RESEARCH ARTICLES

 

Exploring employee satisfaction and leadership attributes within the research industries in Gauteng Province

 

 

K Mutsonziwa^I; P Serumaga-Zake^II

^IDepartment of Finance, Economics and Value Chain; School of Business Leadership, University of South Africa; Information and Research Department, FinMark Trust kingstonemutsonziwa@gmail.com
^IIDepartment of Finance, Economics and Value Chain; School of Business Leadership, University of South Africa serumpa@unisa.ac.za

 

 

---

ABSTRACT

The purpose of this study was to identify the key attributes that affect employee satisfaction in the market and social research industries in Gauteng Province, South Africa. The data used for the analysis was collected face-to-face from employees using a quantitative questionnaire. An overall Cronbach alpha of 0.982 showed that the results are statistical reliable.
The study found that a statistically significant relationship exists between employee satisfaction and leadership attributes within the market and social research industries in Gauteng Province. Issues of corporate climate and leadership were important determinants of overall employee satisfaction. The key drivers of employee satisfaction were identified to be job security, commitment or adherence to agreements, relative remuneration, company culture, conflict resolution and ethical considerations by management. Critical leadership factors that were identified included sharing vision of the company into the future, assisting and giving feedback to employees. Exploratory factor analysis results showed that the data was best explained by four factors. The data was also classified into four strategic quadrants derived from the standardised beta coefficients of the attributes derived from regression analysis.
Leadership and management must put in place well thought systems that allow employees to effectively utilise their potential in order to perform at their level best to optimise labour productivity. The researcher recommends that leadership should be delineated across all the different levels within the companies to ensure that all employees understand the operations of the company in order to achieve their optimum operational levels.

Key phrases: employee satisfaction; factor analysis; Gauteng Province; leadership; market and social research; regression analysis

---

 

 

1. INTRODUCTION

It is generally accepted that the effectiveness of any grouping of people is largely dependent on the quality of its leadership. Nel, Gerber, Van Dyk, Haasbroek, Schultz, Sono and Werner (2004:63) argue that the quest in today's globalised companies is for visionary and transformational leaders, and for managers with an above average risk profile.

Further, Nel et al. (2004:63) advise that the challenge for managerial leadership is to recognise which things need to be done timeously and those that need to be done sequentially. Heifetz, Grashow and Linsky (2009:68) propose that to distribute leadership responsibility more widely, one needs to mobilise everyone to generate solutions, by increasing the information flow that allows people across the organisation to make independent decisions and share the lessons they learn from innovative efforts.

Castro and Martins (2010:1) argue that organisations in the 21^st century are facing more challenges than ever before. These challenges are not unique to any specific organisation, but affect all organisations, regardless of their structure or size. To survive and outdo their competitors, organisations are constantly seeking to improve their performance.

Storey (cited in Nyabadza 2008:4) observes that most textbooks on the subject of leadership do not clearly distinguish or delineate between leadership at various levels, leadership of organisations and leadership within organisations. Leaders by themselves cannot stir an organisation effectively without capable employees. Therefore one key asset of any organisation is its employees and keeping them satisfied is of paramount importance.

This paper focuses on managerial leadership aspects and employee satisfaction within the market and social research industries in Gauteng and it is therefore important to note that the leader in this case would be the head of department (director) of a division. An understanding of how employee satisfaction is affected by leadership within an organisation becomes vital.

The broad objective of this paper is to apply statistical techniques to evaluate the level of employee satisfaction in the market and social research industries in Gauteng Province. Specifically the paper aimed to identify the factors that affect employee satisfaction. The specific objectives of this study were as follows:

■ To identify the employee or job-related and leadership-related attributes that contribute towards the overall level of employee satisfaction in the market and social research industries in Gauteng Province.

■ To determine the effect of working conditions on employee satisfaction in the market and social research industries in Gauteng Province.

■ To specifically determine the effect of employee satisfaction on the market and social research industries in Gauteng Province.

 

2. EMPLOYEE SATISFACTION AND LEADERSHIP

The concepts of leadership and employee satisfaction are not new phenomena in research. Studies on employee satisfaction have identified areas that seem to be important for the satisfaction of employees (Abdullah, Musa, Zahari, Rahman & Khalid 2011:150). Küskü (2003:351) advises that employee satisfaction reflects the degree to which an individual's needs and desires are met and the extent to which this is perceived by the other employees.

Some of the components that are important in understanding employee satisfaction within an organisation are (i) culture and organisational issues (ii) immediate manager factors (iii) colleague group factors (iv) information and communication factors (v) the job (tasks and responsibilities) (vi) employee development factors (vii) image and identification factors of the company (e.g. working condition, pay, fringe benefits) (viii) leadership of the organisation.

2.1 Leadership in the South African context of market and social research industries

Many South African organisations have delivered leaders of outstanding quality, who have made a name for themselves and their organisations (Nel et al. 2004:347). However, South Africa maintains a very low profile in the World Competitiveness Surveys, and this shows a need to identify and develop more leaders, and to create organisational cultures which encourage and support quality leadership in our organisations (Nel et al. 2004:347). Unfortunately until now, South Africa remains in this relatively weak position in these surveys.

Leadership within the market and social research industries, like in any other industry, is essential for proper co-ordination and giving of directions to subordinates. In order for employees or groups to function successfully within an organisation, there is a need for effective group leadership and consistency in terms of leadership from one department to the other within the same company.

In South Africa, most market and social research companies within these industries have other offices throughout the country besides their head offices, which are in most cases situated in Gauteng Province. Coordinating with the head office is relatively easy but care has to be always exercised. Nel et al. (2004:347) identify issues around trust, expectations, cultural difference, work coordination and group dynamics which need the leader's attention.

2.2 Identification of key employee satisfaction attributes

Kano's model, which contains the five categories of quality attributes - attractive, one-dimensional, must-be, indifferent, and reverse quality - is widely used by industries and researchers (Yang 2005:1127). The weakness of this model is that it does not take the importance of the attributes measured into account.

There have been some versions and revisions of Kano's model over the years. For example, Yang (2005:1127) refined Kano's model and published a paper entitled "The Refined Kano's Model and its Application". Dixon and Warner (2010:146) mention that although his model was developed primarily to evaluate a customer's response to different product or service features, scholars have demonstrated that Kano's model could also be applied to evaluate an employee's response to different job features or elements.

Yang and Chen (2010:1619-1628) address the deficiency in the service quality model by integrating the "importance" and "satisfaction" indices to establish the "importance-satisfaction model (I-S model)" and provided an assessment model for improving specific quality attributes. In this paper, the researcher used the derived importance of attributes, using regression analysis, in order to classify the attributes into four segments, named as follows:

■ urgent improvement attributes;

■ leverage attributes;

■ maintain attributes; and

■ long-term improvement attributes.

Key variables emanating from the strategic impact grid analysis of the regression model is used to identify key variables measuring and impacting on the employee satisfaction levels. The researchers' approach used derived importance of attributes that influence employee satisfaction. Figure 1 illustrates the strategic impact grid that will be used in this paper. It helps to classify attributes into four quadrants requiring different actions to be taken.

 

 

The four areas of the strategic impact grid are explained as follows:

1. Long-term improvement area: The attributes in this quadrant are those that are considered as being (i) unsatisfactory by the respondents, and (ii) are unimportant in driving the overall employee satisfaction.

2. Urgent improvement area: This is called an urgent improvement area because attributes in this quadrant are considered to be (i) important in driving overall employee satisfaction, and (ii) are rated as unsatisfactory. Hence from a strategic point of view, there is a need to urgently improve these attributes so that they can positively contribute to overall employee satisfaction.

3. Maintain area: The attributes in this quadrant are classified by employees as (i) satisfactory to them but (ii) unimportant to them as well. This might then imply that the business need not take any action with respect to these attributes unless cost pressures require them to do so, and therefore need to "maintain" the status quo.

4. Leverage area: The attributes in this quadrant are often referred to as belonging to the "excellent" area. All the attributes in this quadrant perform (i) satisfactorily, and are also (ii) important in driving the overall employee satisfaction. These attributes would give leverage to the way in which the firm is managed.

 

3. FACTORS THAT WERE CONTROLLED IN THE STUDY

In any research, there are causal factors that can affect the research process and which therefore need to be given sufficient attention before the commencement of the study. This study was not an exception, and it therefore became necessary, as far as possible, to control for such complex variables that influence the results.

In this study, the following variables were identified that needed to be controlled namely (i) industry effects (ii) organisational size and geographical location (iii) natural cultural effects (iv) corporate governance and reputation-related effects (v) demographic effects of respondents.

 

4. RESEARCH DESIGN

In any study, the design for how the research is conducted is very important, as it clearly indicates how the research can be implemented. Quantitative research methodology was used to gain a deeper understanding of what affects employee satisfaction in the market and social research industries of South Africa. The unit of analysis in this study was an ordinary employee who rated the leadership (manager or director in this case). Multifactor Leadership Questionnaire (MLQ) and employee satisfaction questionnaires were adapted thereby linking employee satisfaction to leadership.

For the employee satisfaction questionnaire, the attributes were mainly derived from the Job Descriptive Index (JDI) Questionnaire of Smith, Kendall and Hulin (1969), as well as the Minnesota Satisfaction Questionnaire (MSQ). Other questions that were included into the questionnaires were used to segment the data. A number of dimensions were measured using a 10-point scale, where 1 represented extremely dissatisfied and 10 represented completely satisfied. Some of the employee satisfaction dimensions in the study were (i) working conditions-related attributes (ii) co-workers' (teamwork) attributes (iii) supervision-related attributes (iv) pay-related attributes (v) promotion-related attributes (vi) communication-related attributes (vii) leadership style-related attributes.

After piloting the questionnaire, the structured questionnaires were administered to respondents face-to-face. The researcher used well-trained interviewers for the collection of the data. The quantitative questionnaire took 20 to 30 minutes to administer. All the different stages of data collection were closely monitored and the data was verified before data analysis was done. The completed questionnaires were captured using EpiData software.

Different quality control checks were done in order to ensure that the data was entered correctly. Data analysis was done using Statistical Package for the Social Sciences (SPSS).

4.1 Reliability, validity and ethical considerations for the study

For this study, the overall Cronbach was 0.982 and the 95% confidence interval was (0.978; 0.985) indicating that the questionnaire and the scale were reliable. In terms of validity, the questionnaire had been designed on the basis of related studies and therefore had high reliability and validity. Validity is the strength of conclusions, inferences or propositions made from the data collected and thus refers to the accuracy or truthfulness of a measurement or study done.

It was emphasised to the respondents that the data collected was confidential and only going to be used for academic research purposes only. The researcher also ensured that no physical or non-physical harm was done to respondents and that their privacy and dignity were observed at all times during the data collection process.

4.2 Achieved sample size

A total of 261 employees from the market and social research industries participated in the study. The response rate at the company level was 68% which was good enough for the purpose of this research.

A sample of 261 respondents yielded an overall error rate of 4.9% at the 95% confidence interval which was acceptable statistically. For example, recently, Voon, Lo, Ngui and Ayob (2011:24) investigated the influence of leadership styles on employees' job satisfaction in public sector organisation in Malaysia and used a sample of 200 respondents.

Ponnu and Tennakoon (2009:25) used a sample size of 174 respondents for a study on "The Association Between Ethical Leadership and Employee Outcomes." Furthermore, Ponnu and Tennakoon (2009:25) argue that reliability estimates of Ethical Leadership Scale (ELS) indicated that ELS had demonstrated excellent internal consistency and were stable over studies with respective sample sizes of n=127, n=184 and n=87.

In all the three studies the reliability estimates were above 0.90. Mintzberg (1990)'s total sample comprised 5 chief executives, 56 US foremen; and 160 British middle and top managers. Therefore a sample of 261 achieved in this study was large enough to make any statistical analyses that were done.

 

5. DATA ANALYSIS

The data was analysed using different statistical tools in SPSS. Pearson correlation coefficients were calculated between overall employee satisfaction and each of the attributes that were measured. Correlation analysis helps to ascertain the level of association of dependent variable (overall employee satisfaction) and the independent variables (attributes) and therefore it was necessary to ascertain this association before regression analysis was used. Stepwise regression analysis was used to determine which key attributes drove overall employee satisfaction.

5.1 Correlation analysis and construct validity

All the attributes were positively correlated with the overall employee satisfaction attribute. The whole correlation matrix of all the attributes measured against overall employee satisfaction attribute in this study. The attribute with highest correlation coefficient with overall employee satisfaction was All in all, I am satisfied with my pay (total wages) with a correlation coefficient of 0.732.

This was expected; overall employee satisfaction would be highly associated with remuneration (total wages) and hardly associated to the variable manager is absent when needed most. From the correlation coefficients, it was evident that both job related attributes and manager related attributes were related to overall employee satisfaction and therefore thorough interrogation of the data was needed to be done in order to identify the most critical attributes impacting on overall employee satisfaction.

Correlation analysis was also used to measure construct validity of the results. Firstly, convergent validity showed that all the attributes within their dimension were indeed highly correlated. This means that the attributes were correlated with other measures that they were theoretically predicted to correlate with. Secondly, discriminant (divergent) validity showed that all the attributes which were not in the same construct or dimension, were weakly correlated. Theoretically, attributes that are measuring different aspects should be weakly correlated or should have no relationship.

5.2 Regression of leadership (or manager) related attributes only

When overall employee satisfaction was regressed with all the manager related attributes, a total of five manager-related attributes out of the 39 attribute-set were found to be retained in the regression model. The attributes retained in the model are shown in Table 1.

 

 

The model had a quotient of determination of R²=0.484 and an adjusted R² of 0.474. The t -statistics in the table 1 shows that the attributes in the fitted regression model are significantly important. The RMSE of the model is 1.405 and shows that the error is deviant from the standard unit value. Therefore the regression model might not adequate enough and thus one can conclude that manager related attributes cannot possibly explain overall employee satisfaction by themselves.

However, it is important to point that the model still gives a very clear indication of the drivers of leadership related attributes on overall employee satisfaction. When a model is fitted, it is always important to check how adequate the model is. In regression analysis, the residuals are an important by-product used for checking the adequacy of the model through plotting a histogram and also the P-P plot as shown in Figure 2.

 

 

The model diagnostics shows the graphs of the standardised residuals of the model fitted.

The residuals were approximately normally distributed {N (0,1)} implying that the model is relatively satisfactory although have weak prediction power.

5.3 Regression analysis of all attributes measured in the study

The regression analyses were done independently in order to try and ascertain the contribution of each of the leadership related dimensions towards overall employee satisfaction. What has become clear is that overall employee satisfaction in the Market and Social research industries is more driven by employee and job related attributes in comparison to leadership related attributes. Table 2 summarises the results of the full regression model when all the attributes were included in the model.

 

 

The model had a quotient of determination of R²=0.686 and an adjusted R² of 0.678 and RMSE of 1.906. A total of six attributes were retained in the model. Of the six attributes, two attributes came from the leader or manager related attributes (Considers the moral and ethical consequences of decisions & Agreements are adhered to by leaders). Some of the attributes (independent variables) were correlated with others, a phenomenon called multi-collinearity. The attributes that were most correlated to the other independent attributes were eliminated from the data analysis. Taking into account that the sample size was robust, the effect of multi-collinearity was not a big problem and therefore did not affect the results of the study.

5.4 Model diagnostics

Figure 3 shows the histogram of the standardised residuals. The standardised residuals have a mean of almost zero and a variance of 0.998 which is very close to 1 and therefore shows that the residuals follows standard normal distribution.

 

 

5.5 Strategic impact grid analysis

After running stepwise regression analysis, the standardised coefficients of the betas of the attributes obtained from the regression models were plotted against their means in a graph with four quadrants in order to classify the attributes measured. Most of the measured attributes (51 attributes) in this research fell in the Maintain quadrant. These attributes in this quadrant were classified by employees as (i) satisfactory to them but (ii) are unimportant to them as well. Yang and Chen (2010:1621) mention that although customers/employees are less concerned about these attributes, the performance of the firm exceeds their expectations. There were five attributes that were in the leverage or excellent quadrant are listed below:

■ The culture of the company is good;

■ Agreements made by leadership with employees are adhered to;

■ Conflicts are openly discussed and resolved;

■ Agreements are adhered to by other employees; and

■ Considers the moral and ethical consequences of decisions.

The attributes in this quadrant performed (i) satisfactorily and were also (ii) important in driving the overall employee satisfaction.

Five attributes that were in the long term improvement quadrant namely:

■ My suggestions and ideas are recognised and fairly rewarded;

■ Is absent when needed most;

■ Only acts when there is a problem;

■ Does not give feedback; and

■ My manager needs to improve leadership style.

The one attribute that was in the urgent improvement quadrant was the following:

■ I am satisfied with my pay in comparison to colleagues who do the same job.

Any attribute that is in this quadrant is considered to be (i) important in driving overall employee satisfaction and (ii) was rated as unsatisfactory.

Statements related to remuneration were generally rated low in the survey. This indicates that some of the employees were not satisfied with their remuneration. From a strategic point of view, there is a need to urgently improve this attribute so that it can positively contribute to overall employee satisfaction.

5.6 Exploratory factor analysis

The sample size of 261 allowed factor analysis to be satisfactorily done. Tabachnick and Fidell (2001:91) quote a sample size of 200 as fair to do factor analysis while Costello and Osborne (2005:2) investigated the effect of different sample sizes on their study even using Monte Carlo simulations.

A Kaiser-Meyer-Olkin Measure of Sampling Adequacy of 0.950 was obtained meaning that the measure of sampling adequacy was very good and that the degree of common variance among the variables is marvellous and that the sample was big enough for factor analysis. The measures of sampling adequacy (MSA) calculate the entire correlation matrix and each individual variable evaluate the appropriateness of applying factor analysis. The Bartlett's test of sphericity confirmed that the null hypothesis of an identity correlation matrix was rejected in this study.

5.7 Factor analysis results based on Maximum Likelihood (ML) extraction method

The results based on ML extraction showed that 4 factors were most suitable to the data as opposed to 5 factors. The total variance explained by the model with 4 factors was 61.273% while the cumulative initial eigenvalues accounted for 64.058% of the variation as shown in Table 3.

 

 

Factor 1 had the same attributes (33 attributes) which are named leader motivator attributes.

Factor 2 had day-to-day working conditions variables.

Factor 3 had the same five attributes. This factor is therefore called co-worker issues.

Factor 4 had the same six attributes. This factor could be called demotivator leadership factor issues.

When comparing the factor analysis results from a model with four factors and the one with five factors, the researchers concluded that the 4-factor model was better suited to this study taking into account various considerations as shown in Table 4:

 

 

Adding another factor from four to five factors reduces the Chi Square but also reduces the degrees of freedom. There is a change in Chi Square Test statistics of ∆x² = 413.135 on 58 degrees of freedom. Using this information, one can test the hypothesis of whether there is a significant improvement in adding one more factor.

The researchers were also mindful of the idea of principle of parsimony of model simplicity (i.e. models should have a few parameters as possible) in coming up with the number of factors for the data. Increasing the number of factors in this analysis would not help in understanding the data further as 4 factors already explained the data well without losing any extra information the researchers wanted to know.

5.8 Use of regression analysis on factor scores created

A useful by-product of factor analysis is the factor scores. Factor scores are composite measures that can be computed for each subject on each factor. They are standardised measures with mean of zero and a standard deviation of one, computed from the factor score coefficient matrix. Bartholomew, Deary and Lawn (2009:581) correctly mention that the idea of a factor score has resurfaced from time to time ever since Charles Spearman proposed his theory of 'two factors' to explain individual differences in human intelligence.

Some scholars argue that the factor scores should not be used in regression analysis while others argue that they should be used. However, Bartholomew et al. (2009: 581) argue that Thomson's regression scores have the more relevant justification to be used in regression analysis.

Furthermore Bartholomew et al. (2009:581) conclude that regression scores derived from factor analysis are reasonable predictors or estimators of the factors. In this research, the factor scores derived from the factor analysis were regressed against the overall employee satisfaction and produced the results in Table 5.

 

 

The model has a quotient of determination of R²=0.610 and an adjusted R² of 0.604 showing that the model fitted using the factor scores is relatively satisfactorily good. The model produced a RMSE of 1.222 showing relative closeness to the standard value of 1. All the four factors were retained in the regression model.

What is also interesting from these results is that of the four factors, factor 2 (day-to-day working conditions) is the most important factor while factor 4 (demotivator leadership factor issues) is the least important. Both factor 1 (leader motivator attributes) and factor 3 (co-worker issues) were considered to be equally important in the study.

Thus, it is clear from this analysis that leadership attributes are also important drivers of employee satisfaction. The model diagnostics showed that the residuals followed a standard normal distribution. Thus these results further confirm the adequacy of the four factors' ability to explain the data structure.

 

6. DISCUSSION OF RESULTS

The results showed that there exists an important relationship between employee satisfaction and job related attributes as well as leadership (or manager) related attributes. According to Abdullah et al. (2011:150), there are a number of areas that are important drivers of employee satisfaction. The areas include a well-managed, supportive and prosperous work environment, on-going professional development, career growth potential, challenging and exciting work, teamwork, acknowledgement of work well done, work life balance and the work culture (Tarasco & Damato 2006:37).

This study also found that issues of corporate climate and leadership are very important determinants of overall employee satisfaction. As argued by Stallard and Pankau (2008:19), when leaders infuse the organisational culture with the element of human value, it has ripple effect. Key variables emanating from the strategic grid analysis of the regression model were used to identify key variables measuring and impacting the employee satisfaction levels within the market and social research industries in Gauteng Province.

The strategic impact grid results showed that most of the attributes were in the maintain quadrant. However, the attribute "satisfied with my pay in comparison to colleagues who do the same job" needed urgent attention. The researchers think that this might be an indication of salary discrepancies among different employees within these industries. Regardless of initial impressions and opinions, experience and research indicate work place satisfaction was primarily rooted and engrained in human factors. Although there were a number of attributes that were measured in the research, only six attributes were found to be critical in driving overall employee satisfaction in the Market and Social Research Industries.

The study also showed that leadership is an important determinant of employee satisfaction but the key drivers were mostly job related. This means leadership is a necessary condition but not a sufficient factor to achieve overall employee satisfaction. A holistic approach that encompasses both leadership and employee factors should be well balanced to achieve employee-leadership equilibrium. This means employees should be properly looked after by leadership for them to deliver on their mandates while employees also play their role effectively of doing what they are hired to do. The specific conclusions are discussed next.

6.1 Relationships between employee satisfaction and leadership practices, employee and job related attributes

This research has found that there is a direct relationship between overall employee satisfaction and leadership attributes and practices. The attributes that drove overall employee satisfaction were as follows:

■ conflicts are openly discussed and resolved;

■ the culture of the company is good;

■ considers the moral and ethical consequences of decisions;

■ satisfied with pay in comparison to colleagues who do the same job;

■ agreements are adhered to by other employees; and

■ agreements between leadership and employees are adhered to within the company.

Of all the leadership (or manager) related attributes, Considers the moral and ethical consequences of decisions and Agreements are adhered to by leaders were the only ones found to be drivers of employee satisfaction. The issue of moral and ethical considerations within the industries can be treated as a much broader aspect which encompasses other issues. When the respondents were asked directly what drives employee satisfaction, it was evident that most of the respondents were of the idea that working environment (54%) was the key driver, followed by remuneration at 49%. The other important drivers of employee satisfaction are career growth (27%) and recognition (26%). When the data was reduced using factor analysis, the key drivers were shown to be mainly coming from the job related attributes.

Complementary qualitative results by the researchers showed that the key drivers of employee satisfaction were (i) enjoying the job (working environment) (ii) recognition (iii) achieving goals (iv) remuneration (v) co-workers (vi) open communication (vii) transparency and equitable treatment (viii) skills development and training (ix) employee empowerment (x) benefits and rewards

6.2 Leadership attributes that are most important in determining the overall employee satisfaction

When manager related attributes were isolated from job related attributes, the following leadership or manager related attributes were found to be the key drivers of overall employee satisfaction:

■ talks about most important values and beliefs;

■ explains the vision of the company;

■ talks optimistically about the future;

■ provides me with assistance in exchange for my efforts; and

■ gives feedback.

Factor analysis showed that the data could be reduced into four factors (2 factors comprised of employee and job related attributes and 2 factors comprised of leader related attributes).These are (i) Leader motivator attributes (ii) Day-to-day working conditions (iii) Co-worker issues (iv) Demotivator leadership factor issues.

The study has shown that leadership is important determinant of employee satisfaction but the key drivers were mostly job related. Most importantly, a one fits all model might not be the best in this case, as some of the exogenous (independent) variables might be considered as endogenous (dependent) in another case and hence structural equation models might be preferred.

6.3 Recommendations

Although money is a very important part of our lives, it is not everything that we need especially in the workplace. This study showed that the monetary attribute is important but is not necessarily a sufficient condition to achieve employee satisfaction. The study recommends that leadership and management should put in place well thought systems that allow employees to effectively utilise their potential in order to perform at their level best to optimise labour productivity.

The researchers recommend that leadership should be effectively delineated across all the different levels within the companies to ensure that all employees understand the operations of the company. The researchers further recommend the following:

■ leaders must mentor, care and support their subordinates properly;

■ the industries can adopt a flexi- time regime in order to avoid complications of workloads and working very late;

■ proper remuneration that is competitive to other industries;

■ employees must be appreciated and recognised in their jobs; and

■ supervisors must be knowledgeable and able to assist their fellow employees.

This paper is extracted from a DBL thesis accepted by the School of Business Leadership; University of South Africa, entitled "A statistical model for employee satisfaction in the market and social research industries in Gauteng Province".

 

REFERENCES

ABDULLAH RB, MUSA M, ZAHARI H, RAHMAN R & KHALID K. 2011. The study of employee satisfaction and its effects towards loyalty in hotel industry in Klang Valley, Malaysia. International Journal of Business and Social Science 2(3):147-155.         [ Links ]

AMABILE TM, KRAMER SJ, DIXON R, CANDELL LM, BONABEAU E, BINGHAM A, SCHACT A, HIDARY JD, LITAN RE, MITCHELL L, JENSEN B, KLEIN J, MULLAINATHAN S, ROMER P & ROSS C. 2010. Breakthrough ideas for 2010. Harvard Business Review 88(1):41-57.         [ Links ]

AYDIN B & CEYLAN A. 2009. Does organisational learning capacity impact on organisational effectiveness? Research analysis of the metal industry. Development and Learning in Organisations: An International Journal (Emerald) 23(3):21-23.         [ Links ]

BALZER WK, KIHM JA, SMITH PC, IRWIN JL, BACHIOCHI PD, ROBIE C, SINAR EF & PARRA LF. 2000. Users' manual for the job descriptive index (JDI - 1997 version) and the job in general scales. Bowling Green, KY: Bowling Green State University.         [ Links ]

BARTHOLOMEW DJ, DEARY IJ & LAWN M. 2009. The origin of factor scores: Spearman, Thomson & Bartlett. British Journal of Mathematical and Statistical Psychology 62:569-582.         [ Links ]

BASS BM & AVOLIO BJ. 1997. Full range leadership development: manual for the multifactor leadership questionnaire. Palo Alto, CA: Mind Garden.         [ Links ]

CASTRO ML & MARTINS N. 2010.The relationship between organisational climate and employee satisfaction in a South African information and technology organisation. South African Journal of Industrial and Psychology 36(1):1-9. [Internet: http://www.sajip.co.za; downloaded on 21 April 2011.         [ Links ]]

GIBSON JL. 2007. Truth and reconciliation as social indicators. social indicators research. An International and Interdisciplinary Journal for Quality of Life Measurement:257-281.         [ Links ]

HEIFETZ R, GRASHOW A & LINSKY M. 2009. Leadership in a (permanent) crisis. Harvard Business Review 87(7/8):62-69.         [ Links ]

KÜSKÜ F. 2003. Employee satisfaction in higher education: the case of academic and administrative staff in Turkey. Career Development International 8(7):347-356.         [ Links ]

MATZLER K, RENZL B & ROTHENBERGER S. 2006. Measuring the relative importance of dimensions in the formation of price satisfaction and service satisfaction: a case study in the hotel industry. Scandinavian Journal of Hospitality and Tourism 6(3):179-196.         [ Links ]

MINTZBERG H. 1990. The manager's job: folklore and fact. Harvard Business Review 68(2):163-176.         [ Links ]

MUTSONZIWA K. 2013. A statistical model for employee satisfaction in the market and social research industries in Gauteng Province. Midrand: University of South Africa. (DBL thesis.         [ Links ])

NEL PS, GERBER PD, VAN DYK PS, HAASBROEK GD, SCHULTZ HB, SONO T & WERNER A. 2004. Human resources management. 6^th ed. Cape Town: Oxford University Press.         [ Links ]

NYABADZA GW. 2008. The lived experience of the strategic leader: what effective CEOs do, how they do it and an exploration into how they think about it? Pretoria: University of South Africa. (DBL thesis.         [ Links ])

PONNU CH & TENNAKOON G. 2009. The association between ethical leadership and employee outcomes - the Malaysian case. Electronic Journal of Business Ethics and Organisation Studies 14(1):21-32        [ Links ]

SMITH PC, KENDALL LM & HULIN CL. 1969.The measurement of satisfaction in work and retirement. Chicago, IL: Rand McNally.         [ Links ]

STALLARD ML & PANKAU J. 2008. Strengthening human value in organisational cultures. Leader to Leader:18-23.         [ Links ]

STOREY J. 2005. What next for strategic-level leadership Research? Leadership 1(1):89-104.         [ Links ]

TABACHNICK BG & FIDELL LS. 2007. Using multivariate statistics. Boston, OH: Allyn & Bacon.         [ Links ]

TARASCO JA & DAMATO NA. 2006. Build a better career path. Journal of Accountancy 201 (5):37.         [ Links ]

TSUI A. 2006. Time to rethink the new employee relationship. Tempe, Phoenix, AZ: Arizona State University (WP Carey School of Business.         [ Links ])

VOON ML, LO MC, NGUI KS & AYOB NB. 2011. The influence of leadership styles on employees' job satisfaction in public sector organisations in Malaysia. International Journal of Business, Management and Social Sciences 2(1):24-32.         [ Links ]

WEGNER T. 2007. Applied business statistics: methods and excel-based applications. 2^nd ed. Juta: Cape Town.         [ Links ]

WRIGHT TA & CROPANZANO R. 1998. Emotional exhaustion as a predictor of job performance and voluntary turnover. Journal of Applied Psychology 83:486-493.         [ Links ]

WRIGHT S. 1921. Correlation and causation. Journal of Agricultural Research 20:557-585.         [ Links ]

YANG CC. 2005. The refined Kano's model and its application. Total Quality Management 16(10):1127-1137.         [ Links ]

YANG KJ & CHEN SH. 2010.The comparison and analysis of employee satisfaction improvement in the hot spring and financial industries. African Journal of Business Management 4(8):1619-1628. [Internet: http://academicjournals.org/AJBM; downloaded on 2 May 2011.         [ Links ]]