that team members’ similarity in terms of age leads to positive aspects such asopen communication, understanding, feedback, more productivity than heterogeneousteams and therefore influences performance positively (Wiersema and Bantel, 1992;Rhodes, 1983). On the other hand, differences in age tend to be negatively associatedwith team performance and social integration (Jackson et al., 1995; Milliken and

CDI13,2

134

Martins, 1996). Second, there is the cognitive resource diversity theory in whichdissimilarities in terms of age among team members may lead to a wider range ofperspectives and experiences that improve team decision quality and teamperformance (Cox and Blake, 1991). This also leads to negative aspects such as thewithholding of individual knowledge, the tendency of members to leave the team or arise in conflicts (Milliken and Martins, 1996; Pelled, 1996; Pelled et al., 1999). Accordingto Horwitz (2005) all research findings yielded mixed results and failed to distinguishthe direct effects of age on both work team processes and performance.

The model of Milliken and Martins (1996)The model developed by Milliken and Martins (1996) will support us to extend thetheory about direct age effects on teamwork processes and performance. Theyconsider in their model two different categories of diversity: observable attributessuch as age and underlying attributes such as skills and knowledge. The outcomeof these types of diversity leads to two categories of consequences: short-term andlong-term consequences. The observable attributes such as age are directly relatedto the short-term and long-term consequences. Milliken and Martins (1996) dividedshort-term consequences into four categories: affective (e.g. satisfaction), cognitive(e.g. innovation), symbolic (behavior) and communication-related consequences.Thus, it follows that diversity attributes and short-term consequences have effectson “long-term consequences”, referring to individual, group and organizationaloutcomes, such as performance. We purposely use this model to specifically studythe effects of age in teams, focusing both on average age and age differencesfollowing Schalk et al. (2007). Furthermore, the model serves as a conceptualframework to extend the theory about direct age effects by making some alterationsof terms and definitions. This is an elaboration of the model since Milliken andMartins (1996) look at diversity effects of age only. We fill in their concepts ofshort-term and long-term consequences based on the variables they suggest whileleaving out symbolic consequences since Milliken and Martins (1996) only referredto race and gender in this case (see the model in Figure 1).

Figure 1.Research model

Consequences ofage in work

teams

135

Elaboration of the research modelFirst, we start with elaborating the relationship between age and variables ofshort-term consequences followed by the relationships between age and variables oflong-term consequences. We refer to age both in terms of average age and agedifferences.

Short term: affective consequencesMilliken and Martins (1996) refer to affective consequences by identification andsatisfaction at both the “individual and group levels of analysis”. The concept ofidentification (Milliken and Martins, 1996) we define by involvement that is the level ofcommitment at team level to belonging to the organization (see Allen and Meyer, 1996).

Job satisfaction as a short-term consequence (Milliken and Martins, 1996) is definedas the level of work-related enjoyment and enthusiasm within the work team (Le Blanc,1994). Elder workers tend to be less satisfied because of gaining more and more routinein their work over the years. For example, an Australian quality of working life surveyconducted by Considine and Callus (2001) revealed that fulltime workers’dissatisfaction increased with age. The most significant finding was that thedissatisfaction was directed at the amount of work people had to accomplish and thelevel of routine in their work. No evidence was found for dissatisfaction related toenjoyment of the type of work in general or enthusiasm for performing these tasks.According to Sterns and Miklos (1995) recent research about job satisfaction revealedthat for older workers job satisfaction is consistently linked with more control overtheir jobs and that they are less interested in working harder to earn promotions, salaryincreases or other rewards. With regard to De Lange et al. (2006) research studiesconsistently yield mixed results referring to the relationship of age effects and jobsatisfaction, therefore we need to explore this relationship further, considering bothaverage age and age differences in work teams.

Short term: cognitive consequencesIn the model of Milliken and Martins (1996), cognitive consequences are explained interms of decision making and concepts such as innovation and creativity which werefer to as mutual learning.

We define mutual learning as the ability of team members to share information witheach other in order to increase the entire work team performance. Cognitiveconsequences rely on stimuli that are created by team members. Accordingly, Warr(2001), who defined learning as cognitive ability, supports this. In this context, mutuallearning helps to improve the abilities of team members, thus increasing their teamperformance. Older members in work teams may contribute to an increase in the levelof experience and knowledge (expertise) among their team members and consequentlyincrease the team performance commensurately. These effects only take place ifexperienced team members are open to communicate, to share knowledge and tocooperate which leads to intellectual capital. Peterson and Spiker (2005), for example,see the experience of elder people as a contributory value in work teams if theirexperience is utilized to build up intellectual capital and to support younger peers tofunction better in teams.

We define decision making as joint performance of managerial tasks (see Kuipers,2005). Decision making is related to team autonomy that is a key mechanism in team

CDI13,2

136

functioning that gives a team the freedom to decide (Molleman, 2005). According toMilliken and Martins (1996), the decision-making process is embraced by perceptionand interpretation of stimuli and the ability of the work team to process information.For example, Jackson et al. (1995) consistently highlights that decision making is aninformation intensive activity which requires an adequate atmosphere within the team.Research on decision-making processes in management teams with older employeesrevealed that older managers made slower decisions but took greater advantage ofavailable information by considering more aspects of the problem (Sterns and Miklos,1995). Further exploration of the relationships between age and cognitiveconsequences in work teams will be required to gain more understanding into theeffects of both average age and age differences.

Short term: communication-related consequencesMilliken and Martins (1996) refer to communication-related consequences as thebehavior of team members resulting from communication internally – coming fromwithin the team, and externally – coming from outside the team. With our study wefocus on feedback inside teams as an important communication-related consequence.

According to Kuipers (2005), feedback is part of a team’s internal relationship andits team members’ team-oriented attitude. Feedback is what team members address toeach other in case of, for example, mistakes in the task performance, how they evaluateeach other and motivate each other. Although Milliken and Martins (1996) report aboutthe relationship between age diversity and feedback from inside and outside the team,little research is available about the effects of both average age and age differences onfeedback of team members within the same team. We will further explore theserelationships with regards to this internal team process.

Long-term consequences on group levelQuality is, with regard to Suzaki (1993), what continuously satisfies customers(internally and externally) and needs to be clearly defined in day-to-day business.Team members are constantly requested to initiate and support product improvementsin order to achieve better quality products and services (e.g. Katzenbach and Smith,1993; Hut and Molleman, 1998; Dunphy and Bryant, 1996). For example, Sterns andMiklos (1995) found that although speed and efficiency of processing might be affectedby age, workers’ stability in their jobs and work environment lead to good qualityoutput and a low error rate. Although these latter aspects relate more to tenure thanage itself, we may expect that increased age implies more (working) experience whichis useful to increase quality and decrease error rates of the entire work team. Our studywill further need to explore the exact relationships with regard to age and quality, aswell as if age differences in work teams will help to improve quality.

Sick leave is generally considered as to be an absence from work due to illness. Theprejudices and stereotypes, that older people take more sick leave than youngeremployees do, still exist. Cummings et al. (1993) found that if the composition of teamsvaries in age, there tends to be a higher rate of absence than is shown by people inthose teams with a more similar age composition. Recently Peterson and Spiker (2005),for example, indicated evidence that older workers are less absent due to illness thantheir younger colleagues.

Consequences ofage in work

teams

137

Burnout, not to be confused with stress, is defined in our paper as work-relatedemotional exhaustion (Kuipers, 2005). When a team is de-motivated by its work andtired of its work, then this is called team burnout. De Lange et al. (2006) stated in herresearch study that burnout of elder workers is lower compared to younger employees.It is more often observed that younger employees report burnout at the beginning oftheir careers that may lead to the conclusion that experience might be an importantbuffer in the development of burnout (De Lange et al., 2006).

All relationships between average age and age differences and all of these long-termconsequences will need further exploration on team level.

MethodParticipantsWe tested our model with data from a longitudinal study carried out by the secondauthor (Kuipers, 2005) at an automotive plant in Sweden. In this plant about 2,200employees in more than 160 teams work in five production departments and variousother departments ranging from logistics up to engineering. The blue-collar productionteams represented about 75 percent of those researched, whereas the supporting teams(pre-dominantly white-collars, such as for engineering and office functions)represented the other 25 percent. Approximately 83 percent of workers at thecompany are male and 10 percent female (7.5 percent relate to missing data) that is afairly normal division of gender in the Swedish automotive industry (cf. Berggren,1993). The average number of members in a team is about 11. With regard to age wecan see that the entrance age of people working in the company is at about 20 years.The age scale ranges from 20 up to 63 years. The mean age is 36 years with a standarddeviation of 7.3. About 60 percent of workers are less than 36 years old and 40 percentare above 36.

Data on age of team members, team processes and team performance were collectedamong more than 150 blue-collar and white-collar work teams with more than 2,000individuals. The main source of data used was based on questionnaires filled out byemployees in five production departments and five supporting departments (responserate more than 70 percent). Team managers or department managers handed out thequestionnaires to their teams during team meetings. The questionnaires containeditems for all aspects. We have specifically selected items that reflect actions andbehavior of teams and not of individual team members.

MeasuresAverage age and age differences (standard deviation) were calculated for all teams onteam level, by aggregating individual age data to the team level. Within teamprocesses, the topics mutual learning consisted of nine items, decision making of fiveitems, and feedback of six items (see the Appendix for item overview). The topicssatisfaction, involvement and burnout consisted each of two items (see the Appendixfor item overview). For all these items a five-point Likert was used and the items wereaveraged and aggregated to team level.

Both sick leave and product quality were measured by the company on team level.Product quality was calculated by the average percentage of zero-defect productsdelivered by teams during a four-week period, and is only available for productionteams. Zero-defect stands for products produced by a team without any errors or faults.

CDI13,2

138

Sick leave was calculated by the average number of sick occasions in a team during aone-year period to avoid seasonal influences. Data on sick occasions were available fora limited number of both production and support teams.

AnalysesData collected were aggregated to team level, where necessary, and analyzed byhierarchical multiple regression models in SPSS (Field, 2000; Miles and Shevlin, 2001).We used this type of model because it is “a technique for modeling the relationships oftwo or more variables” (Miles and Shevlin, 2001). One by one we tested the models forall short-term (affective, cognitive and communication-related) and long-term (grouplevel) consequences as dependent variables. We built the regression models step-wiseby entering team size as a control variable in the first step, followed by both averageage and age differences as independent variables in step two. Furthermore, thereliability of our scales was tested by calculating Cronbach a. All alphas appeared tobe sufficient (all above 0.70; see also Table I).

ResultsThe means, standard deviations, alphas and correlations are reported in Table I. Thecontrol variable team size shows to have a significant negative relationship to almostall variables (except burnout, where a positive significant relationship was found).Both average age as well as age differences show strong and significant relationshipswith team short-term and long-term consequences.

Short-term consequencesThe results of the analyses of relationships between age and short -term consequencesare presented in Table II. A strong significant positive relationship was found betweenage and affective consequences. About 22 percent of the variance in involvement couldbe explained by average age in teams, whereas 30 percent of the variance insatisfaction could be explained by this age variable. Also between age and cognitiveconsequences positive significant relationships were found. With about 24 percent, aconsiderable proportion of the variance in mutual learning was explained by averageteam age. And although only a small percentage of the variance (3 percent) indecision-making was explained by average age, nevertheless the relationship is stillpositive significant, as the beta shows (b ¼ 0:216 with p , 0.05). Thecommunication-related consequence of feedback seems to be positively affected byaverage age as well. A small but significant and positive relationship was foundbetween both variables. For all short-term effects higher average age appears to havesignificant positive effects. However, the data showed no significant relationshipsbetween age differences (SD of age in teams) and any of the short-term consequences.

Long-term consequences on group levelThe results presented in Table III indicate strong and positive significant relationsbetween average age in teams and long-term consequences. In teams with higheraverage age, lower figures for sick occasions and burnout are reported. About 23 percentof the variance in burnout is explained by its relationship to average age, whereasaverage age in teams seems to explain more than 33 percent of the sick occasions. Thenumber of sick occasions in a team clearly benefits from a higher average age

Consequences ofage in work

teams

139

Var

iab

leM

ean

SD

Cro

nb

ach

a1

23

45

67

89

10

1G

rou

psi

ze10

.58

4.44

NA

–2

Av

erag

eag

e36

.24

7.30

NA

20.

275

**

–3

Ag

ed

iffe

ren

ces

7.30

3.05

NA

20.

023

0.37

5*

*–

4M

utu

alle

arn

ing

3.21

0.44

0.84

20.

349

**

0.56

7*

*0.

561

**

–5

Fee

db

ack

3.65

0.38

0.70

20.

185

*2

0.22

6*

*2

0.16

2*

20.

150

–6

Dec

isio

n-m

akin

gp

roce

sses

3.90

0.37

0.76

20.

337

**

0.23

5*

*0.

255

**

0.01

70.

606

**

–7

Pro

du

ctq

ual

ity

96.3

33.

54N

A2

0.29

2*

0.62

8*

*0.

496

**

0.16

00.

442

**

20.

017

–8

Sic

koc

casi

ons

2.75

1.43

NA

20.

016

20.

634

**

20.

562

*2

0.14

92

0.57

0*

*0.

233

*2

0.12

8–

9S

atis

fact

ion

3.07

0.61

0.75

20.

250

**

0.56

2*

*0.

599

**

0.34

5*

*0.

695

**

0.00

80.

461

**

0.15

7–

10In

vol

vem

ent

3.73

0.47

0.83

20.

264

**

0.43

9*

*0.

528

**

0.22

5*

*0.

616

**

0.22

5*

*0.

546

**

0.09

32

0.49

0*

*–

11B

urn

out

2.74

0.45

0.70

0.21

3*

*2

0.46

5*

*2

0.51

4*

*2

0.35

9*

*2

0.49

2*

*0.

054

20.

331

**

20.

177

0.54

9*

*2

0.72

9*

*

Notes:n¼

147

pro

du

ctio

nan

dsu

pp

ort

team

s,ex

cep

tfo

rp

rod

uct

qu

alit

y(n

¼66

)an

dsi

ckoc

casi

ons

(n¼

89);

* p,

0.05

;*

* p,

0.01

Table I.Means, standarddeviation, Cronbach aand correlations

CDI13,2

140

Sat

isfa

ctio

nIn

vol

vem

ent

Mu

tual

lear

nin

gD

ecis

ion

-mak

ing

Fee

db

ack

Var

iab

leb

DR

2F

bDR

2F

bDR

2F

bDR

2F

bDR

2F

Ste

p1

Con

trol

Gro

up

size

20.

211

**

0.06

510

,206

20.

133

*0.

073

11,5

672

0.21

10.

122

20,4

182

0.28

5*

*0.

117

19,5

132

0.23

9*

*0.

034

5,12

0S

tep

2In

dep

end

ent

Av

erag

eag

e0.

519

**

0.30

542

,554

0.49

8*

*0.

224

30,6

680.

500

**

0.23

540

,574

0.21

6*

*0.

028

12,3

840.

197

**

0.04

96,

564

Ag

ed

iffe

ren

ces

0.06

10.

007

29,0

582

0.01

10.

000

20,3

130.

010

0.00

026

,873

20.

089

0.00

68,

615

20.

068

0.00

44,

559

Tot

alR

20.

377

0.29

70.

357

0.15

10.

086

Ad

just

edR

20.

364

0.28

30.

344

0.13

40.

067

Notes:

n¼

147

pro

du

ctio

nte

ams;

* p,

0.05

;*

* p,

0.01

Table II.Regression results for age

effects and short-termconsequences

Consequences ofage in work

teams

141

Qu

alit

yS

ick

occa

sion

sB

urn

out

Var

iab

leb

DR

2F

(2,

64)

bDR

2F

(3,

86)

bDR

2F

(2,

144)

Ste

p1

Con

trol

Gro

up

size

20.

125

0.03

72,

679

20.

138

0.00

00.

017

0.10

20.

040

4,94

0S

tep

2In

dep

end

ent

Av

erag

eag

e0.

568

*0.

253

13,8

802

0.64

7*

0.33

521

,925

20.

451

*0.

232

21,9

75A

ge

dif

fere

nce

20.

095

0.00

59,

353

0.12

90.

013

15,3

332

0.08

60.

006

14,9

46T

otal

R2

0.29

50.

348

0.27

7A

dju

sted

R2

0.26

40.

326

0.25

9

Note:

* p,

0.01

Table III.Regression results for ageeffects and long-termconsequences

CDI13,2

142

(b ¼ 20:647 with p , 0.01). Also in the case of quality a positive effect regarding ageseems to be present. The data show a strong and positive relationship between averageage and this long-term group level consequence. Again for age differences in teams, nosignificant relationships were reported to any of the long-term consequences.

DiscussionIn this paper we have examined and illuminated the direct relationships between ageand short-term and long-term consequences in work teams. We used the model ofMilliken and Martins (1996) to extend the theory about age effects on work team level,by further examination of team variables for affective, cognitive andcommunication-related consequences as well as long-term consequences on grouplevel. The data in our study indicate important effects of average age in teams on thefunctioning of work teams. Higher average age seems to have significant positiveeffects on all short-term and long-term consequences.

For satisfaction and involvement, it was observed that teams with higher averageage find more enjoyment in their work and feel good in their teams and in the company.This might be related to team as well as to company tenure and is in line with researchdone by Sterns and Miklos (1995) who stated that job satisfaction is consistentlypositively related to work attitudes with increasing age.

In the areas of mutual learning, feedback and decision-making, work teams profitfrom higher average age. Higher average age goes hand in hand with accumulatedknowledge through the years and building up intellectual capital (Peterson and Spiker,2005) that can be effectively used for mutual learning. Teams with higher average agealso might be good at addressing and motivating behavior in the team, as we found asmall but significant positive relationship between average age and feedback in teams.Similarly, decision making can be expected to improve by higher average age. Thisexperience can be regarded as knowledge about the work, but perhaps also moreparticularly concerns increased knowledge about cooperating with others in work teamsand better understanding the organization, therefore being able to make decisions in abetter way. More explanations are hard to find, but may have to do with the history ofthe organization. Work teams were implemented about ten years before the start of ourstudy in the organization and it is only in that time that closer attention has been paid togroup dynamics and team building. It seems to pay off that teams with higher averageage are more used to the earlier culture of hierarchy and task orientation, have learned tobe more open in communication, cooperation and giving feedback to peers.

Teams with higher average age seem to produce better quality. This again mayhave to do with more experience, knowledge and responsibility for high productquality, something which was also observed by Sterns and Miklos (1995). Ourobservations also can be supported by the findings from a study done byBorsch-Supan et al. (2006) who concluded that the productivity in a German carmanufactory does not decline with age.

Teams with higher average age have more trouble keeping up with a fast work paceduring several hours of the day and are also less interested in putting high strains uponthemselves; therefore, they choose a more balanced production schedule, with betterresults in terms of sick occasions and burnout. The studies by Peterson and Spiker(2005) and De Lange et al. (2006) support our findings regarding lower levels of sickleave and burnout with increasing age.

Consequences ofage in work

teams

143

Age differences appear to have no effect on the measurements we used for short-and long-term consequences in work teams. These findings are not in line withfindings by other researchers (see Milliken and Martins, 1996). They found that agedifferences have clear relationships to both short- and long-term consequences in workteams. One of the reasons for not finding such relationships might have to do with themodels we tested. Our correlation analysis indeed shows significant relationshipsbetween both average age and age differences and the outcome variables. However, ifwe add both age variables in our regression models, the effect of age differences seemsto disappear. Of course there is a small correlation between the two, but apparently theaverage age in teams reveals much more robust relationships to team consequencesthan age differences do. An alternative explanation for such findings might lay in thefact that age differences in this production environment are not that high.

The most interesting outcome of our study is the positive relationship of averageage in teams to all short- and long-term consequences in work teams. If such findingsare comparable to other organizations or teams, an ageing workforce seems to providepredominantly positive effects, and delivers important inputs for team developmentand team composition.

Limitations and further researchSeveral limitations regarding the current study should be noted. First, our data werecollected in a single production organization. Although a large variety of teams wasincluded in the research, the focus was predominantly on production teams. Therelationships we found may be influenced by the fact that most of the work teamsconsisted of blue-collar workers and teams of white-collar workers may need to beincluded in further research.

Second, and perhaps related to the first limitation, is the fact that the age differenceswe found in the teams were not very high (a standard deviation of 7.3 years). Perhapsthis is specific for this organization or may be a general observation in productionorganizations; more research also in other industries will be necessary to compare toour findings in this.

Third, the data were collected at a company in Sweden and as such there may be acultural dimension involved with the effects of age, or at least effects concerningattitudes about age. Sweden is known for a rather high labor participation andwell-developed labor policies to support labor participation of more vulnerable groups.On the other hand, the average age in teams may not be considered to be specificallyhigh, so it might be unlikely that such effects play a too important role in theexplanation of our findings.

Fourth, in explaining the relationships we found between average age and short-and long-term consequences, we suggest causal effects. Although it is unlikely topresume that these short-term and long-term consequences do affect the average age inteams (reversed causality) it may be taken into account that in unique cases certainteam member selection criteria play a role in the composition of teams as a result of e.g.affective consequences. This might for instance be the case when a team of satisfiedolder workers will seek to influence the selection process favoring new team membersto be of comparable age as well.

Fifth, in our model we also considered a strong relationship between tenure and theage variables, which would have provided multi-collinearity problems if used in the

CDI13,2

144

same statistical analyses. That is because age characteristics, such as tenure andworking experience are always highly correlated with age. In the relationship betweenthese variables the age factor contributes most and plays a dominant role. Taking thatinto account, we decided to focus on the two age variables (average team age and agedifferences in teams), thereby including all age-related characteristics.

Further research in more organizations both in production as well as in otherindustries should be conducted to compare our findings and outcomes. Also, aninternational comparison could lead to clearer findings of age effects on short- andlong-term work team consequences based on different cultural backgrounds.Especially the study of effects of both age differences and average age in teamsneeds further attention, as many studies about work teams focus on differences ratherthan ageing itself (see Milliken and Martins, 1996). Additionally, age effects in teamsshould be compared to individual age effects in organizations and there should beexploration of more advanced models that help to understand the dynamic processesand consequences of age in teams. Further, the relationship between supervisors andtheir team members should receive more attention. Based on the ageing society we mayexpect that in the future supervisors will have to guide more experienced (elder) teammembers and therefore have to deal with ageing workforces more properly. Researchin the area of so-called economic outcomes would help to develop models to understandthe effects of age on economic outcomes and should help organizations to deal withageing workforces without a decline in business performance. In this study we havemade a first start by including objective performance data, such as for quality and sickoccasions, next to perceived performance data such as for burnout.

Implications for practice and researchOur research shows clear short-term and long-term consequences of increasing age inwork teams. These findings are very promising considering the ageing workforce andshould induce companies to nurture their older employees. Teamwork and teamperformance may benefit clearly from older workers and therefore it seems relevant totake a closer look at the management of teams and HRM policies to support teamworkwith increasing numbers of older workers. Especially in the manufacturing industry,where we conducted our study, attention needs to be paid to the circumstances that cankeep older workers active in the production process. Job redesign and automation maybe used to create a working environment that is physically less straining, but to createmore room for decision making and learning. New workplace measures may have to beintroduced, intending to improve employment and participation of the ageingworkforce. Although we did not find any effects of age differences, team compositionmay be used to also put more older workers in younger teams to increase the averageage. Attention should be paid to a developed learning environment in teams to supportthe exchange of knowledge from older to younger workers and to develop teamwork.

The positive findings of our study regarding the effects of older work teams supportthe need to further develop HRM policies in organizations to facilitate older workers.This could result in new programs for recruitment, training, health-prevention andreintegration of elder people as well as in team composition in general. The next decadepresents a major challenge to reverse the trends from exit and exclusion of olderworkers towards strategies to improve employment prospects of the many differentgroups in the ageing workforce. This will demand multi-disciplinary and coordinated

Consequences ofage in work

teams

145

approaches involving governments and especially the social partners, e.g. unions andwork councils, as well as workers themselves. There are signs of change and signalsabout the necessary measures; but significant investment in the ageing workforcerequires a fundamental reorientation of policies, concepts and attitudes. On the otherhand, workers in the second half of working life have to become more aware of the needto improve their employability and career opportunities, overcoming discriminationand other age barriers in employment.

Our research outcome contributes to research studies in HRM as well as tomanagement studies. The research field of ageing effects in work teams is furtherexamined; specific effects are explored and may extend existing theories. Therelevance of our study can be seen as an extension of prior studies (Milliken andMartins, 1996) to deepen the scientific background of the phenomenon of direct ageeffects in work teams and to start further research on new aspects discovered in ourresearch.

References

Allen, N.J. and Meyer, J.P. (1996), “Affective, continuance, and normative commitment to theorganization: an examination of construct validity”, Journal of Vocational Behaviour,Vol. 49, pp. 252-76.

Berggren, C. (1993), Alternatives to Lean Production: Work Organization in the Swedish AutoIndustry, ILR Press, Ithaca, NY.

Borsch-Supan, A., Duzgun, I. and Weiss, M. (2006), “Age and productivity in work teams:evidence from the assembly line”, unpublished paper, University of Mannheim, Mannheim.

Burke, R.J. and Ng, E. (2006), “The changing nature of work and organizations: implications forhuman resource management”, Human Resource Management Review, Vol. 16, pp. 86-94.

Cohen, S.G. and Baily, D.E. (1997), “What makes teams work: group effectiveness research fromthe shop floor to the executive suite”, Journal of Management, Vol. 23 No. 3, pp. 239-90.

Considine, G. and Callus, R. (2001), “The quality of working life of Australian employees: thedevelopment of an index”, working paper 73, ACIRRT, University of Sydney, Sydney.

Cox, T. and Blake, S. (1991), “Managing cultural diversity: implications for organizationalcompetitiveness”, Academy of Management Executive, Vol. 5 No. 3, pp. 45-56.

Cummings, A., Zhou, J. and Oldham, G.R. (1993), “Demographic differences and employee workoutcomes: Effects on multiple comparison groups”, paper presented at the Annual Meetingof the Academy of Management, Atlanta, GA.

De Lange, A.H., Taris, T., Jansen, P., Smulders, P., Houtman, I. and Kompier, M. (2006), “Age as afactor in the relation between work and mental health: results of the longitudinal TASsurvey”, Occupational Health Psychology, Vol. 1.

Dunphy, D. and Bryant, B. (1996), “Teams: panaceas or prescriptions for improved performance”,Human Relations, Vol. 49 No. 5, pp. 677-99.

Elfenbein, H.A. and O’Reilly, C.A. III (2005), “Fitting in: the effects of relational demography andperson-ulture fit on group process and performance”, research paper series, paper no 1728,Stanford Business School, Stanford, CA.

Field, A. (2000), Discovering Statistics Using SPSS for Windows: Advanced Techniques for theBeginner, 2nd ed., Sage Publications, London.

Horwitz, S.K. (2005), “The compositional impact of team diversity on performance: theoreticalconsiderations”, Human Resource Development Review, Vol. 4 No. 2, pp. 219-49.

CDI13,2

146

Hut, J.A. and Molleman, E. (1998), “Empowerment and team development”, Team PerformanceManagement, Vol. 4 No. 2, pp. 53-66.

Jackson, S.E., Joshi, A. and Erhardt, N.L. (2003), “Recent research on team and organizationaldiversity: SWOT analysis and implications”, Journal of Management, Vol. 29 No. 6,pp. 801-30.

Jackson, S.E., May, K.E. and Whitney, K. (1995), “Understanding the dynamics of diversity indecision-making teams”, in Guzzo, R.A., Salas, E. (Eds), Team Effectiveness and DecisionMaking in Organizations, Jossey-Bass, San Francisco, CA.

Katzenbach, J.R. and Smith, D.K. (1993), The Wisdom of Teams: Creating the High-performanceOrganization, Harvard Business School Press, Boston, MA.

Kuipers, B.S. (2005), Team Development and Team Performance, Responsibilities, Responsivenessand Results, Labyrinth Publications, Ridderkerk.

Le Blanc, P.M. (1994), De Steun van de Leiding, Thesis Publishers, Amsterdam.

Miles, J. and Shevlin, M. (2001), Applying Regression and Correlation: A Guide for Students andResearchers, Sage Publications, London.

Milliken, F.J. and Martins, L.L. (1996), “Searching for common threads: understanding themultiple effects of diversity in organizational groups”, Academy of Management Review,Vol. 21 No. 2, pp. 402-33.

Molleman, E. (2005), “Diversity in demographic characteristics, abilities and personality traits:do faultlines affect team functioning?”, Group decision and Negotiation, Vol. 14, pp. 173-93.

O’Connell, M.S., Doverspike, D. and Cober, A.B. (2002), “Leadership and semi-autonomous workteam performance”, Group and Organisation Management, Vol. 27 No. 1, pp. 50-65.

O’Reilly, C.A. III, Williams, K.Y. and Barsade, S.G. (1999), “The impact of relational demographyon teamwork: when majorities are in the minority”, research paper no. 1551, GraduateSchool of Business, Stanford University, Stanford, CA.

Pelled, L.H. (1996), “Demographic diversity, conflict, and work group outcomes: an interveningprocess theory”, Organizational Science, Vol. 7 No. 6, pp. 615-31.

Pelled, L.H., Eisenhardt, K.M. and XIN, K.R. (1999), “Exploring the black box: an analysis ofwork group diversity, conflict and performance”, Administrative Science Quarterly, Vol. 44No. 1, pp. 1-28.

Peterson, S.J. and Spiker, B.K. (2005), “Establishing the positive contributory value of olderworkers: a positive psychology perspective”, Organizational Dynamics, Vol. 34 No. 2,pp. 153-67.

Remery, C., Henkens, K., Schippers, J. and Ekamper, P. (2003), “Managing an ageing workforceand a tight labor market: views held by Dutch employers”, Population Research and PolicyReview, Vol. 22, pp. 21-40.

Rhodes, S.R. (1983), “Age-related differences in work attitudes and behaviour: a review andconceptual analysis”, Psychological Bulletin, Vol. 93, pp. 328-67.

Schalk, R. et al., (2007), “Research on ageing and work in the EU: where do we stand and whereshould we go?”, unpublished position paper, University of Tilburg, Tilburg.

Sterns, H.L. and Doverspike, D. (1989), “Aging and the retraining and learning process inorganizations”, in Goldstein, I. and Katzel, R. (Eds), Training and Development in WorkOrganizations, Jossey-Bass, San Francisco, CA, pp. 229-332.

Sterns, H.L. and Miklos, S.M. (1995), “The ageing worker in a changing environment:organizational and individual issues”, Journal of Vocational Behaviour, Vol. 47, pp. 248-68.

Consequences ofage in work

teams

147

Sundstrom, E., McIntyre, M., Halfhill, T. and Richards, H. (2000), “Work groups: from thehawthorn studies to work teams of the 1990s and beyond”, Group Dynamics, Vol. 4 No. 1,pp. 44-67.

Suzaki, K. (1993), The New Shop Floor Management: Empowering People for ContinuousImprovement, The Free Press, New York, NY.

Tsui, A.S. and O’Reilly, C.A. III (1989), “Beyond simple demographic effects: the importance ofrelational demography in superior-subordinate dyads”, Academy of Management Journal,Vol. 32 No. 2, pp. 402-23.

Waldman, D.A. and Avolio, B.J. (1986), “A meta-analysis of age differences in job performance”,Journal of Applied Psychology, Vol. 71 No. 1, pp. 33-8.

Warr, P. (2001), “Age and work behaviour, physical attributes, cognitive abilities, knowledge,personality traits and motives”, in Cooper, C.L. and Robertson, I.T. (Eds), InternationalReview of Industrial and Organizational Psychology, Wiley, Chichester.

Wiersema, M.F. and Bantel, K.A. (1992), “Top management team demography and corporatestrategic change”, Academy of Management Journal, Vol. 35 No. 1, pp. 91-121.

Williams, K.Y. and O’Reilly, C.A. III (1998), “The complexity of diversity: a review of forty yearsof research”, in Staw, B. and Sutton, R. (Eds), Research in Organizational Behaviour,Vol. 21, JAI Press, Greenwich, CT, pp. 77-140.

Appendix. Overview of itemsShort-term affective consequences

(1) Satisfaction:. Work giving pleasure.. Being enthusiastic about current work.

(2) Involvement:. Preferring to work at this company.. Feeling connected to this company.

Short-term cognitive consequences

(1) Mutual learning:. Share information about the work.. Listen well to each other during team meetings.. Often swap tasks with each other.. Help each other to perform the tasks.. Perform all the necessary/important work.. Decide by itself who performs which task.. Perform routine maintenance tasks.. Keep everything clean and organized.. Take action on mistakes.

(2) Decision making:. Decide by itself who performs which task.. Make decisions about the work.. Take part in making a group decision.. Stick to their agreements.

CDI13,2

148

. Perform the agreed delegated management tasks.

. Take responsibility for the agreed delegated management tasks.

Short-term communication-related consequencesFeedback:

. Evaluate each other’s way of working.

. Motivate each other to carry out the work in a proper way.

. Address each other’s mistakes when carrying out the tasks.

Long-term consequences: work team level performanceBurnout:

. Feeling tired because of the work.

. Feeling burned out because of the work.

About the authorsFranz J. Gellert is a Senior Lecturer at Hanze University of Applied Sciences in Groningen wherehe has taught since 2001. Franz Josef held various manager positions in trade and industry. InHanze University he specializes in marketing, economics and management. His researchinterests include social behavior and all age related issues. Furthermore he is interested in teamdevelopments and team dynamics. Franz J. Gellert is the corresponding author and can becontacted at: f.j.gellert@pl.hanze.nl

Ben S. Kuipers is a PhD in Management and Organization and specialized in teamwork,performance management and organizational change. His dissertation “Team development andteam performance. Responsibilities, responsiveness and results: a longitudinal of teamwork atVolvo Trucks Umea” (2005) has won him multiple awards. After his PhD research at theUniversity of Groningen he combined researching with his own consulting business(Performability) to support companies in developing team performance. As a lecturer andresearcher at Erasmus University Rotterdam he is involved with HRM and change managementprograms.

Consequences ofage in work

teams

149

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints