RESEARCH AND PRACTICE
IN HUMAN RESOURCE MANAGEMENT

Highlight, copy & paste to cite:

Chia, H., Landau, J. & Ong, J. S. (2000). Predictors of Voice: The Moderating Effect of the General Economic Climate, Research and Practice in Human Resource Management, 8(2), 3-28.

Predictors of Voice: The Moderating Effect of the General Economic Climate

Ho-Beng Chia, Jacqueline Landau & Jane Swad-Wen Ong

Abstract

This study investigated the moderating effect of economic conditions (economic growth vs. recession scenario) on the relationship between employee willingness to voice and two predictors, cognitive style preference for adaptation or innovation, and the supervisor’s competence as a voice manager. The results showed that cognitive style preference was only related to willingness to voice in the recession scenario, with innovators expressing more willingness to voice than adaptors. Supervisor’s competence as a voice manager was positively related to willingness to voice regardless of the economic situation, but the relationship was much stronger in the recession scenario.

Introduction

Both academic and trade publications are replete with articles discussing the benefits of organizational learning. One of the conditions that may facilitiate organizational learning is a process for inquiry (Argyris & Shon, 1978; 1996). According to Argyris and Shon (1996) it is the “detection of error ... that triggers awareness of a problematic situation and sets in motion the inquiry aimed at correcting the error” (p. 31). The organizational learning literature is a collection of literature from many domains, many programs have been advocated, such as empowerment, action learning, and employee involvement. However, the fundamental mechanism that allows these programs to work is still the willingness of employees to express their dissatisfaction and uncover errors at the workplace so that changes in work or procedures can take place. One mechanism for bringing errors to the attention of those who can make or sanction corrections is employee voice.

The term “voice” was first identified by Hirshman (1970) in his theory of how people respond to declining conditions in organization or nations. He argued that employees can respond to dissatisfying conditions by either voicing or exiting. The choice between exit and voice strategies depends on loyalty to the work organization and the belief in the possibility of improvement in the dissatisfying conditions. Although much of Hirshman’s ideas focus on the conflict of interest between the organization and the individual, we argue that both parties also have common interests such as economic survival. Instead of assuming that the conflicts of interest are unresolvable, recent authors have begun to look at the possibility of harnessing employee voice and encouraging constructive conflict as a means for change and improvement.

Rusbult, Farrel, Rogers, and Mainous (1988), building on Hirshman’s theory, defined voice as “actively and constructively trying to improve conditions through discussing problems with a supervisor or coworkers, taking action to solve problems, suggesting solutions, seeking help from an outside agency like a union, or whistle-blowing.” (p. 601). Voice alerts management to potential problems, and helps management recognize its failures and make changes critical for organizational survival. This is particularly important during an economic recession. However, in times of economic rccession and organizational retrenchment, the risk of job loss often creates a climate of fear and distrust (Channon, 1996; Slater, 1999). Therefore, voice may be repressed when it is most needed, either through self-censorship or subtle organizational messages not to “rock the boat.”

Several studies in recent years have attempted to identify those factors which might encourage employee voice (e.g. Janssen, de Vries & Cozijinsen, 1998; Lee & Jablin, 1992; LePine & Van Dyne, 1998; Parker, 1993; Saunders, Sheppard, Knight & Roth, 1992; Spencer, 1986). Withey and Cooper (1989) propose that employees consider three variables in deciding whether to voice; (1) the cost of the action; (2) the efficacy of the action; and (3) the attractiveness of the setting in which the action occurs. In this study we make no attempt to test a complete model of employee voice, but instead we focus on a variable which influences the cost of the action, the economic environment.

This study takes place in Singapore. Until the summer of 1997, Singapore had enjoyed a robust economy with one of the lowest rates of unemployment in the world for several years. In the summer of 1997 the economy of Thailand collapsed, followed by Indonesia, Malaysia and Hong Kong. Although Singapore was not affected as badly as other Southeast Asian countries, the unemployment doubled to 4.5 percent by the end of 1998 as organizations restructured and retrenched (Dolven, 1998, 1999). This sudden reversal provided a perfect opportunity to compare willingness to voice during economic growth and recession.

Employees may voice their dissatisfaction in many different ways in terms of both mode and content. For examples, in terms of mode, employees may voice dissatisfaction to their co-workers, their supervisor, an union representative or through a formal organizational grievance procedure. In terms of content, voice may vary in terms of whether it is directed at changing individual circumstances, the group environment, or the organisation as a whole. In this study we limited our investigation to only one type of voice, directly expressing dissatisfaction to a supervisor. To date, in the organizational literature there has been more research on this form of voice than any other.

We examined how the economic recession influences the relationship between the willingness to voice to supervisors and two factors: (1) the employee’s personality; and (2) the quality of the supervisor as a voice manager. We choose these variables because organizational behavior is a function of both what the employee brings with him or her to the work setting, and what happens to the employee within that setting (Britt, Boniecki, Vescio & Biernat, 1996; Forehand & Gilmer, 1964; Lewin, 1936; Walsh, Craik, & Price, 1992). Also, both variables have been found to be important predictors of voice in previous studies (Janssen et al., 1998; LePine & Van Dyne, 1998; Parker, 1993; Saunders et al., 1992).

Furthermore, willingness to voice to supervisors is pertinent to main stream management issues of participation and leadership. Often, employees and subordinates are only willing to participate in debates on organizational issues that are perceived as within the accepted boundaries of superiors. But for a truely participative environment, subordinates must feel comfortable voicing sensitive issues as well. Otherwise, there will be an asymmetry of information flowing upwards - only “sanctioned issues” are discussed, while important and “tabooed issues” are left unquestioned. At a time of rapid change, when old business models are quickly becoming obsolete, the questioning of “sacred cows” and previously taken for granted ways of doing business must be welcomed. This requires new ways of leading and encouraging employees to voice their concerns.

Voice and Adaptation/Innovation

Only a handful of studies have examined the relationship between personality variables and willingness to voice (e.g. Janssen et al. 1998; LePine & Van Dyne, 1998; Parker, 1993). In this study we investigate cognitive style preference for adaptation or innovation as a determinant of voice. According to Kirton (1976) individuals differ in terms of their cognitive styles when they engage in creativity, problem solving, and decision-making. Some individuals tend to be adaptors, while others are innovators. Adaptors tend to be conservative, operating within the confines of generally accepted guidelines. Their behaviors and solutions reinforce the current guidelines and they focus on refining existing processes. On the other hand, innovators see the guidelines themselves as part of the problem. They are “risk-takers” whose solutions tend to incorporate new and often untried processes, such that they threaten or even bring about a change in the guidelines themselves (Janssen et al., 1998).

Janssen et al. (1998) suggest that individual differences in cognitive style preferences for adaptation-innovation may explain why some employees responding to new stimuli tend to advocate conventional ideas for organizational change and fail to see opportunities outside the existing framework. On the other hand, others compulsively challenge the currently held paradigm by presenting novel, revolutionary ideas to their supervisor. Prior research provides empirical evidence that innovators demonstrate more readiness for change and produce more original ideas than adaptors (e.g. Argyris & Schon, 1978; Clapp & de Ciantis, 1989; Haywood & Everett, 1983; Keller & Holland, 1978). Therefore, we hypothesize that:

H1: Innovators will report more willingness than adaptors to voice to their supervisors.

Supervisors as Voice Managers

Perhaps the most important recipient of employee voice is the employee’s immediate supervisor. Both Saunders et al. (1992) and Janssen et al. (1998) found that employees’ willingness to voice is facilitated by supervisors who are effective voice managers. Saunders et al. (1992) identified two different dimensions of supervisory voice management, approachability and responsiveness.

Approachability refers to the extent employees feel they can bring their concerns to their supervisors without being penalized. Unfortunately, previous studies have shown that employee voice often extracts a great cost (Elliston, Keena, Lockhart & Schaick, 1985; Westin, 1981). For example, Lewin (1987, 1992), Lewin and Peterson (1988), and Feuille & Delaney (1993), all found that employees who filed grievances had lower performance appraisal ratings, lower promotion rates, and higher involuntary turnover in the period following grievance settlement than comparably matched employees who were not directly involved in grievance activity. Rowe and Baker (1984) found that professionals were reluctant to complain to their superiors because they feared that their career progress would be stymied. When the costs of voicing are perceived to be high, employees will remain silent. When supervisors are approachable, the costs of voicing significantly decrease.

Responsiveness, the second characteristic identified by Saunders et al. (1992), refers to the extent to which supervisors are prompt and willing to take action to deal with the problems and suggestions voiced by employees. Supervisory responsiveness increases the efficacy of voice. Employees are more likely to voice if they believe they might be successful in influencing the organization to change (Hirshman, 1970; Withey & Cooper, 1989), the organization values their contributions (Eisenberger, Fasolo & LaMastro, 1990) and those in authority take their opinions into account when decisions are made (Parker, 1993). If supervisors are unresponsive, the credibility of the organization’s voice mechanisms are likely to decrease, creating distrust and silence among employees. Therefore, we hypothesize:

H2: Employees who perceive that their supervisors are approachable and responsive will express more willingness to voice than employees who perceive that their supervisors are unapproachable and unresponsive.

Voice and the Economy

During a recession, the economy contracts, and many organizations respond by halting expansion plans, or by restructuring and laying-off employees to reduce excess capacity and cost. This results in fewer jobs, and increased competition for those jobs that remain. Those who lose their jobs may have great difficulty finding alternative employment.

When employees voice their dissatisfaction, they always face the possibility of retributive action. Their dissent may be interpreted as disloyalty, and they may be criticized as not being “team players.” Management may punish them or force them to leave. Thus, the expression of voice carries with it a probabilistic cost. During a recession, when there are fewer alternative employment opportunities outside the organization, the probabilistic cost of voicing increases (Rusbult et. al., 1988, Withey & Cooper, 1989).

In addition, restructuring and downsizing of organizations during a recession often erodes employee trust and morale (Channon, 1996; Doherty, Band & Vinnicombe, 1996; Mishra & Spreitzer, 1998). Employees are less certain about how dissension or expression of dissatisfaction will be received by management. Thus, they may become reluctant to share information (Kennedy, 1996), and afraid to talk about their personal work concerns. However, some studies have shown that not all employees respond to uncertainty and/or downsizing with fear and distress (Mishra & Spreitzer, 1998). Some survivors see such crises as opportunities for personal growth (Isabella, 1989), and therefore, increase their efforts to solve problems and take greater initiatives to change the organization for the better. We argue that during a recession and in the face of organizational distress, the willingness of employees to voice may partially depend on whether they are innovators or adaptors, and whether their supervisors are approachable and responsive voice managers.

Studies have shown that adaptors are less inclined to take risks and seek sensation, more dogmatic and inflexible, and less tolerant of ambiguity than innovators (Carne & Kirton, 1982; Goldsmith, 1986; Gryskiwicz, 1980; Kirton, 1976; Kirton & de Cinatis, 1986). The low tolerance of uncertainty and risk adversity of adaptors suggest that they would succumb to the fear of unemployment, and thus voice less during a crisis, to avoid offending anyone, in particular their supervisors.

On the other hand, the greater appetite for risk and tolerance for ambiguity of innovators suggest that they would be willing to voice despite potential repercussions. Therefore, innovators might continue to voice regardless of the state of the economy. In fact, innovators may view the recession as an opportunity to voice even more, as colleagues and managers might be more receptive to their ideas and suggestions for change. During good times, even if the business is not as efficient as it could be, as long as it is profitable, adaptive-oriented employees and managers might be reluctant to change. Janssen et. al. (1998) stated that innovators tend to challenge the regular patterns and commonly accepted assumptions by voicing extraordinary suggestions for organizational change even in periods where there is no obvious crisis. Consequently, innovators are sometimes seen as the cause of discord and friction and viewed with distaste by more adaptively oriented colleagues and managers (Kirton, 1976). However, in bad times, management is in dire need of novel theories and practices in order to survive, and may be more open-minded and receptive to change (Wong, 1994). Therefore, innovators may view a recession as an opportunity for acceptance of their ideas and would maintain, if not increase, their use of voice. We hypothesize:

H3: Innovators will be willing to voice as much, if not more, during a recession than during a period of economic growth.

H4: Adaptors will be less willing to voice during a recession than during a period of economic growth.

The approachability and responsiveness of the employee’s immediate supervisor could also influence whether employees voice more or less during a recession than during a period of economic growth. If supervisors are unapproachable and unresponsive during a recession, the cost of voicing would increase and the efficacy of voicing would decrease. On the other hand, if supervisors are skillful at handling voice, the efficacy of voice might increase as employees recognize the need for change and improvement to ensure the survival of the organization and their jobs. Therefore, we hypothesize:

H5: Employees who perceive that their supervisors are approachable and responsive voice managers will be as much, if not more willing to voice during a recession than during economic growth.

H6: Employees who perceive that their supervisors are unapproachable and unresponsive will be less willing to voice during a recession than during economic growth

Method

Sample and Procedure

Data for this study were collected from a survey distributed to 1500 alumni from a large university in Singapore. Only alumni with at least two years of work experience were included in the sample so that they would have been members of the workforce both during the economic boom and economic recession. The sample was stratified by gender and by year of graduation. Sixty graduates (30 male and 30 female) were randomly selected for each graduation hatch from 1970 to 1994.

A questionnaire was mailed to the home address of each of the alumni. A letter accompanying the survey explained that they had been randomly selected to participate in a study about responses to dissatisfaction. They were assured that their responses would be confidential, and were asked to return the survey directly to the researchers using the enclosed addressed, postage paid envelopes. Reminder letters were sent out three weeks after the initial mailing.

One hundred thirty-eight completed questionnaires were returned for a response rate of 9.2%. Although this response rate is quite low by Western research standards, it is typical of survey research in Asia. According to Yuen (1995) the low response rate in Asia is attributed to three factors: (1) Asians are not as accustomed to survey research as North Americans; (2) social research is not as popular in Asia as in North America; and (3) Asians have a larger private space than Americans, and thus, are not as willing to share personal information with a stranger.

The sample was 49% female, 51% male, the average age was 38, the average work experience was 13.9 years, the average time in occupation was 8.73 years, and the average time in the firm was 7 years. Respondents did not differ from non-respondents in terms of gender proportions (respondents: 51.45% vs. mailing list: 50%) and age (respondents: 37.8 years vs. mailing list: 38 years).

Measures

Dependent Variable

Willingness to voice to a supervisor was measured using an eight item Likert type scale developed by Saunders et al., (1992). The specific items are listed in Table 1. Response alternatives ranged from (1) very unlikely to (5) very likely. Saunders et al. (1992) reported Cronbach alphas of .80 for one sample, and .89 for the other.

To corroborate Saunder’s finding of a single dimension, we analyzed the items using principal component analysis with varimax rotation. The survey responses for the recession and economic growth scenario were analysed seperately. Results showed the presence of two factors rather than one (see Table 1). The first factor, explaining 50.5% of the variance in the recession scenario and 47.4% of the variance in the economic growth scenario, included concerns about personal manners (Voicing about Personal Matters). The second factor, explaining an additional 12.8% of the variance in the recession scenario and 13.4% of the variance in the economic growth scenario, included concerns about others (Voicing about Others). The items for each dimension were summed and averaged for subsequent statistical analysis.

Table 1
Factor Loadings for Willingness to Voice Scale
Recession Economic Growth
Personal matters Co-workers Personal matters Co-workers
Eigen Value 4.04 1.03 3.79 1.07
Variance Explained (%) 50.51 12.84 47.38 13.36
Cronbach Alpha .83 .76 .80 .74
If something about the policies and procedures of your organization irritated or bothered you, how likely would you be to speak to your immediate supervisor about it? .809 .258 .789 .211
If you had a concern about something to do with your job, how likely would you be to speak to your immediate supervisor about it? .771 .009 .746 .070
How likely would you be to speak to your immediate supervisor about a better way to do your job? .695 .281 .680 301
If your immediate supervisor did something to irritate or bother you, how likely would you be to speak to him or her about it? .691 .229 .704 .229
When something at work irritates (bothers) you, how likely are you to speak to your immediate supervisor about it? .685 .353 .623 .414
How likely would you be to speak to your immediate supervisor about a concern over how another worker was doing his or her job? .138 .891 .127 .864
If you knew a co-worker was not honest, how likely would you be to speak to your immediate supervisor about it? .248 .757 .266 .698
If an employee from another department did something to irritate or bother you, how likely would you be to speak to your immediate supervisor about it? .404 .670 .262 .762

Participants were asked to respond to these items twice. The first time they were instructed to indicate how they would respond given current economic conditions (“now”). The second time around they were asked to indicate how they would respond if the economy was as it had been prior to the Asian financial crisis (“economic growth”). The Cronbach alphas for responses to the “now” scenario for factor 1 and factor 2 were .83 and .76 respectively. Those for response to the “economic growth” scenario for factor 1 and factor 2 were .80 and .74 respectively.

To check if participants discriminated between the two economic states “now” and “economic growth” when they answered the questionnaire, participants were asked to indicate the quality of job alternatives available to them under both conditions. The two items used were: “How difficult would it be for you to find ajob with another employer with approximately the same income and benefits as you have now?” and “How confident are you of finding a satisfactory job if you were to quit this job?” Response alternatives ranged from (1) strongly disagree to (5) strongly agree. The items were summed and subjects reported there were significantly fewer quality job alternatives under the “now” scenario than under the “Economic growth” scenario (t(1,142)=17.45, p<.000).

Independent Variables

Adaptation-Innovation. Employee cognitive style preferences for adaptation-innovation were assessed using Kirton’s (1976) Adaptation-Innovation Inventory (KAI scale). The scale places people on a continuum - adaptors would have a low score whereas innovators would have a high score. The KAI scale consists of 32 items measuring three personality components: originality (13 items), conformity (12 items), and efficiency (7 items). Originality is closely related to Roger’s (1959) creative personality - characterized by little respect for conventions, obsessive playing with ideas, and a high need for social acceptance of these ideas. Efficiency reflects Weber’s (1948) typically bureaucratic person who is precise, reliable, and disciplined. Finally, conformity mirrors Merton’s (1957) conformist who has proper respect for authority and group rules. Kirton (1976) argues that adaptive personalities score high, while innovative personalities score low on efficiency and conformity. Innovators score higher than adaptors on originality. Respondents were asked to indicate how similar or dissimilar they were to the behavior described by each of 32 statements. Response alternatives ranged from (1) very similar to (5) very dissimilar. The KAI has been shown to have adequate internal and test-retest reliability (Kirton, 1976, 1987), be unrelated to social desirability (Kirton, 1976; Goldsmith & Matherly, 1986), and be valid across different cultures (Holland, 1987). In this study the reliability coefficient, Cronbach alpha was .82.

Supervisor as Voice Manager. Perceived supervisor’s ability to competently manage voice was measured using Saunders et al.’s (1992) 14-item scale. The scale consists of seven items measuring approachability and seven items measuring responsiveness. (see Appendix 1). Response alternatives ranged from (1) strongly disagree to (5) strongly agree. The Cronbach alpha for the scale was .91.

Demographic Variables. Age, organizational tenure, gender, and occupational tenure were measured using single items.

Results

Descriptive Statistics

Table 2 show the means, standard deviations, and intercorrelation matrix for the independent and dependent variables under the “now”, and “economic growth” scenarios. As predicted, the supervisor’s competence as a voice manager was positively correlated with willingness to voice in both economic scenarios (voice about personal matters: recession: r=.36, p<.01 ; economic growth: r=29, p<.01;) (voice about others: recession: r=.32, p<.01; economic growth: r=.23, p<.01). However, contrary to expectations KAI scores were not significantly correlated with willingness to voice.

Table 2
Means, Standard Deviations, and Pearson’s Correlations Among Variables
Mean s.d. 1 2 3 4 5 6 7 8
1. KAI 91.07 12.28 1.00
2. Supervisor as Voice Manager 47.66 10.21 -.10 1.00
Voice during Recession
3. All items 27.78 6.06 .16 .38* 1.00
4. Personal Matters 18.26 4.03 .15 .36* .93* 1.00
5. About Others 9.52 2.72 .13 .32* .85* .60* 1.00
Voice during Economic Growth
6. All items 28.88 5.63 .05 .30* .77* .70* .69* 1.00
7. Personal Matters 18.76 3.69 .08 .29* .73* .79* .46* .92* 1.00
8. About Others 9.80 2.66 .01 .23* .62* .38* .82* .84* .56* 1.00

*p<0.01, two-tailed

HYPOTHESES TESTING

Employee Personality and Voice

Hypotheses 1, 3 and 4 were tested using a 2x2 repeated measure ANCOVA. The dependent variable was the respondents’ willingness to voice scores. The between subject variable was the personality of the respondents based on the KAI measure. Respondents were divided into groups of adaptors and innovators, using a median split of KAI-scores (Median = 89). The within subject factor was the state of the economy - recession vs. economic growth. To control for the supervisor’s ability to manage voice, this variable was entered as a covariate. The results for willingness to voice about personal matters and willingness to voice about others are shown in Tables 3 and 4.

Table 3
Repeated Measures ANCOVA: KAI and Willingness to Voice Concerns about Personal Matters.
Source SS df MS F Sig
Between – Subject Effects
Adaptor-Innovator (KAI) 64.632 1 64.632 2.736 0.100*
Cov: Voice Manager 455.643 1 455.643 19.29 0.000***
Error 3283.202 139 23.620
Within – Subject Effects
Economic State (ES) 18.869 1 18.869 6.145 0.014**
ES × KA 8.405 1 8.405 2.737 0.100*
ES × Cov 12.453 1 12.453 4.055 0.046**
Error 426.843 139 3.071

*p<0.1 **p<0.05 ***p<0.01

Table 4
Repeated Measures ANCOVA: KAI and Willingness to Voice Concerns about Others.
Source SS df MS F Sig
Between – Subject EffCcts
Adaptor-Innovator (KAI) 4.953 1 4.953 .406 0.525
Cov: Voice Manager 162.905 1 162.905 13.339 0.000***
Error 1697.518 139 12.2
Within – Subject Effects
Economic State (ES) 6.884 1 6.884 5.585 0.019**
ES × KA 9.076 1 9.076 7.363 0.007***
ES × Cov 4.901 1 4.901 3.976 0.048**
Error 171.324 139 1.233

*p<0.1 **p<0.05 ***p<0.01

Hypothesis 1, innovators will report more willingness than adaptors to voice to their supervisors, was not supported. Results showed no relationship between personality and willingness to voice about others, and only a marginally significant relationship between personality and willingness to voice about personal matters (F(1,139)=2.74, p<.1). However, there was a significant within subjects interaction effect between economic state and personality for willingness to voice about others (F(1,139)=7.36, p<.01), and a marginally significant interaction between economic state and personality, for willingness to voice about personal matters (F (1,139)=2.74, p<.1). Tables 5 and 6 compare willingness to voice means of adaptors and innovators both within and between the two economic scenarios for the two dependent variables. Hypothesis 3, innovators will be as willing to voice during a recession as during economic growth was supported. Hypothesis 4, adaptors will be less willing to voice during a recession than during economic growth, was supported for “concerns about personal matters.” Mean willingness to voice was 3.54 during the recession scenario compared to 3.70 during economic growth (t(1,70)=2.56, p<.01). The interaction effect of personality and economic state for both the dependent variables is illustrated in Figures 1 and 2.

Table 5
Interaction Effect between Adaptation-Innovation Personality and Economic State on Employee Willingness to Voice Concerns about Personal Matters (mean scores).
Adaptors Innovators t(1, 140) Sig.
Economic Growth 3.70 (.84) 3.81 (.62) 0.816 0.208
Recession 3.54 (.91) 3.77 (.68) 1.713 0.045**
t(1,70) 2.59 0.67
Sig. 0.006*** 0.252

*p<0.1 **p<0.05 ***p<0.01; s.d. in brackets

Table 6
Interaction Effect between Adaptation-Innovation Personality and Economic State on Employee Willingness to Voice Concerns about Concerns about Others (mean scores).
Adaptors Innovators t(1, 140) Sig.
Economic Growth 3.29 (.97) 3.23 (.80) 0.345 0.365
Recession 3.08 (1.01) 3.26 (.80) 0.243 0.122
t(1, 70) 2.751 0.564
Sig. 0.008*** 0.287

*p<0.1 **p<0.05 ***p<0.01; s.d. in brackets

Figure 1.
Interaction Effect between Adaptation-Innovation Personality Traits and Economic State on Employee Willingness to Voice Concerns about Personal Matters.
Employee Willingness to Voice Concerns about Personal Matters

Figure 2.
Interaction Effect between Adaptation-Innovation Personality Traits and Economic State on Employee Willingness to Voice Concerns about Others.
Employee Willingness to Voice Concerns about Others

Supervisor’s Competence and Voice

To test hypotheses 2, 5 and 6, we used a 2×2 repeated measure ANCOVA. The dependent variable was the likelihood to voice score. The between subject factor was the respondents’ ratings of their supervisors as voice managers. Supervisors were divided into two groups using median split of the voice manager ratings (median = 48). Those above the median were categorised as good voice managers and those below the median as poor voice managers. The within subject factor was the state of the economy – recession vs. economic growth. The individual’s adaptation-innovation score was used as a covariate to partial out the effects of personality. The results are shown in Tables 7 and 8.

The main effects for supervisor as voice manager, as well as the interaction effect between supervisor as voice managers and economic state were significant for both dependent variables. Hypothesis 2, employees who perceive that their supervisors are approachable and responsive will express more willingness to express voice than employees who perceive their supervisors to be unapproachable and unresponsive, was supported (willingness to voice about personal concerns: F(1,139)=12.05, p<.01; willingness to voice about others: F(1,139)=17.61, p<.01). Tables 7 and 8 also show a significant interaction effect between economic state and supervisor as voice manager for both dependent variables (willingness to voice about personal concerns: F(1,139)=6.19, p<.05; willingness to voice about others: F(1,139)=4.85, p<.05). We analysed the effects of the interaction in Tables 9 and 10 by comparing voice means within and between the two different scenarios for the two dependent variables. Hypothesis H5, employees who perceive that their supervisors are approachable and responsive voice managers will be as much, if not more willing to express voice during a recession than during economic growth, was supported since there were no differences in means between the two scenarios. Hypothesis H6, employees who perceive that their supervisors are unapproachable and unresponsive will be less willing to express voice during a recession than during economic growth, was also supported. Mean willingness to voice about personal matters fell from 3.59 under economic growth, to 3.40 under the recession scenario (t(1,70)=2.60, p<.01). Mean willingness to voice about others fell from 3.03 to 2.84 (t(1,69)=2.62, p<.01). Figures 3 and 4 graphically illustrate these results.

Table 7
Repeated Measures ANCOVA: Supervisor and Willingness to Voice Concerns about Personal Matters.
Source SS df MS F Sig
Between – Subject Ejfrcts
Supervisor: Voice Manager 149.085 1 149.085 12.046 0.001***
Cov: KAI 37.593 1 37.593 3.037 0.084*
Error 1720.338 139 12.377
Within – Subject Effects
Economic State (ES) 19.038 1 19.038 6.292 0.013**
ES × Voice Manager 18.737 1 18.737 6.193 0.014**
ES × Cov 9.880 1 9.880 3.265 0.073*
Error 420.558 139 3.026

*p<0.1 **p<0.05 ***p<0.01

Table 8
Repeated Measures ANCOVA: Supervisor and Willingness to Voice Concerns about Concerns about Others.
Source SS df MS F Sig
Between – Subject Effects
Supervisor: Voice Manager 104.590 1 104.590 17.609 0.000***
Cov: KAI 4.563 1 4.563 0.768 0.382
Error 825.622 139 5.940
Within – Subject Effects
Economic State (ES) 27.148 1 27.148 11.081 0.001***
ES × Voice Manager 11.890 1 11.890 4.853 0.029**
ES × Cov 19.643 1 19.643 8.017 0.005***
Error 340.561 139 2.450

*p<0.1 **p<0.05 ***p<0.01

Table 9
Interaction Effect between Supervisor as Voice Manager and Economic State on Employee Willingness to Voice Concerns about Personal Matters (mean scores).
High Voice Managers Low Voice Managers t(1, 140) Sig.
Economic Growth 3.90 (.70) 3.59 (.75) 0.816 0.208
Recession 3.90 (.73) 3.40 (.81) 1.713 0.045**
t(1,70) 0.672 2.594
Sig. 0.252 0.005

*p<0.1 **p<0.05 ***p<0.01; s.d. in brackets

Table 10
Interaction Effect between Supervisor as Voice Manager and Economic State on Employee Willingness to Voice Concerns about Others (mean scores).
High Voice Managers Low Voice Managers t(1, 141) Sig.
Economic Growth 3.49 (.81) 3.03 (.90) 2.48 0.07*
Recession 3.49 (.86) 2.84 (.84) 3.78 0.000***
t t(1,72) = 0.21 t(1,69) = 2.62
Sig. 0.417 0.006***

*p<0.1 **p<0.05 ***p<0.01; s.d. in brackets

Figure 3
Interaction Effect between Supervisor as Voice Manager and Economic State on Employee Willingness to Voice Concerns about Personal Matters.
Employee Willingness to Voice Concerns about Personal Matters

Figure 4
Interaction Effect between Supervisor as Voice Manager and Economic State on Employee Willingness to Voice Concerns about Others.
Employee Willingness to Voice Concerns about Others

Discussion

This study extends prior research of Saunders et al. (1992) and Janssen, et al. (1998) by examining the effects of the economic state on employees’ willingness to voice. The findings show that the economic state is a significant moderator in the relationship between (1) supervisor’s competence as voice manager and employee willingness to voice; and (2) cognitive adaptation-innovation personality and employee willingness to voice.

The results support the finding of Saunders et al. (1992) and Janssen et al. (1998). Subordinates who perceived that their supervisors were competent voice managers were more likely to voice than subordinates who perceived that their supervisors were unapproachable or unresponsive to voice. In addition, we found that this relationship was stronger in the recession scenario than in the economic growth scenario. During the recession scenario, subordinates of good voice managers were as willing to voice as they were in the economic growth scenario. However, subordinates who perceived their supervisors to be poor voice managers were less willing to voice their opinions and concerns in the recession scenario. The results also showed that respondents were more willing to voice personal concerns than concerns about others. This finding may be cultural specific; Singaporeans are brought up to “mind their own business” and voicing about others may violate that social norm. Supervisors who are good voice managers interact with subordinates to establish a high level of trust and strong social exchange ties. Once these tie are established, subordinates feel comfortable bringing up difficult and potentially contentious issues, even during an economic recession. Poor voice managers, on the other hand, do not establish a similar level of trust or strong social network ties, resulting in fewer input from their subordinates. During a recession, the possibility of retrenchment, and the lack of alternative job opportunities increases the risk of voicing to superiors. Job insecurity creates a climate where people may be afraid to talk about their personal concerns at work and do not feel free to complain about colleagues and supervisors (Foster, 1996). Insecurity also leads employees to lose trust in their supervisors and creates a reluctance to share information (Kennedy, 1996).

Economic conditions also influenced the relationship between employee cognitive style preference and willingness to voice. In economic growth scenario, adapters were as willing as innovators to voice their concerns. However, in the recession scenario adaptors were significantly less willing to voice than in the growth scenario, and significantly less willing to voice than innovators. These results were consistent for both dimension of employee voice. Innovators are more likely to take calculated risk, are more skilled at persuasion, and are generally used to “standing out” among the crowd. In contrast, adaptors are more risk adverse and less tolerant of ambiguity (Carne & Kirton, 1982; Kirton, 1976; Goldsmith, 1986; Gryskiewicz, 1980). The act of voicing carries more risk in a recession than in a period of economic growth, although voice may be more critical during a recession.

Managerial Implications

Employees help ensure the long term success of the firm by assisting in making decisions essential for innovation, quality improvement, and rapid response to change (Parks, 1995). Organisations are increasingly relying on employees to voice their concerns and make suggestions to improve organisational effectiveness. The results of this study clearly suggest that one of the most effective ways to encourage voice is by ensuring that supervisors are approachable and responsive voice managers. Supervisors need to be trained to identify and understand the fears and needs of subordinates and how these factors influence voice behaviour. Ludeman (1983) suggests that managers must be trained to listen to employees without becoming defensive. They also need to be trained in how to best elicit input from employees, as well as how to implement their suggestions when appropriate. If action is not taken, they need to know how to clearly communicate the reasons for the decision so as not to discourage future participation.

Organisations can encourage supervisors to gain competence in managing voice by including voice management as a component in their performance appraisal evaluations. To build a culture that encourages voice, organisations should also consider hiring and promoting managers on the basis of their ability to elicit and respond to voice.

While voice management may be less critical during periods of economic growth, it would make a real difference during an economic recession or crisis when constructive employee inputs are most needed. During periods of economic growth, when resources are more abundant and the environment relatively benign, corporate leaders could invest in developing voice management competencies within the organisation as a hedge against crisis paralysis.

The study also suggests that managers should ensure that they hire and retain innovators as well as adaptors, because innovators are more likely to voice during a crisis. For the same reason, they should also ensure that work groups and project teams are composed of both adaptors and innovators. However, assembling a mix of personalities is only half the battle. Hayward and Everett (1983) found that innovators must be given room to stretch themselves and exercise their creativity, otherwise they would leave the organisation within five years. To retain innovators, managers must be prepared to tolerate some degree of risk and uncertainty by allowing innovators to try the untried.

Goman (1989) suggests that creative problem solving and the generation of new ideas are among the most important and valued traits in employees in a rapidly changing ambiguous environment. Thus, organisations may want to hire innovators from outside the organisation in turnaround situations. Managers with preferences for innovation may also be selected to key positions during major organisational changes. (Kirton, 1984).

Limitations and Implications for Future Research

In this study, all the variables were obtained from a single questionnaire, possibly resulting in common method bias. Common method biases inflates type I error, or errors of finding positive results when a study looks for commonality between variables, but work against hypotheses that look for differences between variables. In this study, the key hypotheses involved finding differences in interactions which are relatively insensitive to common method bias (Aiken & West, 1991; Kerlinger, 1986). Evans (1985) found empirical support for this contention with their Monte Carlo simulation and concluded that common method in collecting both criterion and predictor variables is not a source of spurious interactions.

The second limitation of this study is that we measured willingness to voice under different scenarios rather than documenting actual voice behaviour at work. As with any survey, it is possible that the responses received in this study may actually differ from actual work behaviour; participants might be responding in the way they think they should be responding. Thus, the results may be capturing their implicit theory of the world rather than their actual behaviors. However, attribution theories do suggests that human behaviors are guided by implicit theories held by people. Nonetheless, the reliance on surveys has been an ongoing weakness in the voice literature. In the future, researchers may want to investigate the predictors of actual voice behavior by using methods other than self-report surveys. For example, subordinates could keep diaries of any work concerns or ideas they might have, whether they tell anybody about the ideas, and whom they tell. Supervisors could also keep records of when their subordinates voice their concerns and/or make suggestions for change. In addition, researchers could conduct interviews with both subordinates and supervisors, and observe instances of voice during meetings.

A third weakness of this study is that we limited our measure to willingness to voice directly to a supervisor. Previous studies have also only looked at a narrow range of voice behaviour. Future studies need to consider the different modes and content of voice behaviour. For example, mode of voice may be direct or indirect. An employee can directly voice his or her concerns at a department meeting, or informally talk to individuals before a meeting so that his or her concern is actually voiced by someone else, perhaps someone who is in a better position to be heard. In terms of content, voice can be categorized as conventional or novel, and whether it is directed at changing individual circumstances, the group environment, or the organization as a whole. Both mode and content may be influenced by national culture. However, with the exception of this research, and a paper by Lee and Jablin (1992), studies of voice have been limited to Western cultures. We found that a couple of the same factors that have been found to influence willingness to voice in Western cultures (personality, and the quality of the supervisor as voice manager) influence willingness to voice in Singapore, but we have only looked at direct voice to supervisors. Future studies could examine how mode and content of voice may vary across cultures.

Conclusion

In conclusion, despite its limitations, this study provides some evidence that the state of the economy moderates the relationship between willingness to voice and two factors: cognitive style preference for adaptation or innovation, and the supervisor’s competence as a voice manager. Future studies could include this factors as part of a more fully developed model of voice mode and content.

Appendix 1

Supervisor’s Competence as Voice Manager Scale

Responsiveness

1. My boss listens carefully to what I say when I bring up a concern.
2. My boss gives high priority to handling employee concerns.
3. My boss takes action to correct the concerns that I speak to him or her about.
4. My boss is fair when I take a concern to him or her.
5. I take concerns to my boss because he or she deals with them effectively.
6. My boss is willing to support me if my concern is valid.
7. My boss handles my concerns promptly.

Approachability

8. 1 don’t know how my boss will react when I take a concern to him or her.
9. I do not know how to take a concern to my boss.
10. I am not afraid to take a concern to my boss.
11. I don’t know what to expect when I take a concern to my boss.
12. It is difficult to take a concern to my boss.
13. I don’t know how my boss will behave when I take a concern to him or her.
14. I find it quite stressful to take a concern to my boss.

References

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

Argyris, C. & Schon, D. A. (1978). Organisational learning: A theory of action perspective. Reading, MA: Addison-Wesley.

Argyris, C. & Schon, D. A. (1996). Organizational Learning II: Theory, Method, and Practice, Reading, MA: Addison-Wesley.

Britt, T. W., Boniecki, K. A., Vescio, T. K. & Biernat, M. (1996). Intergroup anxiety: A person * situation approach. Personality & Social Psychology Bulletin, 22(11), 1177-1188.

Came, O. C. & Kirton, M. J. (1982). Styles of creativity, test score correlations between the Kirton Adaptation-Innovation Inventory and the Myers-Briggs Type Indicator. Psychological Reports, 50, 31-36.

Channon, D. (1996). Can eastern HR promise the West a new dawn? People Management, (August), p.21.

Clapp, R. G. & de Ciantis, S. M. (1989). Adaptors and innovators in large organizations: Does cognitive style characterize actual behaviour of employees at work? An exploratory study. Psychological Reports, 65, 503-513.

Doherty, N., Bank, J. & Vinnicombe, S. (1996). Managing survivors: the experience of survivors in BT and the British financial sector. Journal of Managerial Psychology, 11(7), 51-59.

Dolven, B. (1998). Less than a Lion’s share. Far Eastern Economic Review, 161(45), 46-47.

Dolven, B. (1999). Cool it. Far Eastern Economic Review, 162(9), 46.

Eisenbergem, R., Fasolo, P., & LaMastro, V. D. (1990). Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75(1), 51-89.

Elliston, F. J., Keenan, J., Lockhart, P. & Schaick, J. V. (1985). Whistleblowing: Managing dissent in the workplace. New York: Praeger.

Evans, M. G. (1985). A monte carlo study of the effects of correlated method variance in moderated multiple regression analsysis. Organizational Behavior and Human Decision Processes, 36, 315-342.

Feuille, P. & Delaney, J. T. (1993). The individual pursuit of organizational justice: Grievance procedures in nonunion workplaces. In G.R. Ferris and K.M. Rowland, (eds.). Research in Personnel and Human Resource Management, 10. Greenwich, Conn: JAI Press, 187-232.

Forehand, G. A. & Gilmer, B. V. (1964). Environmental variations in studies of organizational climate. Psychological Bulletin, 6, 361-382.

Foster, J. (1996). Home alone. People Management, 2(19), 23.

Goldsmith, R. (1986). Personality and adaptive innovative problem-solving. Journal of Social Behaviour and Personality, 1, 95-106.

Goldsmith, R. E. & Matherly, T. A. (1986). The Kirton Adaptation Innovation Inventory, faking and social reliability. A replication and extension. Psychological Reports, 58, 269-270.

Goman, C. K. (1989). Creativity in business: A practical guide for creative thinking. Menlo Park, CA: Crisp Publications.

Gryskiewicz, S. S. (1980). A study of creative problem solving techniques in group settings. Unpublished Ph.D. thesis, Birkbeck College, University of London.

Haywood, G. & Everett, C. (1983). Adaptors and innovators: Data from the Kirton Adaptation-Innovation Inventory in a local authority setting. Journal of Occupational Psychology, 56, 339-342.

Hirschman, A. O. (1970). Exit, voice and loyalty: Responses to decline infirms, organizations, and states. Cambridge, MA.: Harvard University Press.

Holland, P. A. (1987). Adaptors and innovators: Application of the Kirton Adaptation- Innovation Inventory to bank employees. Psychological Reports, 60, 263-270.

Isabella, L. (1989). Downsizing - survivors assessments. Business Horizons, 32, 35-41.

Janssen, 0., de Vries, T., & Cozijnsen, A. J. (1998). Voicing by adapting and innovating employees: An empirical study on how personality and environment interact to affect voice behavior. Human Relations, 51(7), 945-967.

Keller, R. T. & Holland, W. E. (1978). A cross validation study of the Kirton Adaptation Innovation Inventory in three research and development organizations. Applied Psychological Measurement, 2, 563-570.

Kerlinger, F. N. (1986). Foundations of behavioral research. Orlando, FL: Holt, Rinehart & Winston.

Kennedy, C. (1996). The incredible shrinking company. Director, (April), 62-68.

Kirton, M. J. (1976). Adaptors and innovators: A description and measure. Journal of Applied Psychology, 61, 622-629.

Kirton, M. J. (1984). Adaptors and innovators - Why new initiatives get blocked. Long Range Planning, 17, 137-143.

Kirton, M. J. (1987). Adaptation-Innovation Inventory (KAI) manual (2s ed). Hatfield, Herts: Occupational Research Centre.

Kirton, M. J., & de Ciantis, S. M. (1986). Cognitive style and personality: The Kirton Adaptation-innovation and Cattell’ s sixteen personality factor inventories. Personality and Individual Differences, 7(2), 141-146.

Lee, J. & Jablin, F. M. (1992). A Cross-Cultural Investigation of Exit, Voice, Loyalty and Neglect as Responses to Dissatisfying Job Conditions, The Journal of Business Communication, 29(3), 203-227.

LePine, J. A. & Van Dyne, L, (1998). Predicting Voice Behavior in Work Groups. Journal of Applied Psychology, 83(6), 853-868.

Lewin, K. (1936). Principles of topological psychology. NY: McGraw-Hill.

Lewin, D. (1987). Dispute resolution in the non-union firm: A theoretical and empirical analysis. Journal of Conflict Resolution, 31, 465-502.

Lewin, D. (1992). Grievance procedures in nonunion workplaces: An empirical analysis of usage, dynamics, and outcomes. Chicago-Kent Law Review, 68(1).

Lewin, D. & Peterson, R. B. (1988). The modern grievance procedure in the United States. Westport, CT: Greenwood Press.

Ludeman, K. (1983). Nine steps to resolving employee complaints. Office Administration and Automation, 44(6), 30-32.

Mishra, A. K. & Spreitzer, G. M. (1998). Explaining how survivors respond to downsizing: The roles of trust, empowerment, justice, and work redesign. Academy of Management Review, 23(3), 567-588.

Merton, R. K. (1957). Social theory and social structure. NY: Free Press of Glencoe.

Parker, L. E. (1993). When to Fix It and When to Leave: Relationships Among Perceived Control, Self-Efficacy, Dissent, and Exit. Journal of Applied Psychology, 78(6), 949-959.

Parks, S. (1995). Improving workplace performance: Historical and theoretical contexts. Monthly Labour Review, (May), 18-28.

Roger, C. R. (1959). Towards a theory of creativity. In H.H. Anderson, (Ed.), Creativity and it’s cultivation. NY: Harper

Rowe, M. P. & Baker, M. (1984). Are you hearing enough employee concerns? Harvard Business Review, 62, 127-135.

Rusbult, C. E., Farrel, D., Rogers, G. & Mainous III, A. G. (1988). Impact of exchange variables on exit, voice, loyalty, and neglect: An integrative model of response to declining job satisfaction. Academy of Management Journal, 31(3), 599-627.

Saunders. D. M., Sheppard, B. H., Knight, V. & Roth, J. (1992). Employee voice to supervisors. Employee Responsibilities and Rights Journal, 5(3), 241-259.

Slater, J. (1999). In the Hot Seat. Far Eastern Economic Review, 21, 42-44.

Spencer, D. G. (1986). Employee Voice and Employee Retention. Academy of Management Journal, 29(3), 488-502

Walsh, W. B., Craik, K. H. & Price, R. H. (eds.) (1992). Person-environntent psychology: Models and perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates.

Westin, A. F., Kurtz, H. I. & Robbins, A. (1981). Whistleblowing: Loyalty and dissent in the corporation. NY: McGraw Hill.

Weber, M. (1948). Essays in Sociology. London: Routledge.

Withey, M. J. & Cooper, W. E. (1989). Predicting exit, voice, loyalty and neglect. Administrative Science Quarterly, 34, 521-539.

Wong, M. (1994). Bad times provide good opportunities for business. The Canadian Manager, 19(3), 21.

Yuen, C. C. (1995). The effects of career salience and life-cycle variables on perceptions of work-family interfaces. Human Relations, 48(3), 265-284.