Orientation: In the current economic climate and the resulting fast-changing global business and political environment, trust among different role players in organisations has become critical for survival.

Research purpose: The objective of this study was to explore the impact of different variables such as demographics on trust relationships in South African organisations.

Motivation for the study: Anecdotal evidence and preliminary data collected for a national trust indicator seemed to suggest a shift in trust levels in organisations.

Research design: A trust questionnaire was administered to a convenience sample of 307 respondents in all economic sectors. Parametric and nonparametric analyses were used to determine significant differences among economic sectors, job levels and sample periods.

Main findings: No significant differences were found for job levels or the different sample periods. However, significant differences were found for the economic sectors and, specifically, between government participants and other sectors for the dimensions of change, team management, organisational trust, information sharing and credibility.

Practical implications: In times of change, leadership in organisations need to be aware of the impact on trust levels. It is therefore important that leaders in government focus more on trust-enhancing behaviours needed to repair mistrust in organisations.

Contribution: Although the effect of time on trust levels is inconclusive, the clearly differing levels of trust in various economic sectors point to the importance of appropriate and fitting approaches to building trust and not a 'one-size-fits-all' attitude.

Over the last decade, the phenomenon of trust within and among organisations has been the focus of increasing attention on both an international and national level (Bews & Martins, 2002). Tyler and Kramer (1996) already noted that because of widely publicised organisational practices such as the high compensation of CEOs, organisations have experienced declining trustworthiness in the eyes of their employees and members of other organisations. Robinson and Rousseau (1994) discovered that 55% of their respondents reported that their employer had violated a psychological contract and that this violation significantly reduced trust. Indeed, the popular press has noted that 'a seemingly endless stream of bad news alleging widespread management negligence and malfeasance is chipping away at the trust vital to a free-market system' (Byrne, 2002, p. 31). An international survey in 2005 found that only 25% of Americans, 21% of Europeans, 29% of Japanese, 43% of Chinese and 55% of Brazilians believe that information from their CEO is credible (Kreitner & Kinicki, 2007). According to Salamon and Robinson (2008), the research on the importance of trust in organisations has grown rapidly. Furthermore, researchers such as Lau, Lam and Salamon (2008), highlight that employees' trust in their leaders has been found to have important benefits for organisations, such as improving organisational commitment, organisational citizenship behaviour, team performance and organisational performance. Research in South Africa by Von der Ohe, Martins and Roode (2004) emphasised the importance for South African managers or supervisors of recognising traits such as being organised and hardworking, as well as being dependable, thorough and responsible in order to enhance trust. The opposite can lead to a lack of trust in a manager or supervisor and a low level of credibility being associated with that person. A study of 500 business professionals indicated that the main factor in deciding to stay on in their job or leave was whether they had a trusting relationship with their manager (Kreitner & Kinicki, 2007).

As the continuous changes in the world of work will inevitably influence trust relationships positively or negatively, it was decided to monitor trust levels over time. This national longitudinal trust survey is meant to be a barometer of the levels of trust in a fast-changing South African business environment. During preliminary analysis of the data, it was noticed that certain institutions seem to stand out constantly (Büssing, 2002; Von der Ohe et al., 2004). One of these was the public service.

According to Kroukamp (2008), trust in institutions, especially government, is vital to achieve stability and development in every nation. Trust enhances confidence in institutions and, consequently, enhances the cooperation of citizens to effect the policies and programmes of government.

Ethics and behaviour in the South African public service context occur within a definite legislative and policy framework. Edwards (2008) lists at least 13 different acts and white papers intended to promote and emphasise sound ethical conduct, professionalism and accountability in the public sector. In addition, South African organisations must adhere to a number of implicit regulations in order to have legitimate employment regulations, or organisational rules, in the workplace (Esterhuizen & Martins, 2008). Another important aspect, which might have an influence on trust in organisations, is the worldwide economic recession. Nearly 20 years ago already, Brown (1992) mentioned that during times of recession it is much more difficult for participatory management to succeed as trust decreases because of management actions such as, for instance, freezing salaries or retrenching workers.

The mentioned regulations, continuous changes in the world of work as well as the worldwide economic recession can all have an impact on employment relations and, especially, the trust relationships in organisations.

According to Arrow (1974), trust is an important lubricant of a social system. It is extremely efficient in that trust means that people can rely on other people's promises. Researchers such as Mayer, Davis and Schoorman (1995) posit that a party decides whether to trust another (the trustee) based on expectations about the other party's future behaviour, as determined by the perceived trustworthiness of the trustee. Bäckstrőm (2008, p. 21) defines trust as follows: 'Trust concerns one party's confidence in an exchange partner'.

According to Reina and Reina (1999, p. 11), trust is 'a relationship of mutual confidence in contractual performance, honest communication, expected competence and a capacity for unguarded interaction'.

Kreitner and Kinicki (1995, p. 342) define trust as 'a reciprocal faith in others' intentions and behaviour'. Shaw (1997, p. 21) views trust as 'the belief that those on whom we depend will meet our expectations of them' and Robbins (1996, p. 357) defines trust as the 'characteristic of high-performance teams where members believe in each other's integrity, character, and ability'. Accordingly, Martins (2000) concisely states that various characteristics appear to emerge from the current models and definitions, which focus on features such as integrity, competence, openness, vulnerability, reliability and expectations, as well as on the role of the trustee and trustor.

Butler (1991), Castelfranchi and Falcone (1998), Du Plessis (2006), Lau et al. (2008), Mayer et al. (1995), Shaw (1997) and others have examined the components of trust. Mayer et al. (1995) developed a model of trust of which the key factors are ability, benevolence and integrity. Shaw (1997) developed a model of trust that encompasses the organisation as a whole by also addressing the formal structures and processes and organisational culture. He describes the key leverage points to sustain trust as 'achieving results, acting with integrity and demonstrating concern (Shaw, 1997, p. 33)'. Pennington (1992) states that earning and maintaining the trust of others require leaders to excel in five areas, namely character, competence, communication, consistency and courage. Research done by Martins, Watkins, Von der Ohe and De Beer (1997) led to the assumption that trust in organisations is created by personality factors and managerial practices. The personality factors are agreeableness, conscientiousness, resourcefulness, emotional stability and extraversion; the managerial practices are information sharing, work support, credibility and team management.

Barrick and Mount (1991) confirmed that five personality characteristics, referred to as the 'Big Five' (in this research, referred to as the five-factor model of personality), explain a significant proportion of work performance in industrial settings. These characteristics were viewed as possible antecedents of interpersonal trust among managers and employees (Martins et al., 1997).

A summary of the elements in a relationship of trust, as viewed by various researchers, is displayed in Table 1.

The research by Martins et al. (1997) focused on determining a relationship between personality, managerial practices and trust on a generic level but did not empirically address the specific aspects of personality and managerial practices that are at the core of the positive relationship found with trust. Martins (2000) subsequently developed a comprehensive model that defines the specific personality attributes or characteristics of the trust relationship, also taking into account specific components of managers' behaviour.

Taking cognisance of the above for the purposes of this study, the authors restricted the role and influence of trust to the field of industrial psychology and, specifically, to the employer-employee relationship.

Trust is therefore defined as the process in which a trustor relies on a trustee (a person or group of people) to act according to specific expectations that are important to the trustor, without taking advantage of the trustor's vulnerability (Martins, 2002). This definition provides the foundation for this study, as it supports both the model of trust and the questionnaire that are used in this follow-up study.

According to Lau et al. (2008), relatively few researchers have examined the effect of relational demographics in manager-employee dyads. Tsui and O'Reilly (1989) found that differences in education, gender and race between members of supervisor-subordinate dyads were related to increased role ambiguity, unfavourable performance evaluations and a lower level of attraction on the part of managers to their corresponding subordinates. In a study by Hay (2002), the researcher concluded that the differences in trust in different relationships provided a strong argument for the study of trust at a number of different levels.

In his research, Büssing (2002) found that it is characteristic for the new type of highly qualified and flexible employee to experience a loose relationship between personal trust in the supervisor or organisation and identification with their job and work. Den Hartog, Shippers and Koopman (2002) found that employees on lower levels would often not have enough discretion or resources to pursue visions that deviate significantly from basic organisational goals or those proposed by higher management. Bews and Uys (2002), who found that more respondents trust their immediate supervisors rather than top management, confirmed this. They also found that employees at higher job levels have more trust in their direct supervisors than those at lower job levels. Another important finding by Bews and Uys (2002) was that employees over the age of 52 were inclined to be more positive about management and the unfolding of change than were younger employees. Trust in authority figures, superiors and decision makers are vital as these people make a significant input regarding the distribution or allocation of rewards and resources (Colquitt, Conlan, Wessan, Porter & Ng, 2001).

Research done by Cyster (2005) found significant differences among the results of some biographical groups in a South African company:

• The personal trust between supervisors and subordinates is much higher in some business units than in others.
• Significant differences were found for all five of the trust dimensions among job grades, with management being the most positive group and the skilled and qualified employees (Patterson C band) being the most negative group.
•Significant differences were found among the qualification groups and, especially, between those with less than Grade 8 qualifications and those with diplomas, as well as between those with Grade 8 or 9 qualifications and those with diplomas.

Du Plessis (2006) explains that where distrust exists, for instance among departments or areas within a business, it is likely that there will be pockets of knowledge in the organisation that will not be integrated or free flowing because knowledge is not shared by the individuals working in these pockets. Research by Lau et al. (2008, p. 203) found that

when the composition of vertical dyads follows education and organisational rank norms (i.e. better educated vs. less educated managers, senior managers vs lower-ranked managers), staff members perceive managerial trustworthiness to be higher than they would if the dyads were compared otherwise. (Lau et al., 2008, p. 203)

They furthermore found that demographic dissimilarities have a more significant influence on perceptions of trustworthiness than do similarities.

Another facet that needed to be addressed was the question of the influence that time has on trust. When revisiting their seminal article (Mayer et al., 1995), Schoorman, Mayer and Davis (2007) noted that, against their expectations, the issue of time had not been sufficiently explored in real-life situations. In this study the researchers attempted to take note of this as '[b]y including a consideration of time, studies of trust should lead to more predictable results' (Schoorman et al., 2007, p. 352).

It appears as if job levels, age, differences in education, gender and race have an impact on trust relationships. Although the above discussion shows that demographics appear to have an impact on trust and the way trust is perceived, Bews and Uys (2002) concluded that more empirical research is needed before a clear picture and understanding can emerge. A focus of the current study is to investigate the impact of demographics such as job level, industry and sample periods on the trust levels of employees.

Based on this study, and in accordance with the brief discussion above of the trust construct and demographical impact, the following hypotheses were formulated:

Hypothesis 1:

There is a significant difference between the two sample time periods.

Hypothesis 2:

There is a significant relationship between the dimensions of trust and job levels.

Hypothesis 3:

There is a significant relationship between the dimensions of trust and the economic sector.

Research approach

Research method

Research participants and sampling strategy
The trust questionnaire, consisting of the three biographical questions, was posted on a survey company's website with an open invitation for individuals and organisations to participate, the rationale being to involve as many employees from the whole spectrum of business sectors in South Africa as possible. Another justification for posting on the website was to obtain unbiased information from as diverse a group of participants as possible and it was hoped that the respondents would feel more willing to reveal confidential information, such as the name of their employer, on a third-party website. Secondly, the survey, with an invitation to participate, was also placed on the website of HR Future, a monthly magazine focused on human resources. Thirdly, invitations were sent out with a link to a database of 800 current or previous course participants in the field of human resource or industrial psychology. As an Internet-based survey methodology was used, it was difficult to ascertain whether the correct population was targeted. For this reason, a convenience sampling strategy was followed to reach as big an audience as possible in all economic sectors. The target population was not only human resources managers but also included line managers with an interest in the area of organisational trust.

The data were collected from respondents following one of the links discussed above to the survey instrument. As this is a web-based application, the data were anonymously stored on the server as soon as the respondent had completed the questionnaire. The data were then verified as far as possible by checking for contradictions and obvious misinformation.

In total, 307 participants completed the questionnaire. The fear that the above method of attracting responses would target mainly the younger 'web-wise' generation was unfounded, as about half (51.1%) indicated that they were over 35 years of age. Furthermore, the participants ranged from top management (3.3%), senior management (12.7%), middle management (26.1%) and supervisors (9.8%) to technical staff or professionals (31.6%) and other levels (13.7%).

Because of the large number of respondents that had classified themselves as 'other' (17.3%), it was decided to recode the sectors. It appears as if respondents did not always identify their sectors correctly. As nearly all of the respondents (93%) had supplied the name of their employer to one of the open questions, it was possible to recode the sectors. After recoding, the following main sectors emerged: primary (mines, manufacturing and construction) (21.2%), services (29.6%), government (34.8%), retail and wholesale (7.2%) and unknown (7.2%).

It was decided to use these four newly grouped sectors in the subsequent analysis.

Measuring instrument
The objective of the trust questionnaire was to gather views on the existence of a trust relationship, correlated with personality and managerial practices and behaviour as a whole. With the current focus of the research (see the hypotheses) and the electronic nature of the survey, it was decided to exclude the section on personality, as it would unnecessarily extend the length of the questionnaire. The focus would also not be on the personality constructs per se but rather on trust and change.

Ultimately, the questionnaire consisted of sections covering managerial practices, trust relationships and organisational trust and, lastly, a section that measured any changes that had occurred.

PManagerial practices construct:
The managerial practices section measured the following managerial dimensions, which are defined as follows (Martins, 2002):

[Credibility:] This includes a willingness to listen, consider proposals, allow others the freedom to express feelings, tolerate mistakes and ensure that employees enjoy prestige and credibility in the organisation.

[Team management:] This dimension refers to the effective management of team and individual goal accomplishments and the handling of conflict within groups.

[Information sharing:] This dimension indicates a willingness to give individual feedback on performance and to reveal company-related information in an honest manner.

[Work support:] This dimension deals with the willingness to support employees when needed and provide job-related information for accomplishing objectives. (Martins, 2002)

Trust relationship:
The trust relationship dimension in the questionnaire was directly related to the trust dimension and was measured by five questions dealing with various aspects of trust between employees and their immediate supervisors. The trust relationship dimension reflects the relationship with the immediate supervisors in terms of openness, honesty, fairness and intention to motivate employees.

Organisational trust:
This dimension focused on the trust relationship between top management, the immediate manager and lower level employees.

Changes that had occurred:
In addition to the above dimensions, an additional section was added to the questionnaire to measure participants' satisfaction with changes that had occurred in their organisations.

Reliability and validity of the questionnaire:
The reliability of the questionnaire was shown to be highly satisfactory, with alpha coefficients ranging between 0.82 and 0.94 for the various dimensions (Martins, 2000). In addition, the goodness-of-fit index (GFI) was 0.95, the adjusted-goodness-of-fit index (AGFI) was 0.91 and the parsimony-goodness-of-fit index (PGFI) was 0.50 (Martins, 2000). The goodness-of-fit test is performed in order to determine whether a set of observed data corresponds to some theoretical distribution (Downie & Health, 1983). A GFI with a value of close to 0.90 reflects a good fit, an AGFI with a value of 0.90 reflects a good model fit and a PGFI varies between 0 (no fit) and 1 (perfect fit). It could therefore be deduced that a good model fit was established (Martins, 2000).

Structural equation modelling (SEM) was used to assess the validity of the conceptual model of the manifestation of trust. SEM is a linear cross-sectional statistical modelling technique, which includes confirmatory factor analysis, path analysis and regression analysis (Kline, 1998). The EQS software programme was utilised to test the trust model, using SEM (Martins, 2000).

A confirmatory factor analysis was conducted, following the procedure outlined by Schumacher and Lomax (1996), to determine whether the data were compatible with the assumption that each of the proposed latent variables represented separate constructs.

The results revealed a chi-square of 4Â 404.511, based on 33 degrees of freedom with a probability value of less than 0.001. The normal theory chi-square for this solution is 3Â 538.303. Additionally, the comparative fit index (CFI) = 0.890, the Bentler-Bonet normed fit index (BDNFI) = 0.889 and the Bentler-Bonet non-normed fit index (BBNNFI) = 0.850. These were all very close to the recommended perfect fit, 0 (no fit) to 1 (perfect fit) (Martins, 2002).

Statistical analysis:
To analyse the data, both parametric as well as nonparametric techniques were used where appropriate. Firstly, item and reliability analysis in the form of Cronbach's alpha was conducted. To test the hypotheses, t-tests and analysis of variance were used.

Item and reliability analysis

The results of all 10 dimensions were above 0.80. Typically, this is what is expected for an established questionnaire. According to Nunnally (1967), the Cronbach-alpha scores for an established scale should be at least above 0.70.

An examination of the results of the item analysis shows only two items, which can increase the Cronbach-alpha scores of the dimensions if deleted. However, these increases are very low and therefore it was decided to retain all items.

The statistical programme used to determine the above was the SPSS statistical package, version 16. The descriptive statistics for the overall results of the seven dimensions are also displayed in Table 2.

If the cut-off point of a mean of 3.20 is used, which can be seen as a reasonable cut-off point to differentiate between positive and negative perceptions (Odendaal & Roodt, 1998), only information sharing and change were below the cut-off point with organisational trust very close to the cut-off point with a mean of 3.2158.

Analysis of variance or t-test

G-graphs were compiled to determine whether the results were distributed normally and Levene's test was used to test for homogeneity of variance, as the Levene's statistic tests the null hypothesis that the group variances are equal. These results would indicate when it would be deemed expedient to use non-parametric statistics for further analysis. Non-parametric tests are usually used in the case of small samples, which was not the case in this study, but as this research involved the use of ordinal-level data (Likert-type response scales), it was an acceptable alternative to the more stringent parametric tests (Pett, 1997).

T-tests were used to compare the means of two groups. If the means of more than two groups had to be compared, an analysis of variance (Anova) was used as 'Analysis of variance tests the hypothesis that the group means of the dependent variable are equal' (SPSS, 2007, p. 184).

Table 3 gives an overview of the descriptive statistics of the four sectors. These results show positive results for the dimensions of trust relationships, teamwork and work support but low mean scores for information sharing and organisational trust.

The most positive sector appears to be services, with all the mean scores above 3.30. The lowest mean scores are for information sharing (government and retail or wholesale sectors) and change (government sector). The lowest scoring dimension that stands out is change but only in the case of the government sector. In contrast, it is important to note that information sharing is the lowest scored dimension for all the sectors. The highest scored dimensions are trust relationships and team management for the services sector. Government appears to be the sector with the lowest trust as four of the seven dimensions are below the proposed cut-off point of 3.20.

To determine whether the mean dimension scores differ significantly among the four sectors, Table 4 provides the results of the Anova. Except for the trust relationship and work support dimensions, the other dimensions (change, team management, organisational trust, information sharing and credibility) show significant differences across all four sectors.

The descriptive statistics for the job levels are presented in Table 5. Both an Anova as well as the Kruskal-Wallis test were conducted. No significant differences emerged by job level overall or by job level per economic sector. It is, however, important to note that both supervisors and 'all other levels' experienced change below a mean of 3.00, for three of the job levels (top management and executives, senior management and middle management); information sharing was also perceived as the lowest ranked dimension. Another interesting tendency is that all job levels perceived the dimensions of trust relationship, team management and work support as positive (mean score above 3.20).

To investigate the main question, that is to say, whether the global financial meltdown in the latter half of 2008 has already had an effect on trust levels, two distinct sample periods could be included, namely that before July 2008 and that after July 2008. The descriptive statistics of the two sample periods are presented in Table 7.

According to the Levene statistic (see Table 8), equal variances could not be assumed, so that it was necessary to use the Kruskal-Wallis test as this test makes 'minimal assumptions about the underlying distribution of the data' (SPSS, 2007,

p. 1259) and 'tests whether k independent samples defined by a grouping variable are from the same population' (SPSS, 2007, p. 1266).

The Kruskal-Wallis test (see Table 8) indicated no significant differences for the two sample periods. It is again important to note that information sharing and change were below the 3.20 cut-off point for both periods, while organisational trust was also below this cut-off point for the second period.

This study should be viewed as part of a larger effort by the researchers to understand the possible influence of the passage of time and the impact of certain demographics on trust relationships.

The current results make several contributions. Firstly, concerning the first hypothesis regarding the time dimension, the findings indicate no significant differences between the different trust dimensions for the two measured periods. This could point to the stability of the trust construct, but very low scores for the dimension of organisational trust for the second time period might be an indication that the economic recession may be starting to influence the trust relationship between top management, the immediate manager and employees at a lower level.

There might be various reasons why no significant differences are found between the two time periods, but this still does not support the notion that the trust construct is stable over time. One reason might be the increased economic and financial pressure on companies during the second half of 2008. This might have forced management to focus more on the measured dimensions to maintain a positive relationship and profitability. During the second half of 2008, the economic and financial pressure on organisations had also just started to have an impact and no immediate actions such as liquidations, downsizing or mergers were announced. The authors anticipate that the planned analysis of the data to be collected in 2009 and later might show the influence that the recession has on trust levels. If this effect does not materialise, however, it might point to the stability of the trust construct even during times of economic recession.

Secondly, the differences among the sectors show that government employees experience two of the dimensions as significantly less positive than all three of the other sectors (see Tables 5 and 6), and three dimensions are significantly more negative than the primary and services sectors but, interestingly, similar to the retail and wholesale sectors. It therefore appears as if government sector employees are the most distrustful of their management and the organisation. As indicated in the introduction, this may be attributed to the more focused approach of government to implement employment equity. According to Israelstam (1999), the implementation of employment equity will lead to the changing face of organisations, which will no longer be dominated by white supremacy. At the opposite end of the spectrum, white males feel that they are being victimised by employment equity (Düweke, 2004). These males might respond by leaving the organisation or, worse still, remaining in the organisation and covertly resisting the employment equity plans and their implementation (Israelstam, 1999). Trust levels may also be affected by managers' making use of the generous severance packages being offered as part of employment equity programmes and the resultant outflow of key skills. One would also need to investigate whether the reported incidences in the popular press of nondelivery of services to the public, as a result of real or perceived mismanagement or corruption, lead to feelings of low trust in superiors, or the public sector as employer, by public servants at all levels. All the above could have a more negative effect on the perceptions of government employees.

A study by Careers24.com (2008) in South Africa, completed online by 21 000 respondents, showed similar tendencies. The survey included questions on leadership, which measure the company's commitment to its people, the time managers spend on people issues and the quality of the relationship between managers and employees. All four of the government institutions that participated showed the lowest survey results of all the participating sectors for the dimension of leadership.

The fact that no significant results were found among the job levels shows that all employees experience the results consistently. These results differ from those of research conducted by Bews and Uys (2002), who found that employees at higher job grades have more trust in their direct supervisor, who tends to be part of top management, than do those at lower job grades. The results of this study also differ from those of a recent diagnosis in a South African organisation conducted by Esterhuizen and Martins (2008). These results indicated a perceived lack of trust between employees and management, with significant differences among the results of the job levels for leadership, which includes trust. The reason for the difference in results might be that the present study was done across sectors and not in one organisation, as was the case with the two previously mentioned studies.

Managerial implications and recommendations

• People who mistrust conceal information and act opportunistically to take advantage of others.
•A few mistrusting people can poison an entire organisation.
• Mistrust usually reduces productivity.
• A climate of mistrust tends to stimulate dysfunctional forms of conflict and impede cooperation.
It is therefore important that leaders in government focus more on trust-enhancing behaviours, which Furia (1996) lists as:

• sharing relevant information
• reducing controls
• allowing for mutual influence
• clarifying mutual expectations
• meeting expectations.
Management professor or consultant Fernando Bartolomè offers the following six guidelines for building and maintaining trust (Kreitner & Kinicki, 2007):

• Communication: Keep team members and employees informed by explaining policies and decisions and providing accurate feedback. Be candid about your own problems and limitations. Tell the truth.
• Support: Be available and approachable. Provide help, advice, coaching and support for team members' ideas.
• Respect: Delegation, in the form of real decision-making authority, is the most important expression of managerial respect. Actively listening to the ideas of others is a close second. (Empowerment is not possible without trust.)
• Fairness: Be quick to give credit and recognition to those who deserve it. Make sure that all performance appraisals and evaluations are objective and impartial.
• Predictability: As mentioned previously, be consistent and predictable in your daily affairs. Keep both expressed and implied promises.
• Competence: Enhance your credibility by demonstrating good business sense, technical ability and professionalism.

Research furthermore indicates important findings to repair mistrust in organisations. According to Kim, Ferrin, Cooper and Dirks (2004), individuals will exhibit more trusting beliefs in and trusting intentions toward an accused party if that party apologises rather than denies culpability for a trust violation. The researchers conclude as follows:

Indeed, the implications of such attributions can be so dramatic that they can lead managers to exhibit greater trust in those who have confirmed guilt with an apology than in those who have claimed innocence. (Kim et al., 2004, p. 116)

Another important implication, for government institutions in particular, is the finding that trust is positively correlated with affective commitment. This is a significant predictor of both absenteeism and turnover (Colquitt, Scott & Le Pine, 2007).

According to Dirks and Ferris (2002), when implementing leadership as a mechanism to promote trust in local government, a distinction between different perspectives of trust in leadership should be taken into consideration. This may be either the perceptions of the follower of the relationship with the leaders or a character-based perspective.

The limitations of this study are that only a relatively limited sample participated in the survey. The researchers are currently analysing the results of a follow-up survey, which will lead to a broader database and better comparisons among the sectors and other biographical variables. The main focus of this next phase though is to try to investigate and gain some clarity concerning the apparently elusive effect of time on organisational trust levels. It is suggested that it might be fruitful for other researchers also to investigate the effect of time, as suggested by Schoorman et al. (2007).

Bäckstrőm, L. (2008). Trust relationships: How sales people view the quality of relationship with friends and customers. Management Dynamics, 17(3), 20-27.

Barrick, M.R., & Mount, M.K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44(1), 1-26.

Bews, N., & Martins, N. (2002). An evaluation of the facilitators of trustworthiness. Special edition: SA Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 28(4), 14-19.

Bews, N., & Uys, T. (2002). The impact of organisational restructuring on perceptions of trustworthiness. Special edition: SA Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 28(4), 21-28.

Brown, D. (1992). Why participative management won't work here. Management Review, 81(6): 42-46.

Büssing, A. (2002). Trust and its relations to commitment and involvement in work in organisations. Special edition: SA Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 28(4), 36-42.

Butler, J.K. (1991). Toward understanding and measuring conditions of trust: Evaluation of conditions of a trust inventory. Journal of Management, 17(3), 643-663.

Byrne, J.A. (2002, 24 June). Restoring trust in corporate America. Business must lead the way to real reform. Business Week, pp. 30-35.

Careers24.com. (2008). The salary survey results 2008. Retrieved January 27, 2008, from http://www.careers24.com/TalentSurvey/results.pdf

Castelfranchi, C., & Falcone, R. (1998). Social trust: Cognitive anatomy, social importance, quantification and dynamics. In Proceedings of the Workshop on Deception, Fraud and Trust in Aged Societies: Autonomous Agents Conference, 9 May 1998. St Paul, MN: Springer Verlag.

Colquitt, J.A., Conlan, D.E., Wessan, M.J., Porter, C.O.L.H., & Ng, K.Y. (2001). Justice at the millennium: A meta-analytical review of 25 years of organisational justice research. Journal of Applied Psychology, 86(3), 425-445.

Colquitt, J.A., Scott, B.A., & Le Pine, J.A. (2007). Trust, trustworthiness and trust prosperity: A meta-analytic test of their unique relationships with risk taking and job performance. Journal of Applied Psychology, 92(4), 909-927.

Cyster, S. (2005). A comparative study of the trust audit results of three business units of a South African company. Unpublished thesis, University of South Africa, Pretoria.

Den Hartog, D.E., Shippers, M.C., & Koopman, P.L. (2002). The impact of leader behaviour on trust in management and co-workers. Special edition: SA Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 28(4), 29-34.

Dirks, K.T., & Ferris, D.L. (2002). Trust in leadership: Meta analytical findings and implications for research and practice. Journal of Applied Psychology, 87(4), 611-628.

Du Plessis, M. (2006). The impact of organisational culture on knowledge management. Cambridge: Chandos Publishing.

Düweke, L.D. (2004). The role of employment equity in gender diversity. Unpublished master's dissertation, University of South Africa, Pretoria.

Edwards, T. (2008). The nuts and bolts of ethics, accountability and professionalism in the public sector: an ethical leadership perspective. Journal of Public Administration, 43(3.1), 77-88.

Esterhuizen, W., & Martins, N. (2008). Organisational justice and employee responses to employee equity. SA Journal of Labour Relations, 32(2), 66-85.

Furia, G.L. (1996). Facilitators guide to interpersonal trust surveys. San Francisco: Jossey-Bass, Pfeiffer.

Hay, A. (2002). Trust and organisational change: An experience from manufacturing. Special edition: SA Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 28(4), 43-47.

Israelstam, I. (1999). Problems and pitfalls attached to the implementation of employment equity plans. People Dynamics, 17(7), 14-19.

Kim, P.H., Ferrin, D.L., Cooper, C.D., & Dirks, K.T. (2004). Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations. Journal of Applied Psychology, 89(1), 104-118.

Kreitner, R., & Kinicki, A. (1995). Organizational behaviour. Chicago: Irwin.

Kreitner, R., & Kinicki, A. (2007). Organizational behaviour. New York: McGraw Hill/Irwin.

Kroukamp, H. (2008). Trust building as a leadership pillar in South African local government. Journal of Public Administration, 43(4.1).

Lau, D.C., Lam, L.W., & Salamon, S.D. (2008). The impact of relational demographics on perceived managerial trustworthiness: Similarity or norms? The Journal of Social Psychology, 148(2), 187-205.

Martins, N. (2000). Developing a trust model for assisting management during change. Journal of Industrial Psychology, (3), 27-38.

Martins, N. (2002). A model for managing trust. Journal of Manpower, 23(8), 745-769.

Martins, N., Watkins, M., Von der Ohe, H., & De Beer, M. (1997). Trust audit summary report. Centre for Industrial and Organisational Psychology. Department of Industrial Psychology, Pretoria: University of South Africa.

Mayer, R.C., Davis, J.H., & Schoorman, F.D. (1995). An integrative model of organisational trust. Academy of Management Review, 20(3), 709-734.

Nunally, J.C. (1967). Psychometric theory. New York: McGraw.

Odendaal, A.C., & Roodt, G. (1998). Corporate culture and participation: Independent constructs? Journal of Industrial Psychology /SA Tydskrif vir Bedryfsielkunde, 24(3),14-21.

Pennington, R. (1992). Building integtity-driven organizations. Retrieved September 8, 2001 from http://penningtongroup.com/articles/IntegDrivenLead.asp

Pett, M.A. (1997). Non-parametric statistics for health care research: Statistics for small samples and unusual distributions. London: Sage.

Reina, D.S., & Reina, M.L. (1999). Trust and betrayal in the workplace. San Francisco: Berret-Koehler.

Robbins, S.P. (1996). Organizational behaviour - concepts, controversies and applications. Englewood Cliffs, NJ: Prentice Hall.

Robbins, S.P., Judge, T.A., Odendaal, A., & Roodt, G. (2009). Organisational behaviour: Global and Southern African perspectives. Cape Town: Pearson Education.

Salamon, S.D., & Robinson, S.L. (2008). Trust that binds: The impact of collective felt trust on organisational performance. Journal of Applied Psychology, 93(3), 593-601.

Schoorman, F.D., Mayer, R.C., & Davis, J.H. (2007). An integrative model of organizational trust: Past, present, and future. Academy of Management Review, 32(2), 344-354.

Schumacher, R.E., & Lomax, R.G. (1996). A beginners guide to structural equation modelling. Mahwah, NJ: Lawrence Erlbaum Associates.

Shaw, F.B. (1997). Trust in balance. San Francisco: Jossey-Bass.

SPSS (2007). SPSS 16.0 Command Syntax Reference. Chicago, IL: SPSS Inc.

Tsui, A.S., & O'Reilly, C.A. (1989). Beyond simple demographic effects: The importance of relational demography in superior-subordinate dyads. Academy of Management Journal, 32, 402-423.

Tyler, T.R., & Kramer, R.M. (Eds.). (1996). Trust in organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage.

Von der Ohe, H., Martins, N., & Roode, M. (2004). The influence of credibility on employer-employee trust relations. South African Journal of Labour Relations, 28(2), 4-31.