The effects of service climate and the effective ......moderating model that predicts the interaction effects between service climate and effective leadership behaviour on employee

Copyright © The British Psychological SocietyReproduction in any form (including the internet) is prohibited without prior permission from the Society

The effects of service climate and the effectiveleadership behaviour of supervisors on frontlineemployee service quality: A multi-level analysis

C. Harry Hui1*, Warren C. K. Chiu2, Philip L. H. Yu3, Kevin Cheng4

and Herman H. M. Tse5

1Department of Psychology, University of Hong Kong, Hong Kong2Department of Management and Marketing, Polytechnic University of Hong Kong,Hong Kong

3Department of Statistics and Actuarial Science, University of Hong Kong,Hong Kong

4Department of Politics and Sociology, Lingnan University, Hong Kong5UQ Business School, University of Queensland, Australia

A supervisor’s behaviour may not be the only factor that determines the performanceof team members (Kerr & Jermier, 1978). Taking this postulation as a basis, weformulated a model to describe how service climate moderates the effects of theleadership behaviour of supervisors. When the organization and working environmentare not conducive to providing a good service to colleagues and customers, thesupervisor’s leadership behaviour makes an important difference. However, when theservice climate is good, a supervisor’s leadership behaviour makes no substantialdifference. This hypothesis was supported in an examination of the service quality of511 frontline service providers as sampled from 55 work groups in 6 serviceorganizations. The employee service quality was low when both the service climate andthe supervisor’s leadership behaviour were lacking. However, when the service climatewas unfavourable, effective leadership behaviour played a compensatory role inmaintaining performance standards towards external customers. When the leadershipwas ineffective, a favourable service climate nullified the negative effect on servicequality to internal customers.

In today’s economy, the success of a business largely depends on the quality of service

provided to its customers (Berry, 1995; Gutek, 1995; Zeithaml & Bitner, 1996). Sasserand Arbeit (1976) distinguished between the service that is provided to internal and

external customers, and argued that both are equally important. Quality service to

* Correspondence should be addressed to C. Harry Hui, Department of Psychology, University of Hong Kong, Pokfulam,Hong Kong (e-mail: [email protected]).

TheBritishPsychologicalSociety

151

Journal of Occupational and Organizational Psychology (2007), 80, 151–172

q 2007 The British Psychological Society

www.bpsjournals.co.uk

DOI:10.1348/096317905X89391


internal customers results in empowered and happy colleagues who are likely to carry

out their own work more effectively (Azzolini & Shillaber, 1993). Quality service to

external customers brings customer satisfaction and loyalty (Rogelberg, Barnes-Farrell, &

Creamer, 1999), which will in turn result in repeat business (Kotler, 1999). Hurley (1998)

suggests that employees with whom customers interact directly should act proactively

and exercise discretion as to how they deliver service quality to satisfy or even surprisecustomers. It is therefore important to understand what drives good service. According to

past research, two antecedents to the work output of employees are service climate

(e.g. Schneider, White, & Paul, 1998) and effective leadership of direct supervisors

(e.g. Bettencourt & Brown, 1997; Deluga, 1994; Podsakoff, MacKenzie, Moorman, &

Fetter, 1990).

Although service climate and effective leadership behaviour have been found to have a

significant impact on employee service quality, little work has been undertaken to

integrate these two conceptually related constructs in the same study. In this paper, weexamine the relative contribution of service climate and effective leadership behaviour to

employee service quality, and the interaction effect of these two independent variables

on the outcome variable. The interaction effect is important, given the literature on the

substitutes of leadership model (Kerr & Jermier, 1978). In addition, we note the

importance of conceptualizing service climate and effective leadership behaviour as

group-level variables, and the need to look at their influence on outcome variables beyond

the individual level (e.g. Schneider et al., 1998; Strutton, Pelton, & Lumpkin, 1993;

Williams, Podsakoff, & Huber, 1992; Yammarino & Dubinsky, 1992, 1994). Althoughservice climate and effective leadership behaviour can best be conceptualized as group-

level variables (Ehrhart, 2004; Schneider et al., 1998), to the best of our knowledge, no

study has so far examined the direct and interaction effects of these two constructs at

multiple levels. This gap in the literature motivates the present study.

This study plans to make the following contributions. First, we formulate a

moderating model that predicts the interaction effects between service climate and

effective leadership behaviour on employee service quality. This will contribute to an

ongoing discussion on the substitutes of leadership model of Kerr and Jermier (1978).Second, in response to the call for a multi-level approach, we test the model at both the

individual and group levels. Third, whereas most studies on these two potential

antecedents to employee service quality have been conducted in North America and

Europe, this study considers service climate and effective leadership behaviour in a

Chinese society. As customer satisfaction and service quality are relatively new to many

Chinese workers, data from this population will be useful for the research community in

developing and refining theories for application beyond cultural boundaries. In the

following sections, we briefly survey the literature on the relationships between serviceclimate, leadership behaviour and employee service quality, and then move on to a

description and test of the moderating model at the group level.

Impact of supervisor’s effective leadership behaviour on service qualityThe literature abounds with actions that supervisors can undertake to be effective

leaders. These actions can be crudely organized into three clusters: task-oriented

actions (sometimes called performance or initiating structure), people-oriented

actions (sometimes called maintenance or consideration) and ethical actions

(sometimes called moral character). The first two have been discussed extensively

by earlier researchers such as Blake and Mouton (1978), Fleishman (1967) and Misumi

152 C. Harry Hui et al.


(1985). Some examples of performance actions are making timely decisions, motivating

employees, giving directions, drawing up plans and meeting deadlines. In a service

setting, effective supervisors influence and encourage the service behaviour of

employees by setting targets for their frontline subordinates. They empower, inspire,

reward and serve as role models so that their frontline subordinates understand how to

deliver the best service. Maintenance actions include respecting the decisions ofsubordinates, resolving conflicts, listening to the views of subordinates, helping

subordinates to achieve organizational and sometimes personal goals and being

supportive when subordinates encounter work problems (e.g. Schaubroeck & Fink,

1998). Moral character – or ethical leadership – has recently attracted the attention of

researchers (e.g. Emler & Cook, 2001; Greenleaf, Frick, & Spears, 1996; Kanungo &

Mendonca, 1996; Ling & Fang, 1995; Srivastva, 1988), and can be linked to

transformational leadership (Podsakoff et al., 1990; Turner, Barling, Epitropaki, Butcher,

& Milner, 2002). Moral character includes the supervisor’s fairness (Bettencourt &Brown, 1997) and trust-building behaviour (Deluga, 1994). Good supervisors act in an

impartial and non-favouring manner and do not abuse their power for personal benefits.

Numerous studies have demonstrated that such effective leadership behaviour is

associated with the quality of work of subordinates in service organizations (e.g.

Bettencourt & Brown, 1997; Deluga, 1994; Pillai, Schriesheim, & Williams, 1999;

Podsakoff et al., 1990; Schaubroeck & Fink, 1998). Leaders who develop a good

relationship with their subordinates will in turn influence the level of discretionary

behaviour of these subordinates (e.g. Settoon, Bennett, & Liden, 1996). In a fieldexperiment, Lam and Schaubroeck (2000) found that using credible and influential

people as service quality leaders could enhance the positive attitudes of bank tellers

towards service quality initiatives, which result in a more favourable customer rating of

service. MacKenzie, Podsakoff, and Rich (2001) found that transformational leaders

could alter the service behaviour of frontline salespeople. Not only does effective

leadership behaviour affect individual service quality, it also enhances team service

quality. For example, supervisors’ level of ‘servant leadership’ (Greenleaf et al., 1996) is

associated with unit-level organizational citizenship behaviour in 249 grocery storedepartments (Ehrhart, 2004). Furthermore, proactive, considerate and charismatic store

managers have been found to significantly improve the objective business performance

of the supermarkets that they run (Koene, Vogelaar, & Soeters, 2002). A critical mass of

colleagues delivering good service has a modelling effect on others. Thus, the

supervisor’s effort in raising service quality at the team level can be multiplied in its

impact.

Impact of service climate on service qualityService climate is derived from a consensual understanding, within a company, a

department or a group, of how to behave in different settings and with different

customer populations. Borucki and Burke (1999) defined service climate in terms of

employee cognitive appraisals of the organization’s attitude towards employee well-

being, and the concern of members of the organization about customer well-being. This

includes the extent of the perception that management sets clear performancestandards, provides appropriate training and information, removes obstacles to service,

assists in employee job performance and distributes rewards for good service to

customers. A favourable service climate is associated with excellent interdepartmental

service (Schneider et al., 1998). Service climate is also related to the perception that

Service climate and leadership 153


the organization and its members assist customers, and to outcome variables such as

individual and organizational service performance (Borucki & Burke, 1999) and

customer satisfaction with service quality (Johnson, 1996). Some data suggest that if the

favourable service climate is ‘strong’ (i.e. when employees agree on their perception of

the climate), then there is a low variability in customer satisfaction. However, a weak

service climate is associated with a high variability in customer satisfaction (Schneider,Salvaggio, & Subirats, 2002).

Relationship between effective leadership behaviour of supervisors and serviceclimateSome writers have conceptualized leadership behaviour as a precursor to organizational

climate (e.g. Dickson, Smith, Grojean, & Ehrhart, 2001; Koene et al., 2002; Litwin &

Stringer, 1968) or group climate (Kivlighan & Tarrant, 2001). Kozlowski and Doherty

(1989) integrated the vertical dyad linkage theory of leadership with climate perception,

and showed that subordinates with whom a supervisor relates closely have a more

positive evaluation of the organizational climate than subordinates who have a poor-quality relationship with their supervisors. More recently, Schneider et al. (1998)

included in their formulation of service climate the subordinates’ perception of the

actions of their immediate supervisors to reward, support and expect good customer

service. Furthermore, management’s recognition and appreciation of quality service

appears in the ‘climate for service’ scale. It should be noted, however, that although the

supervisor is a member of the organization, his or her identity is separate from that of

the organization. Although there is a partial overlap between the perception of a

supervisor’s behaviour and the organizational climate, a frontline supervisor does nothave the power to dictate all features contributing to the service climate. Instead, senior

management cultivates and nurtures the service climate by designing the company’s

compensation structure and setting policies on interdepartmental communication, the

management of customer information, customer service policies and so forth. For this

reason, this study treats service climate and the leadership behaviour of direct

supervisors as conceptually separate entities. Moreover, in our measurement, we

attempt to minimize any overlapping that may confound the two constructs.

Interaction between service climate and effective leadership behaviour: A moderatingmodelThere is a large body of literature on the situational approach to leadership (e.g. Evans,1970; Fiedler, 1967; House, 1971). A quarter of a century ago, Kerr and Jermier (1978)

theorized that the effects of leadership can be substituted for, neutralized or enhanced by

certain contextual variables. Many researchers (e.g. Howell, Bowen, Dorfman, Kerr, &

Podsakoff, 1990; Podsakoff, MacKenzie, & Fetter, 1993; Podsakoff, Niehoff, MacKenzie, &

Williams, 1993; Williams et al., 1988; Yukl, 1994) have concurred that Kerr and Jermier’s

substitutes model is preferable to other situational approaches to leadership. It represents

the most comprehensive attempt to identify situational factors and explains why some

types of leader behaviour are effective in some situations but have no effect – or even adysfunctional effect – in other situations (Podsakoff & MacKenzie, 1997).

However, not all empirical tests have been supportive of the substitutes model

(e.g. Farh, Podsakoff, & Cheng, 1987; Podsakoff & MacKenzie, 1995; Podsakoff,

MacKenzie, Ahearne, & Bommer, 1995). The search for contextual variables that may



moderate the impact of leadership on criterion variables is likened by some to looking

for ‘a needle in a haystack’ (Podsakoff et al., 1995). This unenthusiastic view has been

contested by advocates of the substitutes model. According to Villa, Howell, Dorfman,

and Daniel (2003), the failure to detect moderators in leadership research may be

traceable to the theoretical nature of many of the empirical tests. Indeed, an extensive

review of the literature shows that variability in the performance of subordinates may be

explained by a number of non-leader variables, such as the professional orientation of

employees, task feedback, intrinsic job satisfaction, non-routine tasks, group

cohesiveness and spatial distance among workers. When these factors are present,

the competence of the leader becomes relatively unimportant.

In sum, the jury is still out on whether the substitute variables of Kerr and Jermier are

in fact predictive of important criterion variables. Nevertheless, both sides agree that the

substitutes framework should be used to stimulate further research. For example,

Jermier and Kerr (1997) stated that they had never intended to treat the substitutes

framework as a closed system, and that a better specification of the conceptual domain

of substitutes is long overdue. One contextual variable – shared organizational values –

was proposed by Podsakoff and McKenzie (1997), and may have particular promise as a

moderator. They argued that shared values or work norms can replace hierarchy or

coercion, because members of a value-based organization do not need much

supervision.

In this study, we focus on the employees’ shared perception that the organization

places much emphasis on customer well-being, and examine how such a service

climate, like some organizational shared values, can moderate the impact of leadership

behaviour in a service setting. Specifically, leadership effectiveness may be more

strongly correlated with subordinates’ work performance when the organizational

service climate is not favourable than when it is favourable. We speculate that, in

organizations that do not have a positive service climate, not all employees will

understand the need for satisfying customers or acquiring the means to do so, and nor

will they perceive the organization as supporting good service. Under such

circumstances, an important role of supervisors is to clarify what is expected of their

subordinates in terms of customer service, and to provide the necessary coaching and

managerial support to maintain the service quality of the work unit. This is consistent

with the findings of Brubakk and Wilkinson (1996) that, during changes in corporate

culture, managers play an important role in helping frontline employees to decode and

interpret what must be changed. In short, when the organization or the team does not

have a favourable service climate, the frontline supervisors’ effective leadership

behaviour will play an important role helping individual service providers to understand

service values and realize the importance of customer satisfaction. When team members

work individually and collectively with peers to satisfy customers, the quality of service

provided by the entire work team is strengthened.

There are times when leadership may not produce the intended effect. This happens

in settings in which the organizational climate clearly supports, encourages, and

rewards service (Martin, 2002). When there are sufficient mechanisms to guide and

orchestrate employee behaviour, the immediate supervisor’s leadership behaviour is of

relatively minor importance. Furthermore, a competent supervisor who issues

directives that are not completely in line with the organizational values embodied in

the service culture and climate can promote confusion and dysfunction. In such a

situation, Porter and Bigley (2001) predict that role ambiguity results among



subordinates. This, in turn, depresses service quality indirectly, as Hartline and Ferrell

(1996) showed in their study of hotel employees.

Hence, leadership can even backfire: ‘the popular idea that low levels of motivation

can be remedied by increased transformational leadership behaviour appears to be

overly simplistic where organizations are concerned. Indeed, low levels of employee

motivation may be caused by too much transformational leadership behaviour in theorganization’ (Porter & Bigley, 2001, p. 288). Likewise, when a highly acclaimed

leader’s direction does not align with what is prescribed for a well-developed service

climate, the benefits of leadership effectiveness and favourable service climate may be

cancelled out.

We therefore hypothesize that the relationship between effective leadership

behaviour and employee service quality will be stronger in work groups in an

unfavourable service climate, and weaker or even reversed in those that operate in a

positive service climate. We adopt a multi-level approach to test this hypothesis.

Method

Respondents and proceduresTo ensure the generalizability of our prospective findings to a broad spectrum of

companies in the service industry and to maximize the variability of the constructs of

interest, we sampled a variety of organizations: a telecommunications company, a retailchain, two hotels, an auto-repair company and a government department. All of these

organizations pride themselves in their effort to satisfy customers, and some have even

made this a part of their mission statement. The research participants were all Chinese

frontline service staff who worked in service teams of varying sizes.

A team is defined either by function or by locality. For instance, in a hotel, a team may

comprise all of the staff in a restaurant, whereas in a retail organization, it may be

defined by sales outlet. A team is generally led by a supervisor, to whom all team

members report. During our negotiation with the organizations, we requested access toteams that met two conditions: the members of such teams had to have regular contact

with customers, and the team members had to have more than 6 months working

experience in that frontline position. For very large teams, we randomly selected about

75% of the individuals from a name list that the company supplied. We did not request

participation from everyone in order to avoid unnecessary disruption to the company’s

business day-to-day operation.

Fifty-five service teams (511 employees) participated in the study. They included 14

teams from the telecommunications company, 11 from the retail chain, 7 from the twohotels, 13 from the auto-repair company and 10 from the government department. The

teams ranged in size from 5 to 20 members. Slightly over two-thirds had 10 or fewer

members. In this sample, 61% of the participants were female, 49.5% were within the 15

to 30 age bracket, and 32.8% were within the 31 to 40 age bracket.

Data collection sessions were held in groups of 5 to 20 people in either a training

room or a large recreation area near or at their place of work. Each session began

with a researcher explaining that the aim of the study was to gain a better

understanding of service workers. Although attendance was arranged by therespective organizations, the researcher told prospective participants that they could

choose not to take part. Participation rates ranged from 60% to 85% in the sampled

teams. If the participation rate went below 50%, the researcher dropped the team to

move on to another.



The supervisors of the participants separately rated the service quality of their

subordinates. Confidentiality was guaranteed for both the frontline employees and their

supervisors. The questionnaires that the subordinates completed were collected at the

end of the session, and the questionnaires of the supervisors were collected

approximately one week after distribution.

Although we initially secured the cooperation of 1,035 employees, only teams withfive or more members who provided complete answers to our survey questionnaire were

included in our study. This was necessary because a reasonable group size was required

for a reliable estimation of the group-level variables (Bliese, Halverson, & Schriesheim,

2002), which were used in the hierarchical linear modelling (HLM) analysis.

Measures

Supervisor’s effective leadership behaviourThis construct was measured by aggregating the reports of subordinates about their

supervisors on 15 behavioural items (see Appendix), adapted from several

questionnaires used to measure leadership behaviour (Hui & Tan, 1999; Ling & Fang,1995; Misumi, 1985). This selection was in keeping with our earlier argument that

effective leadership behaviour comprises three aspects: task-oriented actions (or

performance), people-oriented actions (or maintenance) and ethical leadership (or

moral character). Most of the items that pertained to task-oriented and people-oriented

actions were adapted from the Misumi (1985) Performance-Maintenance Question-

naire. The items on ethical leadership were derived from the studies of Hui and Tan

(1999) and Ling and Fang (1995). A 6-point scale anchored by strongly disagree and

strongly agree was used. A principal components analysis revealed only one generalfactor, which accounted for more than 59% of the total variance. The Cronbach’s alpha

was .95.

Service climateSchneider (Schneider, Ashworth, Higgs, & Carr, 1996; Schneider et al., 1998)

developed four Climate for Service scales, namely, ‘global service climate’, ‘customer

orientation’, ‘managerial practices’ and ‘customer feedback’. Of these, we used only

the first two subscales (16 items in total). There are three reasons for this decision.

First, we needed to keep the survey questionnaire to a reasonable length to secure

the cooperation of both the companies and their employees. Second, these subscales

are highly correlated with each other (rs ¼ :50 to .74; Schneider et al., 1998). With‘global service climate’ being correlated with ‘customer feedback’ at r ¼ :63, some

subscales can clearly be omitted. Third, as there is already too much conceptual and

textual overlap between the effective leadership behaviour measure and the

‘managerial practices’ subscale, to avoid confounding the effect of the former in our

model, it was better to have left the latter out.

Some of the items that were included were: ‘employees’ job knowledge and skills to

deliver work and service’, ‘the business does a good job keeping customers informed of

changes that affect them’ and ‘top management has a plan to improve the quality of ourwork and service’. There were two items from the ‘climate for service’ measure that

seem to refer to leadership (e.g. ‘the recognition and rewards employees receive for the

delivery of superior work and service’ and ‘leadership shown by management in

supporting the service quality effort’). However, closer examination revealed that they



are more related to the company’s general management practice than the behaviour of

direct supervisors and thus would not confound our two research variables.

A comparison of two models (one with and the other without the two items as part

of the leadership behaviour measure) also confirmed this. For this reason, we decided to

retain the two items in our measure.

The subsequent principal component analysis indicated a unidimensional structure.This is not surprising in view of the high correlation (r ¼ :74) that was first reported by

Schneider et al. (1998). The 7 items of the ‘global service climate’ subscale were rated

on a 5-point scale that ranged from poor to excellent). However, since the items of the

‘customer orientation’ subscale were anchored by strongly disagree and strongly agree,

we used a 6-point format to avoid a ‘neutral’ mid-scale anchor for our Chinese

respondents, who are relatively high on a central tendency response set. To give equal

weighting to each item when aggregating for an overall service climate score, we

transformed the 5-point ratings into a 6-point scale by subtracting the ratings by 1 anddividing by 4. We then multiplied the ratio by 5 and added 1 to the product term. The

Cronbach’s alpha was .90.

Although most of the items used in this study were validated previously, there may

still be questions as to whether there are conceptual differences between organizational

service climate and leadership behaviour. To verify this, we conducted confirmatory

factor analyses on the 15 leadership and 16 service climate items, to test whether the

team members’ ratings would load on two distinct factors. Results showed that the one-

factor model (CFI ¼ :89, GFI ¼ :83, AGFI ¼ :81, RMSEA ¼ :07, x2 ð434Þ ¼ 1; 404:361)offered a significantly poorer fit than the two-factor model (CFI ¼ :90, GFI ¼ :84,

AGFI ¼ :82, RMSEA ¼ :06, x2 ð433Þ ¼ 1; 310:48; x2 difference test statistic ¼ 93:88,

df ¼ 1, p , :0001). This confirms that, although related to each other, service climate

and leadership behaviour are conceptually distinct.

Outcome measuresEmployee quality of customer service was operationalized by the respective supervisor’s

ratings on an Employee Service Quality Scale (Hui, Cheng, & Gan, 2003). The first

subscale, ‘service to internal customers’, had 9 items that explored facilitative behaviour

directed at internal customers. The raters reported the frequency of such actions over

the previous 3 months (1 ¼ once or never; 6 ¼ six or more times). Sample items

included ‘respond immediately to co-workers’ urgent requests’, ‘inform other team

members about changes that might affect service quality’ and ‘voluntarily assist workers

from other departments’. The Cronbach’s alpha was .90. The second subscale, ‘serviceto external customers’, had 7 items, and measured behaviour that was directed at

external clients. Sample items included ‘willingly listen to customers’ requests’, ‘provide

non-standard service to customers’ and ‘liaise with different departments to provide

service in one service contact’. The Cronbach’s alpha was .86. For both subscales, a high

score indicated high service quality. Although the two subscales are correlated (Table 1),

confirmatory factor analyses conducted using AMOS showed that the one-factor model

(CFI ¼ :73, GFI ¼ :70, AGFI ¼ :61, RMSEA ¼ :14, x2 ð119Þ ¼ 1370:30) offered a

significantly poorer fit than the two-factor model (CFI ¼ :85, GFI ¼ :83, AGFI ¼ :76,RMSEA ¼ :11, x2 ð118Þ ¼ 801:83; x2 difference test statistic ¼ 568:47, df ¼ 1,

p , :0001). Service to internal customers and service to external customers are related

but non-identical aspects of an underlying service construct. Hence, we treated these

two subscales as separate outcome measures.



Hierarchical linear modelling (HLM) as a multi-level analytic techniqueHLM is a popular method for analysing data in a nested structure by constructing a

separate submodel at each of the levels in the data structure (Bryk & Raudenbush, 2002;

Bryk, Raudenbush, & Congdon, 1996). It allows us to make simultaneous inferences on

the effects of variations in the independent variables at the individual level and group

level on the dependent variables, and the cross-level moderating effect of the

independent variables on the dependent variable at the individual level.

To understand multi-level models, it is useful to begin with a simple (single-level)

regression model for employee service quality ( y):

yi ¼ b0 þ b1xi þ ei;

where the subscript i represents the i th employee and xi is the employee rating of the

supervisor’s leadership behaviour centred about the supervisor mean. The model

assumes that the differences between supervisors are constant across the range of

employee service quality scores. We can think of the regression line as a supervisor-level

summary of this relationship, which may vary from supervisor to supervisor. This kind

of variation is known as group-level variation. To account for both individual-level and

group-level variation, we consider the following multi-level model with random

intercepts:

Individual level: yij ¼ b0j þ b1xij þ eij;

Group level: b0j ¼ c0 þ c1gxj þ c2gcj þ c3gxj £ gcj þ uj;

where the subscript i indicates the employee and the subscript j indicates the

supervisor, and gxj represents the aggregated ratings of the supervisor’s effective

leadership behaviour, gcj represents service climate and gxj £ gcj represents the

interaction between the two. To see all of the predictors that are implied in the two-level

model, we substituted the group-level model into the individual-level model, which

resulted in a single equation model:

yij ¼ ðc0 þ c1gxj þ c2gcj þ c3gxj £ gcj þ ujÞ þ b1xij þ eij:

At the individual level, a linear model was built to predict employee service quality

based on the individual ratings of the supervisor’s effective leadership behaviour xij.

Table 1. Means, standard deviations and correlations at individual level and group levela

Individuallevel Group level

Variables Mean SD Mean SD 1 2 3 4

1. Service quality to internal customers 3.00 1.32 2.98 0.89 .90 .68*** 2 .08 .022. Service quality to external customers 3.57 1.32 3.45 1.11 .60*** .86 .09 .053. Effective leadership behaviour 4.01 0.92 4.03 0.41 .09* .09* .95 .48***4. Service climate 3.77 0.81 3.79 0.43 .08 .04 .53*** .90

The theoretical range of all ratings is 1–6; the Cronbach’s alphas are presented on the diagonal;elements below the diagonal are correlations between constructs at individual level; elements abovethe diagonal are correlations at group level.

*p , :05; **p , :01; ***p , :001aN ¼ 511 at the individual level and N ¼ 55 at the group level



As xij is centred about the supervisor mean, each intercept parameter b0j that was

obtained from the individual level indicated the corresponding supervisor’s mean rating

of the service quality of their subordinates. At the group level, another linear model was

built to relate these intercept parameters and the aggregated ratings of the supervisor’s

effective leadership behaviour and service climate. The model had two residuals: the

individual level residual eij, which measured the difference between actual employee

service quality and expected service quality based on the individual-level model, and the

group-level residual uj, which measured the difference between a supervisor’s mean

rating of the service quality of subordinates and the supervisor’s expected service

quality rating based on the group-level model.

Results

Preliminary analysesThe employees in the sample were slightly older (t ¼ 22:55, p , :001) and moreexperienced in the industry (t ¼ 33:78, p , :001) than those that were not selected for

the study. They had also been employed longer in the organization (t ¼ 68:43, p , :001).

Despite this, the two groups did not differ in their rating of their supervisor’s leadership

behaviour (t ¼ 1:10,ns) and service climate (t ¼ 0:02,ns), nor did they differ in quality of

service for internal (t ¼ 1:64, ns) and external customers (t ¼ 21:04, ns).

All subsequent analyses were performed on the sample of 511 employees. Table 1

shows the means, standard deviations and correlations of the variables at the individual

and group levels.At both levels, the quality of service that was delivered to internal customers is

associated with that delivered to external customers, but not to service climate. The

ratings of service climate and effective leadership behaviour are also correlated. Service

quality is significantly related to effective leadership behaviour at the individual level,

but not at the group level. To account for the impact of different types of business and

organizational structures on service quality, we controlled for organization type (i.e.

public vs. for-profit) in our analyses. After introducing the multi-level structure to the

model, we found the organization type effect to be significant for service quality to

external customers, but not for service quality to internal customers.

Data aggregation: Analysis of within-group consistency and between-groupdifferencesBoth h2 and intra-class correlation (ICC(1)) as computed through a random intercepts

model indicated the proportion of total variance between teams (Bliese, 2000; Bryk &

Raudenbush, 2002). According to Bliese, ICC(1) values are typically under .20 and areusually smaller than the corresponding h2 values. In our case, the ICC(1) and h2 are .26

(range ¼ :06– :50) and .20, respectively, for service climate, and .14 (range ¼ 2:02– :33)

and .22, respectively, for effective leadership behaviour. ICC(2) was used to estimate the

reliability of the mean values that we obtained from aggregation. The ICC(2) values for

service climate and effective leadership behaviour are .51 and .58, respectively. There is

thus a substantial variance in service climate and effective leadership behaviour between

teams.

The mean rwg values ( James, Demaree, & Wolf, 1993) for the two group-level variables

(service climate and effective leadership behaviour) are .62 (SD ¼ 0:24) and .80



(SD ¼ 0:11), respectively. Both values are higher than the .60 cut off recommended by

James (1982), and indicative of adequate agreement among the team members.

Taken together, the rwg and ICC values reveal an overall picture of between-team

difference and some within-team similarities. The broad range of agreement values

reflects reality that, in certain teams and for various reasons, members do not always

concur in their perceptions. Although somewhat below the recommended benchmark,ICC (1) ¼ .12 (James, 1982) and ICC (2) ¼ .60 (Glick, 1985), the ICC estimates for

service climate are above those that are reported by Schneider et al. (1998), with ICC

(1) ¼ .09 and ICC (2) ¼ .47. As Schneider et al. were still able to aggregate for group

data and obtain meaningful analytic results at that level of agreement, given our present

ICC estimates, aggregation is still a reasonable method for estimating the two group-level

variables of service climate and effective leadership behaviour.

Hierarchical regression analysisWe conducted a two-level hierarchical regression analysis to predict service quality for

internal customers and another to predict service quality for external customers. The

proportion of variance in service quality for internal customers between supervisors is

36.4%, and the proportion of variance in service quality to external customers between

supervisors is 60.7%. This means that around 36% to 61% of the variation occurs over the

employees nested within the supervisors. A significant chi-squared statistic for the errorvariance that is estimated at the group level indicates that there is systematic variance in

the intercept across supervisors (residual variance for service quality to internal

customers ¼ 0:64, p , :0001; residual variance for service quality to external

customers ¼ 1:13, p , :0001). Tables 2 and 3 contain the regression coefficients of

the hierarchical regression analyses.

Three multi-level regression models were considered, with the effects of

organization type, team size and effective leadership behaviour being controlled for at

the individual level. Model 1 depicts the impact of effective leadership behaviour only,and is supported at the individual-level for both dependent variables. Model 2 includes

both predictors at the individual and group levels. However, the addition of this service

climate variable did not improve the prediction of either dependent variable at either

level. Model 3 differs from Model 2 in that it includes the interaction term at both the

individual and group levels, although only the interaction term at the group-level is

significant for both dependent variables. Testing Models 2 and 3 revealed that the

psychological service climate at the individual-level did not have a significant effect. This

is probably because the construct is either non-existent or because there is too muchmeasurement error. If the latter is true, then there is a reason for aggregating the data

and for examining the effects at the group-level only. In fact, the removal of the

psychological service climate at the individual level yielded similar results, which are not

shown here due to space limitations.

To interpret the interaction effect on service quality in Model 3, we considered two

levels of effective leadership behaviour and two levels of service climate by taking the

standard deviation from its own mean and forming high and low levels of effective

leadership behaviour and of favourable service climate. Figure 1 shows the estimatedquality of service to internal and external customers for these four combinations.

When the service climate is poor, a supervisor who displays effective leadership

behaviour enhances employee service quality. A simple slope test (Aiken & West, 1991)

showed that this effect is especially pronounced for service to external customers



Table

2.

Two-lev

elhie

rarc

hic

alre

gres

sion

anal

ysis

pre

dic

ting

serv

ice

qual

ity

toin

tern

alcu

stom

ersa

Coef

fici

ents

Leve

lVar

iable

Model

1M

odel

2M

odel

3St

andar

diz

edco

effici

ents

ofM

odel

3

Indiv

idual

Public

vers

us

for-

pro

fit

org

aniz

atio

nb

2.2

3(.16)

2.2

5(.19)

2.2

2(.19)

–G

roup

size

.01

(.04)

.01

(.04)

2.0

0(.04)

–Effec

tive

lead

ersh

ipbeh

avio

ur

.20

(.05)*

**.1

9(.07)*

*.1

9(.07)*

*.1

6Psy

cholo

gica

lse

rvic

ecl

imat

e.0

2(.07)

.02

(.07)

.01

Effec

tive

lead

ersh

ipbeh

avio

ur£

Psy

cholo

gica

lse

rvic

ecl

imat

e2

.05

(.07)

2.0

3

Gro

up

Inte

rcep

t2.8

4(.16)*

**2.8

3(.18)*

**2.9

3(.18)*

**–

Effec

tive

lead

ersh

ipbeh

avio

ur

2.2

1(.32)

2.1

8(.35)

2.2

1(.26)

2.0

9Se

rvic

ecl

imat

e2

.08

(.36)

2.0

4(.36)

2.0

2Effec

tive

lead

ersh

ipbeh

avio

ur£

Serv

ice

clim

ate

2.9

4(.38)*

2.2

3

Public

org

aniz

atio

n(c

oded

1):

ago

vern

men

tdep

artm

ent.

For-

pro

fit

org

aniz

atio

n(c

oded

21):

hote

ls,a

tele

com

munic

atio

ns

com

pan

y,an

auto

-rep

air

com

pan

yan

da

reta

ilch

ain.

*p,

:05;**p,

:01;**

*p,

:001

a Stan

dar

der

rors

are

inpar

enth

eses

.bT

his

contr

olva

riab

leca

nbe

ente

red

atei

ther

the

indiv

idual

leve

lor

the

group

leve

lw

ithout

chan

ging

the

valu

eofth

epar

amet

eres

tim

ates

.



Table

3.

Two-lev

elhie

rarc

hic

alre

gres

sion

anal

ysis

pre

dic

ting

serv

ice

qual

ity

toex

tern

alcu

stom

ersa

Coef

fici

ents

Leve

lVar

iable

Model

1M

odel

2M

odel

3St

andar

diz

edco

effici

ents

ofM

odel

3

Indiv

idual

Public

vers

us

for-

pro

fit

org

aniz

atio

nb

2.3

6(.18)*

2.4

5(.20)*

2.4

2(.21)

–G

roup

size

.10

(.03)*

*.0

9(.03)*

*.0

8(.03)*

–Effec

tive

lead

ersh

ipbeh

avio

ur

.12

(.04)*

*.1

6(.05)*

*.1

5(.05)*

*.1

2Psy

cholo

gica

lse

rvic

ecl

imat

e2

.07

(.06)

2.0

7(.06)

2.0

5Effec

tive

lead

ersh

ipbeh

avio

ur£

Psy

cholo

gica

lse

rvic

ecl

imat

e2

.05

(.07)

2.0

3

Gro

up

Inte

rcep

t3.3

5(.18)*

**3.2

9(.20)*

**3.3

9(.21)*

**–

Effec

tive

lead

ersh

ipbeh

avio

ur

.30

(.34)

.47

(.38)

.42

(.33)

.17

Serv

ice

clim

ate

2.3

9(.45)

2.3

4(.44)

2.1

5Effec

tive

lead

ersh

ipbeh

avio

ur£

Serv

ice

clim

ate

2.9

8(.44)*

2.2

4

Public

org

aniz

atio

n(c

oded

1):

ago

vern

men

tdep

artm

ent.

For-

pro

fit

org

aniz

atio

n(c

oded

21):

hote

ls,a

tele

com

munic

atio

ns

com

pan

y,an

auto

-rep

air

com

pan

yan

da

reta

ilch

ain.

*p,

:05;**p,

:01;**

*p,

:001

a Stan

dar

der

rors

are

inpar

enth

eses

.bT

his

contr

olva

riab

leca

nbe

ente

red

atei

ther

the

indiv

idual

leve

lor

the

group

leve

l,w

ithout

chan

ging

the

valu

eofth

epar

amet

eres

tim

ates

.



(slope ¼ :85, t ¼ 3:11, p , :001), although the slope is not significantly different from

zero for service to internal customers. When the service climate is positive, the

supervisor’s effective leadership behaviour does not make a difference in the service of

subordinates to external customers, but surprisingly is associated with low service

quality to internal customers (slope ¼ 2:60, t ¼ 22:02, p , :05).

Discussion

This study aimed to examine how the two key variables of effective leadership

behaviour and service climate are related to the work performance of frontline

employees. We hypothesized that the relationship between supervisors’ effective

leadership behaviour and subordinates’ service quality would be more prominent in anunfavourable than in a favourable service climate. The results provide preliminary

support to our prediction of an interaction effect. In an unfavourable service climate,

compared with teams led by less effective supervisors, teams led by effective

supervisors delivered better service (especially to external customers). In a favourable

service climate, however, effective leadership behaviour does not seem to enhance the

service that subordinates provide to external customers. It may even be negatively

related to the quality of service provided to internal customers. In the following

paragraphs, we interpret the interaction effects, speculate on why combining effectiveleadership behaviour with a favourable service climate does not always lead to better

team outcomes, discuss why neither variable has a main effect on team outcomes and

describe the limitations of this study.

Interaction effect on quality of service to external customersIn the present study, teams led by effective supervisors and having a favourable service

climate did not receive the highest ratings on quality of service to external customers.The first explanation is that supervisors (also the raters) in such teams had very high

performance expectations for their subordinates, and were therefore harsh when rating

their subordinates’ service quality. Second, as we pointed out in the Introduction, a co-

occurrence of a good leader and a favourable service climate may lead to role conflict;

2.0

2.5

3.0

3.5

4.0

Ineffective leadershipbehaviour

Effective leadershipbehaviour

Ser

vice

qua

lity

to in

tern

al c

usto

mer

s

Unfavourable service climateFavourable service climate

2.0

2.5

3.0

3.5

4.0

Ineffective leadershipbehaviour

Effective leadershipbehaviour

Ser

vice

qua

lity

to e

xter

nal c

usto

mer

s

Unfavourable service climateFavourable service climate

Figure 1. Subordinates’ service quality as a function of service climate and supervisors’ effective

leadership behaviour.



that is, when the leader points at a direction different from that of the company. Third,

perhaps a favourable service climate does not alter the level of service very much.

Instead, it merely maintains a certain level of service critical for the organization to

survive, such that the service quality is consistent even when the team is under the

supervision of someone not very effective in leadership. However, an unfavourable

climate becomes a backdrop for competent leaders to demonstrate their skills andcharisma in full strength, and to motivate their subordinates into productive activities.

Furthermore, when organizational service climate is not as favourable, employees may

not know exactly what their performance goals should be. Those who are willing to

work extra hard could be prompted by a good leader to display more service behaviour.

This effect was evident in some dot-coms that were starting up during the 1990s. While

those companies did not have in place a perfect set of practices and policies (which are

a component of favourable service climate), at the inspiration of their visionary bosses

employees were still willing to work extra hours to serve their customers.Of course, we would not recommend that management should keep service climate

unfavourable in order that frontline employees have high external service quality.

Maintaining an unfavourable service climate would put businesses at risk. This is

because teams that have poor service climate may deliver no service at all, if the

supervisor is a very ineffective leader, as can be visualized by extrapolating the left-hand

bottom part of Figure 1. Unfavourable service climate is a liability to any organization,

while a favourable climate provides a buffer against poor leadership. In fact, strong

management should be introduced and competent managers selected for places whereservice climate is poor.

Interaction effect on quality of service to internal customersThe fact that employees deliver a better service to internal customers when their

supervisors are ineffective leaders than when their supervisors are effective in a

favourable service climate is not unexplainable. This phenomenon is consistent with thesuggestion of Netemeyer, Boles, Mckee, and McMurrian (1997) that colleagues help one

another to enhance customer relations when supervisors cannot provide timely

assistance and advice. As our data show, such mutual assistance when supervisors are

ineffective exists only in a service climate that provides members with structures,

standards and implicit rules on satisfying customers. A positive organizational climate

may have cultivated a collegial spirit and motivated employees to make up for any

shortcomings that are due to supervisors not functioning as they should. In the absence

of such a climate, however, members are unlikely to become good organizationalcitizens.

Ironically, when the service climate is favourable and leadership is effective, the

quality of service may also be low. Teams with both ‘desirable’ conditions recorded a

lower than expected performance. This finding is nevertheless in line with that which

others have reported in the literature. For example, in their study of over 100 teams in

the pharmaceutical industry, Gibson and Vermeulen (2003) found that when a team

‘already has an impetus to [perform], a performance management push by its : : : .leader may be superfluous’ (p. 214). This set of findings corroborates the postulationthat leaders are not needed under certain conditions (De Vries, Roe, & Taillieu, 2002;

Kerr & Jermier, 1978).

Although this pattern of interaction confirms the predictions of the substitutes

model, it challenges the deep-seated belief that an effective leader and a favourable



climate should be additive in their positive impact on employees’ service to their

colleagues and peers. For this somewhat novel finding, we offer further theoretical and

analytical discourse.

First, a highly effective (and therefore possibly persuasive) supervisor and a

favourable service climate together may be confusing to the subordinates, if their

directions are not consistent with each other. This proposition deserves furtherinvestigation in future research.

Second, when the supervisor and the service climate are working in alignment, the

result would probably be an emphasis on serving external customers over the internal

ones. This is because few organizations explicitly enforce and reward extra-role

behaviour (Organ, 1988, 1997). Therefore, working beyond their prescribed duties to

assist their colleagues is entirely the employees’ discretion. These discretionary acts will

be less frequent when two conditions coexist: a positive service climate that directs the

attention of employees to their tasks, which for our participants working in thefrontline, largely involves serving external customers, and a supervisor who effectively

reinforce this idea. After all, management assesses and rewards frontline employees

primarily on the basis of how well they serve external customers, but not internal

customers, even though it is important to do both.

Third, as in the case of rating of service to external customers, this rating could be

harsh among effective supervisors and less so among the less effective ones. The

emphasis of service in the organization or team may also elevate the expectations, thus

resulting in a lower rating of employees who were in teams where the service climatewas favourable and the leader was highly competent. The above explanations should be

taken with a grain of salt, because although the interaction effect was significant, neither

slope was significantly different from zero.

Absence of group-level main effectsIn this study, effective leadership behaviour predicted employee service quality only atthe individual level but not at the group level. Two explanations can be offered for this

apparent inconsistency. First, the correlation at the individual level may merely be a

result of mutual appreciation between the supervisor and the subordinate spilling over

into the subordinate ratings of leadership behaviour and the supervisor ratings of

service quality. Second, individual ratings of the leadership behaviour of supervisors

reflect the supervisor’s actions towards the subordinate concerned, and therefore have a

stronger impact on the subordinate’s work performance than an abstract measure of

leadership behaviour that is displayed towards a range of subordinates and aggregatedfrom ratings provided by these subordinates.

We did not find any main effect of service climate. A positive service climate does not

always result in the favourable evaluation of the work performance of service providers.

Although one may attribute this to the fact that we did not measure service climate with

the ‘customer feedback’ subscale, we note that Schneider et al. (2002) found no main

effect for service climate on customer perception either, despite using the full scale. We

believe that service climate is more ‘remote’ and intangible than leadership behaviour

from the subordinates’ perspective, and thus is less potent. Another explanation is that,although there may be a link between service climate and individual employee service

performance in theory, the operation of a myriad other factors and moderators may have

masked the relationship. Unfortunately, many of these moderators are as yet unknown.

This study, which sampled a variety of organizations, takes a step towards understanding



the boundary conditions that surround the effect of service climate. As we have shown,

a positive service climate may well go some way towards preserving service quality

when leadership is mediocre, but paradoxically may not help when the supervisor

displays behaviour that is regarded as effective.

One further explanation for the lack of effect of service climate pertains to

methodology. In the teams in which good service was accorded a high value,supervisors might have been conditioned to use an unduly high standard when they

rated the performance of their subordinates. The net result of this stringency bias would

then be a depression of performance ratings in the teams that operated within a

favourable service climate. However, although this is a possible explanation, future

research should continue to search for substantive explanations for this phenomenon.

At a conceptual level, these findings cast doubt on the robustness of a direct causal

link between leadership behaviour and organizational service climate on the one hand

and employee service quality on the other. On the contrary, they provide support for thesubstitutes model that is discussed at the beginning of this article. More research is

needed to replicate the present findings and to identify other boundary conditions of

these determinants.

Limitations, future research and practical implicationsLike any study, this one is not without limitations. We have slightly over-sampled olderand more experienced employees, and many of the teams that were sampled had fewer

than 10 members and some as few as 5. This sampling procedure presented two

problems. First, the data that were aggregated from a small group might have been

corrupted by one or two outliers, which would have resulted in unreliable findings.

Second, the generalizability of our findings to larger teams may be restricted.

Fortunately, subsequent correlation analyses showed no systematic relationship

between team size and any of the major variables except service quality to external

customers (r ¼ :26, p , :001). Furthermore, as team size was controlled in all of theregression analyses, we can be confident that its effect on our findings was minimal. All

things considered, our findings are probably generalizable to working environments

with small and medium-sized service teams that are staffed by ‘seasoned’ customer

service personnel.

Another limitation of the present study is that employees’ service quality was

assessed by supervisors only. While the use of supervisory rating is already an

improvement over the use of self-rating as in most early organizational citizenship

behaviour studies, supervisors’ rating stringency could partially account for theobservation that effective leadership did not bring good performance ratings

when service climate is excellent. Future research should therefore examine the

interaction effect with other rating sources, such as using customers’ and possibly 3608

assessment.

This study has several practical implications. In today’s economy, the bottom line of

commercial organizations depends on how service is delivered at the frontline. To excel

in the service industry, supervisors are taught leadership skills and management

consultants are brought in to help cultivate a service climate. Sometimes these initiativesare uncoordinated, with supervisors and management consultants emphasizing their

own interest domain. For example, leadership workshops may inspire supervisors to be

attentive to the needs of their subordinates and to encourage them to innovate, whereas

management consultants may implement standardized ways in which to serve



customers better. These two driving forces are not necessarily opposed to one

another, but organizations should be aware of any potential incompatibility or

redundancy.

In sum, the present study is probably one of the few to suggest that a good supervisor

and a positive service climate may work against one another. Practitioners are thus

advised to assess how these two performance drivers can best be combined andmanaged in their own organizations. However, readers are reminded that results and

discussions of this study, while intriguing, are exploratory and need to be replicated in

future research.

Acknowledgements

This research as partially supported by a research grant (Project B-Q416) from the

Hong Kong Polytechnic University. All five authors contributed equally to the research and

writing of this paper. We wish to thank Fleur Sequeira and David Wilmshurst for comments on an

earlier draft.

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.

London: Sage.

Azzolini, M., & Shillaber, J. (1993). International service quality: Winning from the inside out.

Quality Progress, 26, 75–78.

Berry, L. L. (1995). Relationship marketing of service – Growing interest, emerging perspectives.

Journal of the Academy of Marketing Science, 23, 236–245.

Bettencourt, L. A., & Brown, S. W. (1997). Contact employees: Relationships among

workplace fairness, job satisfaction and prosocial service behaviors. Journal of Retailing,

73, 39–61.

Blake, R. R., & Mouton, J. S. (1978). The new managerial grid. Houston, TX: Gulf.

Bliese, P. D. (2000). Within-group agreement, non-independence and reliability: Implications for

data aggregation and analysis. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory,

research, and methods in organizations: Foundations, extensions, and new directions

( pp. 351–424). San Francisco, CA: Jossey-Bass.

Bliese, P. D., Halverson, R. R., & Schriesheim, C. A. (2002). Benchmarking multilevel methods in

leadership: The articles, the model and the data set. Leadership Quarterly, 13, 3–14.

Borucki, C. C., & Burke, M. J. (1999). An examination of service-related antecedents to retail store

performance. Journal of Organizational Behavior, 20, 943–962.

Brubakk, B., & Wilkinson, A. (1996). Agents of change? Bank managers and the management of

corporate culture change. International Journal of Service Industry Management, 7, 21–43.

Bryk, A. S., & Raudenbush, S. W. (2002). Hierarchical linear models: Applications and data

analysis methods (2/e). Newbury Park, CA: Sage.

Bryk, A. S., Raudenbush, S. W., & Congdon, R. T. (1996). HLM: Hierarchical linear modeling with

the HLM/2L and HLM/3L programs. Chicago: Scientific Software International.

Deluga, R. J. (1994). Supervisor trust building, leader-member exchange and organizational

citizenship behaviour. Journal of Occupational and Organizational Psychology, 67,

315–326.

De Vries, R. E., Roe, R. A., & Taillieu, T. C. B. (2002). Need for leadership as a moderator of the

relationships between leadership and individual outcomes. Leadership Quarterly, 13,

121–137.

Dickson, M. W., Smith, D. B., Grojean, M. W., & Ehrhart, M. (2001). An organizational climate

regarding ethics: The outcome of leader values and the practices that reflect them. Leadership

Quarterly, 12, 197–217.



Ehrhart, M. G. (2004). Leadership and procedural justice: Climate as antecedents of unit-level

organizational citizenship behavior. Personnel Psychology, 57, 61–94.

Emler, N., & Cook, T. (2001). Moral integrity in leadership: Why it matters and why it may be

difficult to achieve. In R. Hogan & B. W. Roberts (Eds.), Personality psychology in the

workplace ( pp. 277–298). Washington, DC: American Psychological Association.

Evans, M. G. (1970). The effects of supervisory behavior on the path-goal relationship.

Organizational Behavior and Human Performance, 5, 277–298.

Farh, J. L., Podsakoff, P. M., & Cheng, B. S. (1987). Culture-free leadership effectiveness

versus moderators of leadership behavior: An extension and test of Kerr and Jermier’s

substitutes for leadership model in Taiwan. Journal of International Business Studies, 18,

43–60.

Fiedler, F. E. (1967). A theory of leadership effectiveness. New York: McGraw-Hill.

Fleishman, E. A. (1967). Performance assessment based on an empirically derived task taxonomy.

Human Factors, 9, 349–366.

Gibson, C., & Vermeulen, F. (2003). A healthy divide: Subgroups as a stimulus for team learning

behavior. Administrative Science Quarterly, 48, 202–239.

Glick, W. H. (1985). Conceptualizing and measuring organizational and psychological climate:

Pitfalls in multilevel research. Academy of Management Review, 10, 601–616.

Greenleaf, R. K., Frick, D. M., & Spears, L. C. (1996). On becoming a servant-leader. San Francisco:

Jossey-Bass.

Gutek, B. A. (1995). The dynamics of service: Reflections on the changing nature of

customer/provider interactions. San Francisco: Jossey-Bass.

Hartline, M. D., & Ferrell, O. C. (1996). The management of customer-contact service employees:

An empirical investigation. Journal of Marketing, 60, 52–70.

House, R. J. (1971). A path goal theory of leader effectiveness. Administrative Science Quarterly,

16, 321–338.

Howell, J. P., Bowen, D. E., Dorfman, P. W., Kerr, S., & Podsakoff, P. M. (1990). Substitutes for

leadership: Effective alternatives to ineffective leadership. Organizational Dynamics, 19,

21–38.

Hui, C. H., Cheng, K., & Gan, Y. (2003). Psychological collectivism as a moderator of the impact of

supervisor-subordinate personality similarity on employees’ service quality. Applied

Psychology: An International Review, 52, 175–192.

Hui, C. H., & Tan, G. C. (1999). The moral component of effective leadership: The Chinese case. In

W. H. Mobley (Ed.), Advances in global leadership, Vol. 1 ( pp. 249–266). Stamford, CT: JAI.

Hurley, R. F. (1998). Customer service behavior in retail settings: A study of the effect of service

provider personality. Journal of the Academy of Marketing Science, 26, 115–127.

James, L. R. (1982). Aggregation bias in estimate of perceptual agreement. Journal of Applied

Psychology, 47, 219–229.

James, L. R., Demaree, R. G., & Wolf, G. (1993). Rwg: An assessment of within-group interrater

agreement. Journal of Applied Psychology, 75, 306–309.

Jermier, J. M., & Kerr, S. (1997). Substitutes for leadership: Their meaning and measurement –

Contextual recollections and current observations. Leadership Quarterly, 8, 95–101.

Johnson, J. W. (1996). Linking employee perceptions of service climate to customer satisfaction.

Personnel Psychology, 49, 831–851.

Kanungo, R. N., & Mendonca, M. (1996). Ethical dimensions of leadership. Thousand Oaks, CA:

Sage.

Kerr, S., & Jermier, J. M. (1978). Substitutes for leadership: Their meaning and measurement.

Organizational Behavior and Human Decision Processes, 22, 375–403.

Kivlighan, D. M., Jr., & Tarrant, J. M. (2001). Does group climate mediate the group leadership-

group member outcome relationship? A test of Yalom’s hypotheses about leadership priorities.

Group Dynamics, 5, 220–234.



Koene, B. A. S., Vogelaar, A. L. W., & Soeters, J. L. (2002). Leadership effects on organizational

climate and financial performance: Local leadership effect in chain organizations. Leadership

Quarterly, 13, 193–215.

Kotler, P. (1999). Kotler on marketing: How to create, win, and dominate markets. New York:

Free Press.

Kozlowski, S. W. J., & Doherty, M. L. (1989). Integration of climate and leadership: Examination of

a neglected issue. Journal of Applied Psychology, 74, 546–553.

Lam, S. K., & Schaubroeck, J. (2000). A field experiment testing frontline opinion leaders as

change agents. Journal of Applied Psychology, 85, 987–995.

Ling, W., & Fang, L. (1995). Theories on leadership and Chinese culture: The formulation of the CPM

model on leadership behavior and a Chinese implicit theory on leadership traits. In H. S. R. Kao,

D. Sinha & S. H. Ng (Eds.),Effective organizationsand social values ( pp. 269–286). New Delhi,

India: Sage.

Litwin, G. H., & Stringer, R. A., Jr. (1968). Motivation and organizational climate. Boston, MA:

Harvard University, Graduate School of Business Administration.

MacKenzie, S. B., Podsakoff, P. M., & Rich, G. A. (2001). Transformational and transactional

leadership and salesperson performance. Journal of the Academy of Marketing Science, 29,

115–134.

Martin, J. (2002). Organizational culture: Mapping the terrain. Thousand Oaks, CA: Sage.

Misumi, J. (1985). The behavioral science of leadership. Ann Arbor, MI: University of Michigan.

Netemeyer, R. G., Boles, J. M., Mckee, D. O., & McMurrian, R. (1997). An investigation into the

antecedents of organizational citizenship behaviors in a personal selling Context. Journal of

Marketing, 3, 85–98.

Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome.

Lexington, MA: Lexington.

Organ, D. W. (1997). Organizational citizenship behavior: It’s construct clean-up time. Human

Performance, 10, 85–97.

Pillai, R., Schriesheim, C. A., & Williams, E. S. (1999). Fairness perceptions and trust as mediators

for transformational and transactional leadership: A two-sample study. Journal of Manage-

ment, 25, 897–933.

Podsakoff, P. M., & MacKenzie, S. B. (1995). An empirical examination of the effects of substitutes

for leadership and leader behaviors at the individual-level and group-level of analysis.

Leadership Quarterly, 6, 289–328.

Podsakoff, P. M., & MacKenzie, S. B. (1997). Kerr and Jermier’s substitutes for leadership model:

Background, empirical assessment, and suggestions for future research. Leadership Quarterly,

8, 117–132.

Podsakoff, P. M., MacKenzie, S. B., Ahearne, M., & Bommer, W. H. (1995). Searching for a needle in

a haystack: Trying to identify the illusive moderators of leadership behaviors. Journal of

Management, 21, 422–470.

Podsakoff, P. M., Mackenzie, S. B., & Fetter, R. (1993). Substitutes for leadership and the

management of professionals. Leadership Quarterly, 4, 1–44.

Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader

behaviors and their effects on followers’ trust in leader, satisfaction, and organizational

citizenship behaviors. Leadership Quarterly, 1, 107–142.

Podsakoff, P. M., Niehoff, B. P., MacKenzie, S. B., & Williams, M. L. (1993). Do substitutes for

leadership really substitute for leadership? An empirical examination of Kerr and Jermier’s

situational leadership model. Organizational Behavior and Human Decision Processes, 54,

1–44.

Porter, L. W., & Bigley, G. A. (2001). Motivation and transformational leadership: Some

organizational context issues. In M. Erez, U. Kleinbeck & H. Thierry (Eds.), Work motivation

in the context of a globalizing economy ( pp. 279–291). Mahwah, NJ: Erlbaum.



Rogelberg, S. G., Barnes-Farrell, J. L., & Creamer, V. (1999). Customer service behavior: The

interaction of service predisposition and job characteristics. Journal of Business and

Psychology, 13, 421–435.

Sasser, W. E., & Arbeit, S. (1976). Selling jobs in the service sector. Business Horizons, 19, 61–66.

Schaubroeck, J., & Fink, L. S. (1998). Facilitating and inhibiting effects of job control and social

support on stress outcomes and role behavior: A contingency model. Journal of

Organizational Behavior, 19, 167–195.

Schneider, B., Ashworth, S., Higgs, A., & Carr, L. (1996). Design, validity, and use of strategically

focused employee attitude surveys. Personnel Psychology, 49, 695–705.

Schneider, B., Salvaggio, A. M., & Subirats, M. (2002). Climate strength: A new direction for climate

research. Journal of Applied Psychology, 87, 220–229.

Schneider, B., White, S. S., & Paul, M. C. (1998). Linking service climate and customer perceptions

of service quality: Tests of a causal model. Journal of Applied Psychology, 83, 15–163.

Settoon, R. P., Bennett, N., & Liden, R. C. (1996). Social exchange in organizations: Perceived

organizational support, leader-member exchange, and employee reciprocity. Journal of

Applied Psychology, 81, 219–227.

Srivastva, S. (1988). Executive integrity: The search for high human values in organizational

life. San Francisco, CA: Jossey-Bass.

Strutton, D., Pelton, L. E., & Lumpkin, J. R. (1993). The relationship between psychological climate

and salesperson-sales manager trust in sales organizations. Journal of Personal Selling and

Sales Management, 13, 1–14.

Turner, N., Barling, J., Epitropaki, O., Butcher, V., & Milner, C. (2002). Transformational leadership

and moral reasoning. Journal of Applied Psychology, 87, 304–311.

Villa, J. R., Howell, J. P., Dorfman, P. W., & Daniel, D. L. (2003). Problems with detecting

moderators in leadership research using moderated multiple regression. Leadership

Quarterly, 14, 3–23.

Williams, M. L., Podsakoff, P. M., & Huber, V. (1992). Effects of group-level and individual-level

variation in leader behaviours on subordinate attitudes and performance. Journal of

Occupational and Organizational Psychology, 65, 115–129.

Williams, M. L., Podsakoff, P. M., Todor, W. D., Huber, V. L., Howell, J. P., & Dorfman, P. W. (1988).

A preliminary analysis of the construct validity of Kerr & Jermier’s Substitutes for Leadership

scales. Journal of Occupational Psychology, 61, 307–333.

Yammarino, F., & Dubinsky, A. J. (1992). Superior-subordinates relationships: A multiple levels of

analysis approach. Human Relations, 46, 575–599.

Yammarino, F., & Dubinsky, A. J. (1994). Transformational leadership theory: Using levels of

analysis to determine boundary conditions. Personnel Psychology, 47, 787–816.

Yukl, G. (1994). Leadership in organizations. Englewood Cliffs, NJ: Prentice-Hall.

Zeithaml, V. A., & Bitner, M. J. (1996). Services marketing. New York: McGraw-Hill.

Received 18 October 2003; revised version received 7 November 2005

Appendix. Translation of items used to measure effective leadershipbehaviour

(1) Your supervisor rarely ‘stabs people in the back’ or seeks personal vengeance.

(2) Your supervisor often introduces unique insights and plans.

(3) Even if problems at work arise, your supervisor rarely scolds subordinatesunreasonably.

(4) Your supervisor rarely curries favour with senior management or submits

unquestioningly.

(5) Your supervisor can flexibly cope with any unpredictable circumstances.

(6) Your supervisor rarely gets too irrational and emotional at work.



(7) Your supervisor rarely abuses his/her power and authority.

(8) Your supervisor respects the decisions made by subordinates in their area of

authority and responsibility.

(9) Your supervisor rarely takes undeserved credit for other people’s work.

(10) Your supervisor is mindful of meeting deadlines in his/her work.

(11) Your supervisor listens to subordinates’ views and accepts their criticism with ahumble attitude.

(12) Your supervisor handles issues decisively.

(13) Your supervisor rarely works through inappropriate relations or uses underhanded

means to accomplish things for his/her own interest.

(14) Your supervisor provides useful suggestions when a subordinate encounters

problems at work.

(15) Your supervisor encourages his/her subordinates to freely voice their opinions.


Documents

The effects of service climate and the effective ......moderating model that predicts the interaction effects between service climate and effective leadership behaviour on employee