TRUST AS A FACTOR IN DETERMINING HOW TO ATTRACT, MOTIVATE AND RETAIN TALENT

The aim of this study was to validate a questionnaire that can be used to determine how employees select the best company to work for. The second focus was to determine the role of trust in a relationship where employers must attract, motivate and retain employees. The confirmatory factor analysis resulted in 10 dimensions that supported most of the theoretically constructed dimensions. A second-order factor analysis was done and it became clear that there are two second-order factors underlying factor 1, namely leadership and trust. A strong correlation was found between trust and the dimensions of job satisfaction, relationships and leadership.

calculative trust -salience of trustworthy exchanges for individual benefit relational trust -an organisation's history and mythology of trustworthy behaviour institutional trust -an established system of rewards and sanctions corresponding to trustworthy behaviour a key characteristic of high performance teams trust in leadership as well as the leader's trusting other people to do what needs to be done influence of trust in institutions at the societal level on the organisational trust climate, (Trust climate is seen as one of the dimensions of organisational trust) a multi-dimensional typology at individual, interpersonal and institutional levels The present study examines trust as one dimension of the factors that employees consider when choosing a company to work for. In this context aspects such as the trust relationship between the immediate manager and subordinates, trust between employees, respect, care and motivation were included. The relationship between trust and other dimensions was also determined.
In the last few years South African organisations have experienced drastic changes, re-engineering, downsizing, ferocious global competition and the implementation of various new labour laws. These changes have led to a shift on both sides of the employment contract which is leading to a widening of the trust gap between employees and their employers (Martins, 2000). Employers can only speculate what the impact of the widening trust gap is on attracting and retaining talented workers. According to Pfeiffer (1991) trust is emphasized in the philosophies of renewed organisations.
Various criteria are used to determine the best or most admired companies to work for, for instance: pay and benefits, opportunities, job security, pride in work and company, openness and fairness, camaraderie and friendliness (Lau & May, 1998) completing a trust index, receiving supplementary material such as newsletters, employee handbooks and videos (Branch, 1999) quality of management, quality of products and services, innovativeness, long term investment value, financial soundness, ability to attract, develop and retain talent, community responsibility, use of corporate assets and global business acumen (Stein, 2000) macrocriteria: financially stable, enjoy success in its markets, enjoys status and reputation, salary/benefits/work environs, recruitment, career opportunities and development, job security/handling corporate change and the human face of the company (Donald, 2000) The above suggest that various methods are used to determine what factors attract, motivate and help retain talented people. In an attempt to establish what these methods are, research was conducted to determine the factors that attract and retain employees. The changing workforce composition and changing work environment also encouraged the researchers to investigate the role of trust as a factor in attracting and retaining employees.

METHOD Participants
In total, 1984 employees from 42 companies participated in the survey. Fifty-five percent of respondents had less than four years of service: the majority (75,9%) had been employed for less than 11 years. More than 60% (62,6%) of the population was over the age of 30 with the majority (36,5%) falling between the ages of 31 and 45 years. A quarter of the respondents (24,7%) were from the ranks of middle management, professionals/specialists (26,6%) and general staff (26,2%).
Developing the measuring instrument Deloitte and Touche, Human Capital Corporation sponsored a project to determine the best companies to work for in South Africa. As this was the first project of this nature in South Africa it was decided to first develop a theoretical model. The theoretical model was used to develop dimensions and statements to compile a questionnaire. In order to differentiate the survey from other similar initiatives it was decided to include both employer and employee perceptions and also external market perceptions in the final analysis. This study focuses only on the validation of the employee questionnaire.
The following steps were used to compile the employee perception questionnaire: A project team selected aspects that might have an impact on attracting and retaining employees. A theoretical model (Dicks, Smith & Martins, 2001, p. 51) was developed based on work being done in the USA in particular (figure 1). Questions were compiled and grouped into dimensions. An independent panel of experts from labour and the private sector and academics checked the prepared questionnaire for face validity. Questions and dimensions that did not satisfy the panel were excluded from the questionnaire.
The final questionnaire consists of 101 questions that were grouped into 12 dimensions.
The following dimension were identified: diversity change trust management information leadership rewards and recognition policies and procedures development and training job satisfaction change which has occurred communication sources The items (questions) are endorsed in a five point interval scale, the lowest anchor signifying not applicable, the next anchor signifying strongly disagree and the highest strongly agree. The scale ranges from 0 to 4.

Procedure
The questionnaire was distributed to interested companies for completion. A covering letter explaining the purpose of the survey was included with the questionnaire. A contact person in each company assisted with the distribution and collection of the questionnaires. The guideline for completion was 50 to 200 randomly selected employees. The respondents were assured that the data would be kept confidential and anonymous.
The overall purpose of this phase was to validate the original theoretical model of attracting and retaining employees (figure 1). The second phase was to determine if there are any correlations between trust and the dimensions of attracting and retaining employees. The business focus was to determine the top companies to work for.

RESULTS
In the organisational literature and commonly in role-based research, factor analysis is frequently used to assess whether instruments measure substantive constructs (Cortina, 1993;Drasgow & Miller, 1982). Hair, Anderson, Tatham and Black (1995) argue that the analyst has perceived thoughts about the actual structure of the data, based on theoretical support or prior research. The analyst may test hypothesis involving issues such as which variables should be grouped together in a factor or the precise number of factors.
In these instances, the analyst requires factor analysis to take a confirmatory approach -that is, to assess the degree to which data meets the expected structure of the analyst. With the original factor structure in mind, 10 factors were used as the criteria for the factor analysis. A priori criterion is based on the reasonable criterion of the process applied to construct the theoretically constructed factors. A varimax rotation was used and the 10 extracted factors accounted for 49% of the variance (table 1). The factor loadings of the items within each sub-scale are given in table 2. The items with their factor loadings are reported and are sorted by dimension and strength of factor loading within dimension. Factor loadings greater than 0,30 were considered to meet the researchers minimum level to be included in a factor. In the assessment of how items load on the 10 factors identified with the varimax rotation, factor 1 has one item that does not meet the 0,3 criterion. All the items of factor 2 meet the 0,3 criterion, while two items of factor 3 do not meet the criterion. One item of both factor 4 and 5 do not meet the criterion. Two items from factors 6 and 7 do not meet the 0,3 criterion. All the items of factors 8 and 9 meet the criterion. The results of the factor analysis indicate that 10 items can be excluded from the measuring instrument.  Re-structuring of the organisation. 0,483 The implementation of gender equality as part of affirmative action. 0,432 Change to a total package approach to pay. 0,428 Granting of more decision-making power to lower organisation levels.  A second-order factor analysis was conducted for factor 1. The results and eigenvalues show two factors above the 1,0 eigenvalue (table 3). The scree plot also indicated only two factors. A correlation analysis to determine the correlation between the dimensions of trust and the other dimensions is displayed in table 4. Trust correlates the highest with the dimensions of job satisfaction and leadership. Conceptual naming of factors Conceptual naming of the 10 factors produced the names as displayed in tables 4 and 5. Each factor will be discussed briefly.

Leadership/Trust
In the preliminary model trust and leadership were seen as two separate dimensions. The factor analysis revealed a close relationship between leadership and trust as determined by the results of the second-order factor analysis. This supports researchers such as Charlton (2000), Katzenbach and Smith (1993) and Martins (2000), who argue that trust influences leadership. Furnham and Gunter (1993) identified trust as one of the eight dimensions of the universe of psychological perceptions. The people I work with are committed to quality work. 0,322 There is good co-operation between my department and other 0,240 departments in my organisation.
The internal system is overloaded with unnecessary paperwork, 0,232 e-mails, reports, forms and memos. In this organisation promotions go to those who best deserve them. 0,299 (2000) the continuing capacity for an organisation to change has an impact on employees' decisions to remain with an organisation.

Job satisfaction
This factor retained eight of the original nine statements. Sparrow (2001) sees job satisfaction as an important part of the psychological contract between the employee and the employer.

Management of change
Only one statement was excluded from the original factor. This factor focuses on the impact of change on individuals and the organisation. Various researchers have included the concept of change in organisational assessments, for instance Dastmalchin, Blyton and Adamson (1991) (in Furnham &Gunter, 1993 andDenison, 2001).

Policies and procedures
The factor analysis identified 10 policies and procedures which are mostly in line with the original factor. Only one statement loaded too low on the item analysis to be included. This factor is not typically included in organisational surveys that focus on climate/culture. However, the number of new Acts that have been implemented in South Africa have forced organisations to update, change and improve their policies and procedures and to align themselves with the latest legislation. These new policies and procedures have an effect on the organisational environment and on internal-external relationships. According to Sparrow et al. (2001), climate instruments tap into the practices and procedures.

Management communication
The original 10 statements were retained after the factor analysis. Communication is seen as one of the important components of most organisational assessments (Furnham & Gunter, 1993).
Relationships Table 5 shows that this dimension differs from the original theoretical-concept ual factor. The newly formed factor focuses on the relationship between employees and their immediate managers and on the relationships between departments. Kays and De Cottis (1991) (in Furnham & Gunter, 1993, p. 152) summarised the various psychological climate dimensions and found that recognition is labelled by many researchers as the summary dimension for rewardpunishment relationships.

Rewards and recognition
The original five statements were maintained after the factor analysis. Rewards and recognition are included in most climate or organisational surveys to determine employees satisfaction/dissatisfaction with rewards and recognition (Furnham & Gunter, 1993).

Development and training
This factor grouped all aspects relating to the development and training of employees together. Six statements were grouped together. The concepts of jobs/skills match, support and empowerment are all seen by researchers as dimensions that relate to the theme of development and training (Charlton, 2000;Kays & De Cottis (in Furnham & Gunter, 1993); Kraut, 1996).

Diversity
Only five of the original 11 statements were retained. One statement was excluded due to a low loading on the item analysis. The theme of diversity was not included in most original organisational assessments. But in view of the changing South African environment it is now included.

DISCUSSION
The results obtained in the present study supported most of the theoretically constructed dimensions. The confirmatory factor analysis resulted in 10 dimensions that were, however, grouped somewhat differently. The main changes in the grouping of the items occurred for factors 1 and 7.
The grouping of the remaining eight dimensions generally corresponded with the original theoretically constructed dimensions. The factors yielded reliabilities ranging from 0,796 to 0,932. The 10 dimensions that emerged were identified as dimension 1: leadership/trust, dimension 2: change occurred, dimension 3: job satisfaction, dimension 4: management of change, dimension 5: policies and procedures, dimension 6: management communication, dimension 7: relationships, dimension 8: rewards and recognition, dimension 9: development and training and dimension 10: diversity.
It is clear from the second order factor analysis that was done on the first factor, that there are two second-order factors underlying factor 1, namely leadership and trust. Both these second-order factors loaded positively on the scree plot and the eigenvalues are above 1. The second-order factor of trust focuses mainly on aspects such as manager/subordinate trust relationships, departmental/sectional trust relationships, departmental motivation, respect and care. This supports some of the components of trust as identified by other researchers working on the concept and characteristics of trust (Charlton, 2000;Katzenbach & Smith, 1993;McKnight & Webster, 2001).
A correlation analysis bet ween trust and the other dimensions indicates high relationships bet ween job satisfaction, relationships and leadership. The present study thus generally confirms research which indicates that trust is created by leadership which influences relationships and job satisfaction.
As far as the theoretical model is concerned, it was found that all the factors do not correlate equally and that there is a stronger focus on some dimensions (leadership, change occurred, policies and procedures, management communication and relationships) than on others (rewards and recognition, diversity and development and training). The implication of this is that the original framework (figure 1) needs to be adapted to the present research. Structural equation modelling could be considered to confirm and/or create a new model.
In conclusion, the present study opens up new research possibilities, for instance understanding the criteria that influence an individual's attraction to an organisation and how to retain talent. The present study emphasises the importance of leadership in attracting and retaining talent. The concept of trust needs to be expanded to include more items that focus on aspects such as loyalty, integrity, congruence between individual and organisational values and the role of organisational culture in retaining and attracting talented people.