Challenger Space Shuttle Disaster Analysis

Leadership and Effectiveness in the Context of Gender: The Role of Leaders’

Verbal Behaviour*

Gisela Mohr and Hans-Joachim Wolfram University of Leipzig, Institute of Psychology, Department of Work and Organizational Psychology,

Seeburgstra�e 14-20, D-04103 Leipzig, Germany Corresponding author email: [email protected]

This field study focuses on verbal consideration, which is a leadership behaviour that expresses esteem for the follower and her or his work, knowledge and opinion. It was

assumed that the relationship between verbal consideration and various outcomes is

moderated by the leader’s gender. One-hundred-and-forty leaders and 455 of their

direct followers were surveyed in a one-wave questionnaire study in Germany. Male and female leaders showed the same degree of verbal consideration as rated by their

followers. Verbal consideration is related to three out of four outcome variables for both

sexes. One unexpected moderating effect of leaders’ gender was found: followers of

male leaders displaying verbal consideration report less ‘irritation’ (a state of exhaustion considered a threat to good task fulfilment). One explanation may be that

male leaders get ‘extra credit’ for showing verbal consideration as it may be thought to

entail special effort, whereas for female leaders it may be seen as normal and routine.

This assumption should be examined in further studies in order to get more information about the different mechanisms by which female and male leaders reach the same

quality of outcomes.

Meta-analyses have shown that certain types of leadership are positively related to outcomes (for transformational leadership see Lowe, Kroeck and Sivasubramaniam, 1996; for leader–member exchange see Gerstner and Day, 1997). In addition, there seem to be slight but consistent differences between male and female leaders in terms of leadership behaviour (Eagly, Johanne- sen-Schmidt and van Engen, 2003; Trinidad and Normore, 2005). There is some evidence that leadership style and gender are jointly related to outcomes (Eagly, Karau and Makhijani, 1995). In this paper, a closer look is taken at the possible

moderating effects of gender on the relationship between leadership and outcomes.

Leadership behaviour: the concept of verbal consideration

Different views on leadership exist in literature and research, with most of them focusing on an interactive approach. A similar approach is taken here, though this study focuses on a very specific type of interaction, namely verbal communica- tion. Managerial activities are, for the most part, fragmented in nature and have an interactive character (Yukl, 2002), and leading subordinates is only one of a broad variety of managers’ roles (Mintzberg, 1973). The study presented here is focused on this aspect of managerial activity, and as such the term leader seems to be more appropriate than the term manager. Guest, as

*The authors thank Marc van Veldhoven, Birgit Schyns and Tina Paul for their helpful comments on an earlier version of this paper. The paper is based on research funded by the Deutsche Forschungsgemeinschaft (DFG) No. Mo 440/4-1.

British Journal of Management, Vol. 19, 4–16 (2008) DOI: 10.1111/j.1467-8551.2007.00521.x

r 2007 British Academy of Management. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA, 02148, USA.

early as 1956, had already reported that leaders at the lower management level (foreman) used nearly half of their daily working hours ‘simply’ talking to others. Mintzberg (1973) reported that communication occupied 78% of the working day. Yukl (2002, p. 24) argues that research of managerial activities has shown that lower and middle managers are involved in oral commu- nication for anywhere between 27% and 82% of their work time, during which they are busy exchanging information or influencing people. Presumably, leaders would engage less in com- munication were it less crucial to the success of their leadership. The importance of the role of communication is evident when leaders carry out activities such as communicating goals to their followers, clarifying standards of fulfilment, giving feedback etc., all of which can be done effectively using (verbal) communication. Inter- action and communication seem therefore to be very important tools for leading successfully. Leaders have different means of interaction, e.g. the written word, the telephone, electronic messages or different types of meetings. However, leaders tend to prefer face-to-face interaction and oral communication for receiving and giving the up-to-date information needed to aid effective- ness (Yukl, 2002). Consequently, followers’ perceptions of leaders’ behaviours will be based to an important extent on the personal verbal communication behaviour of their leaders. Although many studies deal with the quantity

of communication in which leaders generally engage, not much research concerns itself with how followers perceive and rate their leaders’ verbal leadership behaviour and how this kind of behaviour is related to performance. In this study, a measurement of leadership behaviour, especially designed to explore the verbal interac- tion between leaders and followers from the perspective of followers, was included. Even if leaders are content with the way they are communicating with followers, the effect of their communication still depends on the evaluation of this communication by the recipients (i.e. fol- lowers). Therefore, it seems necessary to focus on the followers’ perceptions of their leader’s communication. Communicative behaviours have two main

functions for the fulfilment of work demands. To begin with, communicative behaviours convey information that is needed to understand the

objective of work tasks, and they include feed- back about the state of fulfilment of a task. Furthermore, communication transmits attitudes towards the interaction partner. For example, by asking followers for their opinions or knowledge, leaders are conveying their respect for the expertise of their followers (or at least showing that they take their followers seriously). With this in mind, the main interest of this study was to analyse the effectiveness of ‘verbal consideration’ which is more than just giving information. Verbal consideration contains a positive attitude towards the follower. In addition to the advantages posed by the

specific focus mentioned above, several reasons exist for not choosing well-established instru- ments of leadership research, such as those used for transformational leadership (Multifactor Leadership Questionnaire (MLQ), Bass and Avolio, 1995) or leader–member exchange (Liden and Maslyn, 1998). One main reason is that these traditional scales merge leadership behaviours and the effects of these behaviours (mostly from the perspective of followers). Leader–member exchange refers to the relation- ship quality between leader and follower (Liden and Maslyn, 1998), which is already an outcome of leadership behaviour. Similarly, transforma- tional leadership is defined ‘in terms of the leader’s effect on the followers, and the behaviour used to achieve this effect’ (Yukl, 1999, p. 286). Thus, these leadership constructs, which already include outcomes of leadership behaviour, cause methodological problems to arise in research focusing on leaders’ behaviour and outcomes, as independent and dependent variables are not separated. This methodological problem is ag- gravated by the fact that the items of these scales are highly evaluative. They do not ask for a description of concrete behaviours but for highly aggregated behaviour (e.g. transformational lea- dership’s subscale ‘Idealized Influence Attribu- ted’ refers to displaying a sense of power and confidence; MLQ, Bass and Avolio, 1995). Furthermore, such descriptions of aggregated behaviour (by untrained observers like followers) cannot be considered particularly ‘objective’, as they might be influenced by a wide range of perceptual distortions, e.g. assimilation and con- trast effects (cf. Biernat, 2003). The latter issue aside, the fact remains that the more aggregated a category of description is, the more decisions

Leadership and Effectiveness in the Context of Gender 5

r 2007 British Academy of Management.

have to be made by the observer when labelling an observable behaviour. That is, appraisals have to be made, opening the door even wider for the influence of the follower’s personal values. For this reason, a more descriptive way of asking for leaders’ behaviour was chosen in the present study. There is some empirical evidence suggesting

that transformational leadership is more effec- tive in unstable environments (Podsakoff, MacKenzie and Bommer, 1996) or in higher level positions (cf. Lowe, Kroek and Sivasu- bramaniam, 1996), but less successful in stress and crisis situations (Halverson, Murphy and Riggio, 2004). Verbal consideration is probably less context dependent because it focuses on what leaders do for most of the day: com- munication (Yukl, 2002). Verbal consideration applies to the daily work context for every level of leadership and in both routine and specific situations. It was assumed that the construct of verbal consideration would relate to other measures of leader–member exchange and transformational leadership, because this kind of communication between leader and follower is also a means of transmitting the respect a leader has for a follower as well as the vision the leader has. These correlations will be reported in the Preliminary Results section. When naming a scale ‘verbal consideration’,

the two subdimensions ‘consideration’ and ‘in- itiating structure’, which stem from the early work of the Michigan group (Fleishman, 1957), may come to mind. This initial conception involving two independent factors did not ulti- mately find empirical support (Bass, 1990). The above-mentioned twofold function of leaders’ communication – to give information for pur- poses of better task fulfilment and to transmit the leader’s attitude towards the followers – supports the construction of a scale which includes consideration (mostly considered to be a rather feminine behaviour; Cann and Siegfried, 1990) but also aspects of ‘initiating structure’, by means of communication.

Verbal consideration and expectations towards female leaders

Verbal consideration comprises communicative behaviours such as asking for the views of others

and noting others’ contributions, which are commonly seen as being more typical of women than of men (Diekman and Eagly, 2000). Thus, verbal consideration can be seen as being closer to the communal than to the agentic dimension of leadership behaviour and can therefore be expected to be shown more by women (commu- nal versus agentic, cf. Bakan, 1966; Parsons and Bales, 1955). Therefore it can be assumed that female leaders will receive higher ratings in verbal consideration.

H1: Female leaders receive higher follower ratings

in verbal consideration.

Leadership and effectiveness

The concept of verbal consideration takes into account that communication is always more than giving or receiving information (cf. Watzlawick, Beavin and Jackson, 1969). It seems reasonable to assume that verbal behaviour expressing esteem for another – which is what consideration is – will also foster the acceptance of the content of the message. Moreover, it should give followers the feeling that the way they are working is accepted, thus giving them a greater sense of security. It seems logical then to expect leaders whose communication style is assessed as being positive in verbal consideration to be more effective. The question concerning the measurement of

leadership effectiveness is an ongoing debate. Many definitions of leadership even incorporate the notion of effectiveness, such as Yukl’s (2002) definition: ‘Leadership is the process of influen- cing others to understand and agree about what needs to be done and how it can be done effectively, and the process of facilitating indivi- dual and collective efforts to accomplish the shared objectives’ (p. 7). The discussion concern- ing the selection of valid criteria has a long history (Lent, Aurbach and Levin, 1971; Smith, 1976; Witte, 1995). Yukl (2002) proposed three types of leadership effectiveness indicators: (a) ‘the extent to which a leader’s organizational unit performs its task successfully and attains its goals’ (p. 8), (b) the attitude of followers towards the leader, and (c) the contribution of the leader to the quality of group processes. These types of indicators imply different sources of assessment (e.g. groups, followers, leaders).

6 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.

Outcomes of verbal consideration

It is, of course, impossible to assess all kinds of effectiveness criteria in one study, and therefore a reasonable choice has to be made. Follower and leader rated outcomes were included in this study. The leaders provided their evaluation of the performance of the group (goal fulfilment). Fol- lowers rated two prominent examples of followers’ attitudes (job satisfaction and commitment). Moreover, a non-conventional criterion was added, which has so far been under-represented in leader- ship effectiveness studies and is an indicator of followers’ well-being (irritation, see below). In many studies, leadership has been shown to

influence job satisfaction (e.g. DeGroot, Kiker and Cross, 2001; Podsakoff, MacKenzie and Bommer, 1996). Presumably, verbal considera- tion is related to job satisfaction because this style refers to the positive communication between leader and member wherein the leader indicates that she or he values the follower’s opinion and thus expresses esteem for the follower. The leader thereby also shows an interest in getting to know the follower’s area of expertise. Being taken seriously in this way is a general human need and, when it is experienced on the job, may lead to job satisfaction.

H2: Verbal consideration is positively related to job

satisfaction.

Commitment refers to the emotional bond an employee feels towards the organization for which she or he works. Three forms of commit- ment are generally distinguished in research: normative, calculative and affective commitment (Meyer, 1997; Meyer, Allen and Gellatly, 1990). Leaders have the task of convincing their followers of the value of the organization’s goals. The goals of the organization should become the followers’ goals. When this happens, a bond develops between the follower and the organiza- tion, and this is seen to be one of the primary aims of leadership. Along with other researchers (e.g. Mathieu and Zajac, 1990), the leader is expected to influence employees’ affective re- sponses, e.g. affective commitment.

H3: Verbal consideration is positively related to

commitment.

Irritation is a state of mental exhaustion from which short daily breaks are no longer sufficient

for recovery, but recovery may nevertheless still be possible by means of longer breaks, changes in task assignments, changes of deadlines etc. (Mohr et al., 2006). A longitudinal study has shown that irritation is a precursor of depressive reactions in situations where people experience stress at work (Dormann and Zapf, 2002). The inclusion of this variable as an effectiveness criterion is in response to the fact that, particu- larly in Germany (where the figure is 25%), a not insignificant proportion of the working popula- tion retires before reaching legal retirement age due to health problems (Ruland, 2004), including stress-related illness. The task of leaders is and will increasingly be to prevent ‘exploiting’ of the mental and physical health of their followers and to protect followers from stressful over- or under- stimulation in their workload. Communication between a leader and her or his followers should in general lessen stress, since leaders receive up- to-date information concerning a follower’s workload and can then adapt task assignments to the follower’s present capacity and state of exhaustion. As Kalimo, Pahkin and Mutanen (2002) show in a longitudinal study, support from a supervisor is one of the most important work- related predictors of (low physiological) strain. Thus, one aspect of instrumental support could be the regulation of task assignments by taking into account the existing workload of followers. It can be assumed that verbal consideration gives followers the chance to let leaders know about their workload and gives leaders the chance to adjust task assignments. This, in turn, can prevent the development of irritation.

H4: Verbal consideration is negatively related to

irritation.

Communication between leader and follower can be seen as a means of clarifying the task that has to be done, how it has to be done and what the standard of fulfilment will be. Verbal consideration includes providing information whenever asked, giving feedback in a constructive way and encouraging followers to ask questions when they are uncertain. In this way role ambiguity or task unclarity for the follower is reduced, and the followers have a better chance of fulfilling the tasks in an effective way instead of relying on trial and error. Consequently, goal fulfilment should increase.

Leadership and Effectiveness in the Context of Gender 7

r 2007 British Academy of Management.

H5: Verbal consideration is positively related to

goal fulfilment.

Gender as a moderator between leadership and outcomes

There is a wide range of literature concerning the different evaluations of leaders depending on the sex of both the leader and the evaluator and also depending on whether the leader’s beha- viour is considered to be congruent or in- congruent with her or his gender role (e.g. Eagly and Karau, 2002; Eagly, Makhijani and Klonsky, 1992). Both women and men may be punished for violating gender stereotypes (Cost- rich et al., 1975). As stated above, verbal consideration is assumed to be closer to the communal and thus closer to the female stereo- type than to the agentic male stereotype (Cann and Siegfried, 1990). Women are rated as being more popular when they behave in a gender role congruent manner (Rudman, 1998; cf. Eagly and Karau, 2002; Sczesny et al., 2004). Women’s feminine leadership style could be rewarded because it is in line with gender stereotypic expectations. This positive effect may be miti- gated, however, by the mismatch that their role as a leader presents with respect to gender role expectations: even today, the leadership role is still conceived as being a male role (Schein, 2001; Sczesny, 2003), and this seems to be especially true in Germany (Schein and Mueller, 1992; Sczesny et al., 2004). In the context of this empirical research, it is assumed that verbal consideration utilized by women – being more in line with what followers expect from female leaders – will be pertinent to feeling comfortable and doing one’s job. No additional energy will be required to cope with incongruent leaders’ behaviour, and this may contribute to satisfac- tion and commitment and may also be helpful in preventing irritation. Goal fulfilment may thus be influenced in particular by female verbal consideration. Verbal consideration is expected to have a stronger relationship with these dependent variables when shown by female leaders.

H6: The relationship between verbal consideration

and effectiveness is moderated by leaders’ gender.

For female leaders the relationship is stronger than

for men.

Method

Sample

Participants from 46 German companies took part in the study. Among these were 10 insurance companies (22%), 11 housing companies (24%), 12 banks (26%) and 13 other organizations (28%) from a diverse range including the engineering, electrical and optical industries. Sixty-two female and 78 male leaders partici-

pated in this study. On average, they were 41.6 years of age (SD 5 8.3), had a tenure of 13.4 years (SD 5 8.6) and had worked for 9.8 years as a supervisor (SD 5 7.4). Most of these leaders came from the lowest level of organizational hierarchy (82%). Seventy-four leaders had a college of higher education diploma or a uni- versity diploma (approximately 53%). This re- flects that in Germany leaders from lower levels often have a rather academic background. It is assumed that – at this level – personal interaction between leaders and followers is more frequent than at higher levels, because at the lowest level of hierarchy leaders must foster and monitor the fulfilment of the tasks by their followers directly in order to be able to report to their own supervisors about reached targets. Four-hundred-and-fifty-five followers were

questioned (286 women, 169 men). On average they were 38.4 years of age (SD 5 10.2), had a tenure of 12.1 years (SD 5 8.4) and had worked for their supervisor for 3.5 years (SD 5 3.5). On average, three followers per leader took part (range 1–10 followers). Table 1 shows the distribution of the sex of leaders with that of followers.

Procedure

The followers’ questioning took place in small groups of between two and ten followers. In all groups one researcher was present in order to

Table 1. Distribution of followers’ sex across their leaders’ sex

Followers’ sex Leaders’ sex

Female Male

(S 5 62) (S 5 78)

Female n1 5 149 n3 5 137

Male n2 5 60 n4 5 109

8 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.

explain the purpose of the study, the confidenti- ality of data treatment and the procedure. In the case of questions, she or he provided standar- dized answers. The followers’ questionnaires contained questions about the verbal considera- tion of their leaders and about their own job satisfaction, commitment and irritation. The leaders were asked to answer their questionnaires at home. Their questionnaires included items concerning the performance of their work group. Filled-in survey sheets were sent back via regular mail. The leaders’ response rate was 76%.

Instruments

The data come from two data sources. The independent variable (verbal consideration) and three of the dependent variables are based on data from the followers. Data for one of the dependent variables, goal fulfilment, were based on the responses given by the leaders. In order to assess leadership, a scale referring

to verbal consideration was used (Mohr et al., 2004). To develop the scale, a special pre-study was conducted. Six teams were videotaped during a routine meeting of their department. Half of the teams had a female leader. The mean size of the groups was nine followers. About 11 hours of videotape were recorded. To analyse the data, a combination of time- and event-sampling meth- ods was chosen. A ten-minute sequence from the beginning, another ten-minute sequence from the middle, and one from the end of every meeting were taken. The sequences were evaluated by experts in leadership research with special focus on verbal and non-verbal interaction. Out of a broad pool of items, six were classified by experts as having a special focus on verbal consideration. These six items yielded a reliability of 0.81 (M 5 3.94, SD 5 0.61). A factor analysis (max- imum likelihood) revealed one factor accounting for 52.3% of the variance in the data. An example of an item is ‘My supervisor encourages me to speak up about matters that are important to me’. The scaling ranged from 1 (strongly disagree) to 5 (strongly agree) (see the Appendix for all items). Job satisfaction was assessed using an instru-

ment tested by Baillod and Semmer (1994). The instrument comprises eight items based on the research of Oegerli (1984). An example of one of these items is ‘I hope that my working situation

remains as good as it is at the moment’. The scale ranges from 1 (virtually never) to 7 (virtually always). The reliability for the eight items was 0.75 (M 5 5.35, SD 5 0.83). In order to assess commitment, the affective

commitment scale by Allen and Meyer (1990; German translation Schmidt, Hollmann and Sodenkamp, 1998) was used. The instrument focuses on the emotional bond a follower has with her or his organization. An example item reads ‘I would be very happy to spend the rest of my career in this organization’. The scale ranges from 1 (does not apply at all) to 7 (applies completely). The eight items yielded an alpha of 0.70 (M 5 5.18, SD 5 0.96). The scale ‘irritation’ (Mohr et al., 2006) was

included, which describes a state of mental exhaustion that occurs before the onset of mental illness. This scale was developed in the context of occupational stress research. Good statistical values have been demonstrated for many differ- ent samples (Mohr, Müller and Rigotti, 2005). Irritation is related to psychosomatic complaints (Garst, Frese and Molenaar, 2000). The long- itudinal data gathered by Grebner (2001) indicate that systolic blood pressure – an indicator of stress – is a predictor of irritation. The five-item scale yielded a reliability of 0.78

(M 5 2.08, SD 5 0.84) on a seven-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). An example item is ‘I get grumpy when others approach me’. Goal fulfilment was assessed with five items

referring to the fulfilment of goals set in advance. A sample item is ‘We reach the goals set by the top management’. The scale ranges from 1 (does not apply at all) to 5 (applies completely). Data for goal fulfilment had to be z-transformed for technical reasons (a change of the scale’s range during the study). The internal consistency was 0.79. Table 2 shows the correlations between all

variables.

Analyses

In order to test H1–H4, followers’ data were used. In order to test H5, the followers’ ratings of their leaders’ verbal consideration were averaged per group and correlated with the leaders’ rating of the groups’ performance. H6 was tested using moderated regression analyses.

Leadership and Effectiveness in the Context of Gender 9

r 2007 British Academy of Management.

Results

Preliminary results

In a first step, female and male leaders were compared with respect to their followers’ job satisfaction, commitment and irritation, as well as their teams’ goal fulfilment. Results of t tests do not show any significant difference between female and male leaders (for job satisfaction, female leaders M 5 5.34 (SD 5 0.85) versus male leaders M 5 5.34 (SD 5 0.82), t(442) 5 �0.10, p40.92; for commitment, female leaders M 5 5.16 (SD 5 1.02) versus male leaders M 5 5.20 (SD 5 0.91), t(432.52) 5 �0.52, p40.60; for irritation, female leaders M 5 2.10 (SD 5 0.84) versus male leaders M 5 2.06 (SD 5 0.84), t(451) 5 0.53, p40.59; for goal fulfilment, female leaders M 5 �0.18 (SD 5 1.02) versus male leaders M 5 0.12 (SD 5 0.97), t(136) 5 �1.77, p40.07). Moreover, differences between female and

male followers’ ratings were tested. Although female followers reported significantly higher job satisfaction than male followers (M 5 5.41 (SD 5 0.82) versus M 5 5.24 (SD 5 0.58); t(441) 5 2.01, po0.05), none of the remaining differences was significant. The intraclass correlation coefficient for fol-

lowers’ ratings of their leaders’ verbal considera- tion was calculated to gain a measure of inter- rater reliability (N 5 127 teams). A rather low intraclass correlation was expected, because verbal consideration has to be seen in the context of dyadic interaction, i.e. leaders may show different verbal consideration to each follower, and leaders may in turn be perceived differently by each follower. The ICC(1) result shows that 21% of a follower’s rating is due to group membership. This value is – as expected – rather

low, indicating that there is considerable varia- bility in the scores at the individual level. The correlations between verbal consideration

and the well-known constructs transformational leadership and leader–member exchange were evaluated as well. As expected, considerable intercorrelations exist (see Table 3), indicating that verbal consideration is related to a follower’s positive evaluation of the leader–follower rela- tionship. The correlations are substantial, but not so high that identity could be assumed between these constructs and verbal consideration. Nearly all correlations between verbal consideration and the subscales of leader–member exchange and transformational leadership are of similar

Table 2. Intercorrelations between independent and dependent variables

Variable N 5 455 N 5 140

1 2 3 4 5

1. Verbal consideration 1.0 0.29** 0.16** �0.19** 0.04 2. Job satisfaction 1.0 0.50** �0.23** 0.08 3. Affective commitment 1.0 �0.05 �0.08 4. Irritation 1.0 �0.07 5. Goal fulfilment 1.0

*po0.05; **po0.01.

Table 3. Intercorrelations between verbal consideration and

other leadership constructs (transformational leadership, leader–

member exchange)

Variable Verbal consideration

1. Transformational leadership 0.60**

1a. Intellectual stimulation 0.57**

1b. Idealized influence attributed 0.56**

1c. Idealized influence behaviour 0.51**

1d. Inspirational motivation 0.47**

1e. Individual consideration 0.54**

2. Leader–member exchange 0.57**

2a. Contribution 0.33**

2b. Professional respect 0.44**

2c. Affect 0.55**

2d. Loyalty 0.53**

Notes: N 5 455 followers. *po0.05; **po0.01. The internal consistencies for transformational leadership (Multifactor Leadership Questionnaire (MLQ), Bass and Avolio, 1995; German translation and modification, Felfe, 2006) and leader–member exchange (leader–member exchange multidimensional measure (LMX MDM), Liden and Maslyn, 1998; German translation, Paul and Schyns, 2002) were 0.94 and 0.90 respectively.

10 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.

strength. Thus, no further distinct information could be derived from the interrelations between verbal consideration and the aforementioned subscales. However, the correlation pattern con- firms that the scale has a positive connotation. The comparatively low correlation between verbal consideration and leader–member ex- change’s subscale contribution might be due to a specific of this subscale: it assesses the extent to which one exerts oneself for one’s leader. This extra-role behaviour might not necessarily mirror the quality of the leader–follower relationship.

Hypotheses testing

In order to test H1 (male and female leaders differ in verbal consideration), a t test was conducted. No differences were found (female leaders M 5 3.94 (SD 5 0.64) versus male leaders M 5 3.93 (SD 5 0.59); t(447) 5 0.17, p40.86). Thus, the first hypothesis was not confirmed. Hypotheses 2–5 were tested using correlation

analyses. As expected, verbal consideration was positively related to job satisfaction (r 5 0.29, po0.01; Table 4) and commitment (r 5 0.16, po0.01; Table 5), and negatively related to irritation (r 5 �0.19, po0.01; Table 6). In order to test H5 (verbal consideration is

positively related to goal fulfilment), the fol- lowers’ data were averaged for each leader and related to the leaders’ rating of goal fulfilment. Thus, the data set was reduced from 455

followers to 140 teams. In general, data from different sources and smaller samples provide lower correlations. Contrary to our assumption, there was no correlation at all (r 5 0.04, p40.62; Table 7). Before testing H6 (the relationship between

verbal consideration and effectiveness is moder- ated by leaders’ gender), the variables involved were centred (cf. Aiken and West, 1991), and an interaction term for verbal consideration and leaders’ gender was calculated. Regression analyses for each effectiveness criterion were conducted. The interaction verbal consideration � leaders’

gender was significant only with regard to followers’ irritation (po0.02), but not with regard to any of the other remaining dependent variables (job satisfaction p40.41; commitment p40.87; goal fulfilment p40.91). As illustrated in Figure 1, the relationship is negative for male leaders and almost zero for female leaders. This result indicates that women’s verbal considera- tion is not related to the group’s irritation, whereas men showing verbal consideration lower the level of the group’s irritation. This finding is contrary to our expectation.

Summary and discussion

This study focused on the effects of a communica- tion-focused leadership style: verbal consideration.

Table 4. Moderated regression on job satisfaction

Variable B SE B b

Constant �0.02 0.05 Verbal consideration (vc) 0.30** 0.05 0.29

Leaders’ gender (lg) 0.00 0.05 0.00

vc * lg 0.04 0.05 0.04

Notes: N 5 441, R 2 corr 5 0.08.

*po0.05; **po0.01.

Table 5. Moderated regression on commitment

Variable B SE B b

Constant �0.01 0.05 Verbal consideration (vc) 0.16** 0.05 0.16

Leaders’ gender (lg) 0.03 0.05 0.03

vc * lg 0.00 0.05 �0.01

Notes: N 5 448, R 2 corr 5 0.02.

*po0.05; **po0.01.

Table 6. Moderated regression on irritation

Variable B SE B b

Constant 0.01 0.05

Verbal consideration (vc) �0.20** 0.05 �0.20 Leaders’ gender (lg) �0.02 0.05 �0.02 vc * lg �0.12* 0.05 �0.12

Notes: N 5 447, R 2 corr 5 0.04.

*po0.05; **po0.01.

Table 7. Moderated regression on goal fulfilment

Variable B SE B b

Constant �0.01 0.09 Verbal consideration (vc) 0.05 0.09 0.0

Leaders’ gender (lg) �0.15 0.09 �0.15 vc * lg 0.01 0.09 0.01

Notes: N 5 138, R 2 corr 5 0.03.

*po0.05; **po0.01.

Leadership and Effectiveness in the Context of Gender 11

r 2007 British Academy of Management.

Its main interest was to examine whether or not this leadership behaviour is related to positive outcomes and whether or not leaders’ gender has a moderating effect on the relationship. Contrary to stated expectations, the results

showed that verbal consideration is not a leader- ship style that is shown especially by female leaders. Verbal consideration is indeed related to three

of the four effectiveness measures. The correla- tions – though significant – were low. This is not unexpected considering that outcome variables are usually multi-determined, as Arvey, Carter and Buerkley (1991) have shown for job satisfac- tion and Mathieu and Zajac (1990) for commit- ment. Also goal fulfilment depends not only on leaders’ communicative behaviour, but also on the resources available to fulfil a task (e.g. human resources, proper tools, sufficiency and good quality of raw materials). With regard to the interrelation between verbal consideration and goal fulfilment, the fact that data refer to different levels of aggregation must be taken into account. Furthermore, data have come from different sources: the followers rated their leaders’ verbal consideration, while the leaders rated their working groups’ goal fulfilment. A multi-source approach generally leads to lower correlations

compared to correlations between two constructs originating from the same source. Nevertheless the zero correlation between verbal consideration and goal fulfilment indicates that this is not simply a reduction caused by the elimination of common method and common source variance. This zero correlation leads to the assumption that verbal consideration has no direct influence on goal fulfilment. A moderating effect of leaders’ gender was

found only on the interrelation between verbal consideration and followers’ irritation. The direc- tion of this moderating effect was unexpected. It was assumed that the relationship would be stronger for female leaders than for male leaders. However, followers of male leaders using verbal consideration report less irritation. For female leaders, using verbal consideration had no effect on their followers’ irritation at all. Thus, con- cerning the moderating effect of leaders’ gender, it can be concluded that the same leadership behaviour attributed by the follower has different effects on followers’ irritation depending on who is showing this behaviour: if the behaviour is shown by a woman, it has no effect on followers’ irritation, but if it is shown by a man, a positive effect on followers’ irritation can be found. It should be borne in mind, however, that female and male leaders are equally effective in overall comparisons (see Preliminary Results section): followers of male and female leaders do not differ in irritation. This is true for the other dependent variables as well. Two questions worthy of further investigation arise from these results. (1) How can the negative interrelation between

verbal consideration and irritation only for male leaders be explained? Irritation was included because leaders should influence the amount of strain felt by their followers. Leaders have the responsibility of assigning tasks to their followers in an appropriate manner, so the follower is able to fulfil the task at a high quality level and over a long period of time without being too exhausted or no longer being able to replenish mental resources by normal breaks. Caution is advised about interpreting the results in a causal way because the study is cross-sectional. Nevertheless, selection processes can be excluded, such as male leaders in general having more mentally stable followers. The followers of the female leaders are not different in their general level of irritation. Furthermore, a selection effect would not help to

Verbal consideration

1.51.00.50.0–0.5–1.0–1.5–2.0–2.5

Ir ri ta

tio n

0.8

0.6

0.4

0.2

– 0.0

– 0.2

– 0.4

– 0.6

Leader's gender

male

female

Figure 1. Moderated regression on irritation

Note: Variables were centred before calculating the moder- ated regression analysis.

12 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.

explain why there is a negative correlation between verbal consideration shown by male leaders and their followers’ irritation. Perhaps our unexpected result can be seen as a ‘male advantage’. Followers may not expect their male leaders to show verbal consideration whereas such a behaviour may be expected as the norm for female leaders. (We found that female and male leaders do not differ in their level of verbal consideration. But this does not contradict the assumption above, because even when male and female leaders show the same amount of verbal consideration as rated by their followers, these followers’ expectations can be different towards male and female leaders.) When male leaders show verbal consideration, because it is unex- pected it may be regarded as particularly caring, and followers may therefore value it more than when female leaders behave this way. In such a way, the ‘caring male leader’ could become an extra source of mental strength in the workplace. (2) If verbal consideration shown by male

leaders is important for the low irritation of their followers, how should female leaders foster low irritation in their followers? Again, it is worth keeping in mind that followers of male and female leaders do not differ in their level of irritation. Thus, when it comes to female leaders they must use means other than verbal considera- tion to influence irritation. For example, fol- lowers may exhibit less competitive behaviour towards a female leader, or a female leader may show more concern for work–family conflict and help followers manage this more effectively etc. As interesting as these questions may be, they are tangential to the main focus of our study. Therefore, returning to the present study, limita- tions and need for future research have to be discussed before coming to some conclusions.

Limitations and future research

To begin with, it should be borne in mind that the presented results were based on leaders from low levels of organizational hierarchy. These low- level leaders were chosen because of the high amount of interaction that occurs with their followers. Verbal consideration captures the quality of interaction, and as such the results should not be generalized to leadership behaviour in general. This is particularly true for the puzzling result that only male leaders get the

credit of a positive relationship between verbal consideration and their followers’ mental health (i.e. irritation). Nevertheless, the results deserve further investigation because interacting with followers constitutes a large part of leadership behaviour at low levels. The study design was cross-sectional. The

possibility that followers’ low irritation, i.e. low emotional exhaustion, evokes leaders’ verbal consideration cannot be excluded. Then why should this happen only with male leaders? And would it not be much more plausible that a leader shows concern when her or his follower shows signs of exhaustion? It is also possible that high goal fulfilment by the group evokes verbal consideration. Verbal consideration could be seen as a kind of reward for high goal fulfilment. It makes sense to assume that verbal considera- tion can be the beginning of a self-supporting feedback loop: it helps to reach good goal fulfilment and this, in turn, leads to even more appreciation of the follower by further verbal consideration from the leader. This idea can only be proved with additional longitudinal data. A follow-up study should make use of different

outcome criteria. Although a broad range of outcomes was examined, ones with a stronger interrelation with the communicative behaviour of a leader could have been chosen. Outcomes such as role conflict, problem-solving and social stressors may be more related to the followers’ experience of the leaders’ verbal behaviour. In addition, a different, more objective performance assessment would be useful in a further study: The assessment of goal fulfilment based on leaders’ judgements is prone to bias, as leaders may refrain from judging themselves and their group as ineffective. Furthermore, it would be worthwhile ascer-

taining whether or not followers rate verbal consideration as a feminine or communal leader- ship style and to check the assumption that showing a feminine leadership style leads to more credit for male than female leaders. Pounder and Coleman (2002) emphasize the

necessity to control for the influence of national culture in leadership research. Schein (2001), Schein and Mueller (1992) and Sczesny et al. (2004) showed that, especially in Germany, leadership is still gender stereotyped (even when ratings are given by women). Thus, a further step could be to control for gender role attitudes when

Leadership and Effectiveness in the Context of Gender 13

r 2007 British Academy of Management.

analysing how followers perceive their leaders (cf. Grinnell, 2002).

Conclusion and implications

The results allow the conclusion that verbal consideration is an important leadership beha- viour and should thus be taken seriously. A lot of leadership training which focuses on communica- tion with followers already appreciates its pro- minent role. The concept of verbal consideration is related to everyday interactive behaviour, not specifically for managing crisis situations. It can be directly observed by followers. Thus, it may be more accessible for training than highly aggre- gated concepts such as respect, trust etc. Additionally, one can argue that male and

female leaders achieve positive consequences to the same extent, but that the way in which they are reached seems to be different. Finding out more about the different ways female and male leaders achieve the same effect may be a challenge for further research and may give support for more specific and thus more effective leadership training. It seems that many roads lead to Rome, and so it still remains to find out which way may be more appropriate for male leaders and which one for female leaders.

Appendix: Items of the verbal consideration scale

1. My supervisor asks for the views of her or his subordinates.

2. My supervisor encourages me to speak up about matters that are important to me.

3. My supervisor precisely notes my contributions. 4. My supervisor gives me opportunities to ask

questions. 5. My supervisor constructively criticizes me. 6. My supervisor willingly explains when I ask.

Note: The items were translated into English for this manu-

script. Reported results refer to a German version of the scale.

References

Aiken, L. S. and S. G. West (1991). Multiple Regression:

Testing and Interpreting Interactions. Thousand Oaks, CA:

Sage.

Allen, N. J. and J. P. Meyer (1990). ‘The measurement and

antecedents of affective, continuance and normative commit-

ment to the organization’, Journal of Occupational Psychol-

ogy, 63, pp. 1–18.

Arvey, R. D., G. W. Carter and D. K. Buerkley (1991). ‘Job

satisfaction: dispositional and situational influences’, Inter-

national Review of Industrial and Organizational Psychology,

6, pp. 359–383.

Baillod, J. and N. Semmer (1994). ‘Fluktuation und Berufsver-

läufe bei Computerfachleuten [Turnover and career paths of

computer specialists]’, Zeitschrift für Arbeits- und Organisa-

tionspsychologie, 38, pp. 152–163.

Bakan, D. (1966). The Duality of Human Existence. Chicago,

IL: Rand McNally.

Bass, B. M. (1990). Stogdill’s Handbook of Leadership: Theory,

Research and Managerial Applications. New York, NY: Free

Press.

Bass, B. M. and B. Avolio (1995). MLQ Multifactor Leadership

Questionnaire. Technical Report. Redwood City, CA: Mind

Garden.

Biernat, M. (2003). ‘Toward a broader view of social

stereotyping’, American Psychologist, 58, pp. 1019–1027.

Cann, A. and W. D. Siegfried (1990). ‘Gender stereotypes and

dimension of effective leader behavior’, Sex Roles, 23, pp.

412–419.

Costrich, N., J. Feinstein, L. Kidder, J. Maracek and L. Pascale

(1975). ‘When stereotypes hurt: three studies of penalties for

sex-role reversals’, Journal of Experimental Social Psychol-

ogy, 11, pp. 520–530.

DeGroot, T., D. S. Kiker and T. C. Cross (2001). ‘A meta-

analysis to review organizational outcomes related to

charismatic leadership’, Canadian Journal of Administrative

Sciences, 17 (4), pp. 356–371.

Diekman, A. B. and A. H. Eagly (2000). ‘Stereotypes as

dynamic constructs: women and men of the past, present and

future’, Personality and Social Psychology Bulletin, 26, pp.

1171–1181.

Dormann, C. and D. Zapf (2002). ‘Social stressors at work,

irritation, and depressive symptoms: accounting for unmea-

sured third variables in a multi-wave study’, Journal of

Occupational and Organizational Psychology, 75, pp. 33–58.

Eagly, A. H. and S. J. Karau (2002). ‘Role congruity theory of

prejudice toward female leaders’, Psychological Review, 109

(3), pp. 573–598.

Eagly, A. H., M. C. Johannesen-Schmidt and M. L. van Engen

(2003). ‘Transformational, transactional, and laissez-faire

leadership styles: a meta-analysis comparing women and

men’, Psychological Bulletin, 129 (4), pp. 569–591.

Eagly, A. H., S. J. Karau and M. G. Makhijani (1995). ‘Gender

and the effectiveness of leaders: a meta-analysis’, Psycholo-

gical Bulletin, 117 (1), pp. 125–145.

Eagly, A. H., M. G. Makhijani and B. G. Klonsky (1992).

‘Gender and the evaluation of leaders: a meta-analysis’,

Psychological Bulletin, 111 (1), pp. 3–22.

Felfe, J. (2006). ‘Validierung einer deutschen Version des

‘‘Multifactor Leadership Questionnaire’’ (MLQ 5 X Short)

von Bass und Avolio (1995) [Validation of a German version

of the ‘‘Multifactor Leadership Questionnaire’’]’, Zeitschrift

für Arbeits- und Organisationspsychologie, 50 (2), pp.

61–78.

Fleishman, E. A. (1957). ‘Leader behavior description for

industry’. In R. M. Stogdill and A. E. Coons (eds), Leader

14 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.

Behavior: Its Description and Measurement, pp. 103–119.

Columbus, GA: Columbus State University.

Garst, H., M. Frese and P. C. M. Molenaar (2000). ‘The

temporal factor of change in stressor–strain relation-

ship: a growth-curve model on a longitudinal study in

East Germany’, Journal of Applied Psychology, 85, pp.

417–438.

Gerstner, C. R. and D. V. Day (1997). ‘Meta-analytic review

of leader–member exchange theory: correlates and con-

struct issues’, Journal of Applied Psychology, 82 (6), pp.

827–844.

Grebner, S. (2001). ‘Stress at work, well-being, blood pressure

and cortisol: two field studies’. Unpublished doctoral

dissertation. University of Berne, Switzerland.

Grinnell, J. P. (2002). ‘Effects of leaders’ and evaluators’ sex-

role stereotyping of charismatic leaders’, Psychological

Reports, 91, pp. 1247–1252.

Guest, R. H. (1956). ‘Of time and the foreman’, Personnel, 32,

pp. 478–486.

Halverson, S. K., S. E. Murphy and R. E. Riggio (2004).

‘Charismatic leadership in crisis situations: a laboratory

investigation of stress and crisis’, Small Group Research, 35,

pp. 495–514.

Kalimo, R., K. Pahkin and P. Mutanen (2002). ‘Work and

personal resources as long-term predictors of well-being’,

Stress and Health, 18 (5), pp. 227–234.

Lent, R. H., H. A. Aurbach and L. S. Levin (1971). ‘Predictors,

criteria, and significant results’, Personnel Psychology, 24, pp.

519–533.

Liden, R. C. and J. M. Maslyn (1998). ‘Multidimensionality of

leader–membership exchange: an empirical assessment

through scale development’, Journal of Management, 24,

pp. 43–72.

Lowe, K. B., K. G. Kroeck and N. Sivasubramaniam (1996).

‘Effectiveness correlates of transformational and transac-

tional leadership: a meta-analytic review of the MLQ

literature’, Leadership Quarterly, 7 (3), pp. 385–425.

Mathieu, J. E. and D. M. Zajac (1990). ‘A review and meta-

analysis of the antecedents, correlates, and consequences of

organizational commitment’, Psychological Bulletin, 108 (2),

pp. 171–194.

Meyer, J. P. (1997). ‘Organizational commitment’. In C. L.

Cooper and I. T. Robertson (eds), International Review of

Industrial and Organizational Psychology, Vol. 12, pp.

176–228. New York: Wiley.

Meyer, J. P., N. J. Allen and I. R. Gellatly (1990). ‘Affective

and continuance commitment to the organization: evaluation

of measures and analysis of concurrent and time-

lagged relations’, Journal of Applied Psychology, 75 (6), pp.

710–720.

Mintzberg, H. (1973). The Nature of Managerial Work. New

York: Harper and Row.

Mohr, G., A. Müller and T. Rigotti (2005). ‘Normwerte

der Skala Irritation: Zwei Dimensionen psychischer Bean-

spruchung [Standardization data of the irritation scale:

two dimensions of mental strain]’, Diagnostica, 51, pp.

12–20.

Mohr, G., H.-J. Wolfram, B. Schyns and T. Paul (2004). ‘Verbales

Führungsverhalten aus Sicht der MitarbeiterInnen (VFV-MA)

[Verbal consideration of the leader from the follower’s

perspective]’. In A. Glöckner-Rist (ed.), ZUMA-Informations-

system. Elektronisches Handbuch sozialwissenschaftlicher Er-

hebungsinstrumente, ZIS Version 8.00. Mannheim: Zentrum

für Umfragen, Methoden und Analysen.

Mohr, G., A. Müller, T. Rigotti, Z. Aycan and F. Tschan

(2006). ‘The cross-cultural assessment of psychological strain

in work contexts: concerning the measurement equivalency of

11 adaptations of the irritation scale’, European Journal of

Psychological Assessment, 22 (3), pp. 198–206.

Oegerli, K. (1984). ‘Arbeitszufriedenheit. Versuche einer

qualitativen Bestimmung [Job satisfaction. A qualitative

approach]’. Unpublished doctoral dissertation. University

of Berne, Switzerland.

Parsons, T. and R. F. Bales (1955). Family, Socialization and

Interaction Process. Glencoe, IL: Free Press.

Paul, T. and B. Schyns (2002). ‘Mehrdimensionale Skala zur

Erfassung des Leader–Member-Exchange (LMX-MDM

nach Liden and Maslyn, 1998) – Übersetzung [Multidimen-

sional scale for the assessment of leader–member exchange

(LMX MDM according to Liden and Maslyn, 1998) –

translation]’. In A. Glöckner-Rist (ed.), ZUMA-Informa-

tionssystem. Ein elektronisches Handbuch sozial-wissenschaf-

tlicher Erhebungsinstrumente, Version 6.00. Mannheim:

Zentrum für Umfragen, Methoden und Analysen.

Podsakoff, P. M., S. B. MacKenzie and W. H. Bommer (1996).

‘Transformational leader behaviors and substitutes for

leadership as determinants of employee satisfaction, commit-

ment, trust, and organizational citizenship behaviors’,

Journal of Management, 22 (2), pp. 259–298.

Pounder, J. S. and M. Coleman (2002). ‘Women – better leaders

than men? In general and educational management it still ‘‘all

depends’’’, Leadership and Organization Development Jour-

nal, 23, pp. 122–133.

Rudman, L. A. (1998). ‘Self-promotion as a risk factor for

women. The costs and benefits of counterstereotypical

impression management’, Journal of Personality and Social

Psychology, 74 (3), pp. 629–645.

Ruland, F. (2004). ‘Aktuelle Ergebnisse zu den Wirkungen der

bisherigen Rentenreformen auf den Übergang von der

Erwerbs- in die Ruhestandsphase [The latest results on the

influence of the pension reform on the transition from

working life to retirement]’, Deutsche Rentenversicherung, 57,

pp. 34–53.

Schein, V. E. (2001). ‘A global look at psychological barriers to

women’s progress in management’, Journal of Social Issues,

57, pp. 675–688.

Schein, V. E. and R. Mueller (1992). ‘Sex role stereotyping and

requisite management characteristics: a cross-cultural look’,

Journal of Organizational Behavior, 13, pp. 439–447.

Schmidt, K.-H., S. Hollmann and D. Sodenkamp (1998).

‘Psychometrische Eigenschaften und Validität einer

deutschen Fassung des ‘‘Commitment’’ – Fragebogens von

Allen und Meyer (1990) [Psychometric properties and

validity of a German version of Allen and Meyer’s (1990)

commitment questionnaire]’, Zeitschrift für Differentielle und

Diagnostische Psychologie, 19, pp. 93–106.

Sczesny, S. (2003). ‘A closer look beneath the surface: various

facets of the think-manager think-male stereotype’, Sex

Roles, 49 (7/8), pp. 353–363.

Sczesny, S., J. Bonsak, D. Neff and B. Schyns (2004). ‘Gender

stereotypes and the attribution of leadership traits: a cross-

cultural comparison’, Sex Roles, 51, pp. 631–645.

Smith, P. (1976). ‘Behavior, results, and organizational effec-

tiveness: the problem of criteria’. In M. D. Dunnette (ed.),

Leadership and Effectiveness in the Context of Gender 15

r 2007 British Academy of Management.

Handbook of Industrial and Organizational Psychology, pp.

745–775. Chicago, IL: Rand McNally.

Trinidad, C. and A. H. Normore (2005). ‘Leadership and

gender: a dangerous liaison?’, Leadership and Organization

Development Journal, 26, pp. 574–590.

Watzlawick, P., J. H. Beavin and D. D. Jackson (1969). Human

Communication: Forms, Disturbances, Paradoxes. Oxford:

Hans Huber.

Witte, E. (1995). ‘Effizienz der Führung [Efficiency of leadership]’.

In A. Kieser, G. Reber and R. Wunderer (eds), Handwörter-

buch der Führung, pp. 264–276. Stuttgart: Schäffer-Poeschel.

Yukl, G. (1999). ‘An evaluation of conceptual weakness in

transformational and charismatic leadership theories’, Lea-

dership Quarterly, 10, pp. 285–305.

Yukl, G. (ed.) (2002). Leadership in Organizations, 5th edn.

Upper Saddle River, NJ: Prentice Hall.

Dr Gisela Mohr is Professor for Work and Organizational Psychology and Director of the Institute of Psychology II, University of Leipzig, Germany. Her research interests include stress at work, unemployment, job insecurity, women’s employment, leadership, and the effect of psychological contracts on health.

Dr Hans-Joachim Wolfram works as a research fellow at the Institute of Psychology II, University of Leipzig, Germany. His main areas of interest are work and organizational psychology, social psychology, and social psychology perspectives within organizational settings.

16 G. Mohr and H.-J. Wolfram

r 2007 British Academy of Management.