This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
In this chapter, the results of the empirical study are reported. Results are presented in respect of the relationship between satisfaction with performance appraisal system and employee outcome, in the form of work performance within the Mauritian context. The results will provide the basis for rejection or confirmation of the research hypothesis indicated in the Literature Review.
The following empirical aims have been identified:
a) To determine the empirical relationship between satisfaction with performance appraisal system and employee outcome in term of the work performance of the employees at the Companies Division that forms part of the public sector.
b) To determine if intrinsic motivation will mediate the relationship between satisfaction with performance appraisal system and employee outcome.
The hypotheses are as follows:
H1: Satisfaction with PAS will positively influence Work Performance.
H2: The relationship between Satisfaction with PAS and Work Performance is mediated by Intrinsic Motivation.
4.2.0 Frequency Analysis
4.2.1 Analysis of Population in terms of Gender
Fig 3: Gender
The sample involved in this present study was 100 individuals who are working in 14 departments at the Companies Division. From the total respondents of 100, it can be noticed that there is not an equal proportion of male as compared to female. The Male section represented a percentage of only 16% representing 16 Men and the Female section represented a percentage of 84% representing 84 Women as shown in Figure 3 and Appendix E. This clearly shows that the ratio of Man compared to Women is 4:21. But still the researcher tried as far as possible to get the response of all the men working at the Companies Division.
4.2.1 Frequency Analysis of Age Group
Fig 4: Age Group
From the sample, majority of the respondents were from the middle age group that is the range of 25-29, 30-34 and 35-39 (76 percent combining the three age range together) representing a total of 76 people out of 100 whereas only a small percentage (24 percent) was from the younger and older group representing 24 people out of 100 as shown in Figure 4 and also in Appendix E. This represents a ratio of 19:6(Middle Age: Others)
4.2.2 Frequency Analysis of the Education Level of the employees
Fig 5: Educational Level of Staff
From the sample it can be noted that 51 respondents representing 51 % of the respondents hold a degree and 15 respondents representing 15 % of the respondents have a diploma. This is shown in Fig 5 and Appendix E. From these data, the researcher has found that the Companies Division has highly educated staff. It represents a ratio of 33:17(Degree and Diploma combined: Others)
4.2.3 Frequency Analysis for the Years of Experience of the Employees
Fig 6: Years of Experience of Staff
With reference to Figure 6 and Appendix F, it was noted that the respondents working at the Companies Division were dedicated to their Company. As per the data obtained it can be deduced that about 76%( combining the experience range of 6-10 years and More than 10 years) of the respondents representing 76 people out of 100 has been working there for 6 or more years. This represents a ratio of 19:6 (Age range 6-10 & 10 or More combined: Other).
4.2.4 Frequency Analysis of the Job Position of the employees
Fig 7: Distribution of Sample based on Job Position
From Figure 7 and Appendix F, it can be noted that 56% of the respondents were either Compliance officers or Clerical officers. This is because there is a majority of Compliance and Clerical Officers as compared to other staffs and this can be clearly observed in the Company Structure of Companies Division( Please refer to Appendix A). But still the researcher tried as far as possible to take 100% of respondents from the job positions having the least staff. This represents a ratio of 14:11(Compliance Officer and Clerical Officer combined: Others).
4.3.0 Factor and Reliability Analysis
Factor analysis is the explanation of multivariate investigation that requires a large number of variables or items to identify a small number of factors that gives an explanation about the interrelations among the variables or items. This is used to find out the number of components in a set of data. These components are then categorised according to their specific characteristics. This process gives the researcher the opportunity to convert information into statistical groups.
For the present study, the researcher performed a principle component analysis on three types of variables being studied namely
the independent variable which is Satisfaction with PAS,
the mediating variable which is Intrinsic Motivation,
the dependent variable which is Employees' Work Performance
The Researcher performed principle component analysis with Varimax rotation on these variables. The main aim of conducting the principle component analysis is to find out whether the items found in the questionnaire considerably contribute to the variables which they evaluate. In addition, this analysis was used to refine the data obtained to the required scales.
Eigen values of greater than 1.0 and item loading of greater than 0.30(Coakes & Steeds, 2007) are generally the established criteria. The researcher omitted the variable item that is questions that failed to exceed 0.30 from consideration. KMO measures sampling adequacy which is greater than 0.5 is acceptable as per Kaiser (1974). Moreover, the Barthlett "Test of Sphericity" is highly significant when p<0.001 and hence the factor analysis is appropriate. To be able to measure the consistency of the scale used in the questionnaire, the researcher used Cronbach's alpha as a measure of reliability. Cronbach's alpha allowed the researcher to measure how well a set of variables measure a single factor.
Cronbach's alpha is usually low when data have a multidimensional composition. In principle, Cronbach's alpha is not a statistical analysis but a coefficient of reliability.
Hair, Black, Babin, Rolph, Anderson and Tatham (2006) noted that coefficients of 0.7 or more are considered satisfactory.
4.3.1 Factor and Reliability Analysis on Satisfaction with Performance Appraisal System. (Independent Variable)
With reference to Appendix G, as anticipated by the researcher, the items for the Satisfaction with PAS fitted perfectly into one factor, with an initial Eigen value of 3.830 and the total variance described was 54.713% of the total variance. Sufficient inter-correlations have been noted because the KMO measure of sampling adequacy was 0.742. Bartlett's Test of Sphericity illustrated Chi-Square of 450.158 and was statistically significant as p<0.001. Since the Cronbach's alpha coefficient is 0.831, we can say that the scale is reliable.
4.3.2 Factor and Reliability Analysis on Intrinsic Motivation (Mediating Variable)
With reference to Appendix H, as expected by the researcher, the items for Intrinsic Motivation fitted perfectly into one factor, with an initial Eigen value of 3.422and the total variance described was 57.039% of the total variance. Sufficient inter-correlations have been noted because the KMO measure of sampling adequacy was 0.722. Bartlett's Test of Sphericity illustrated Chi-Square of 388.587 and was statistically significant as p<0.001. Since the Cronbach's alpha coefficient is 0.825, we can say that the scale is reliable.
4.3.3 Factor and Reliability Analysis on Work Performance (Dependent Variable)
With reference to in Appendix I, as expected by the researcher, the items for Work Performance fitted perfectly into one factor, with an initial Eigen value of 3.296 and the total variance described was 54.926% of the total variance. Sufficient inter-correlations have been noted because the KMO measure of sampling adequacy was 0.688. Bartlett's Test of Sphericity showed Chi-Square of 321.113 and was statistically significant as p<0.001. Since the Cronbach's alpha coefficient is 0.817, we can say that the scale is reliable.
4.4.0 Descriptive Analysis
For the purpose of this analysis, Mean and Standard deviation is used by the researcher to explain the characteristics of the data. The calculated average of scores is the Mean. It is the most popular measure of central tendency. Mean is calculated by summing all the scores and dividing the total by the number of scores. It is suitable when the data represents either an interval or a ratio scale and is a more accurate, constant guide than both the median and the mode.
Standard deviation is a measure of distribution of a collection of values. Its application can be found in probability distribution, random variable, a population or a set of data. It is the most frequently used measure of statistical dispersion. It is used to measure how widely spread the values are in a set of data. Standard deviation is small if many data points are close to the mean, but many data are far from the Mean, and then the SD is large. The SD is going to be zero if all the data values are the same.
From calculations done on Excel and by observation, the researcher noted that among the three variables, Satisfaction with PAS was perceived to be the highest with a Mean of 4.4143 and a SD of 0.5218. It was followed by Work Performance with a Mean of 4.4050 and a SD of 0.51201 and Intrinsic Motivation with a Mean of 4.3817 and a SD of 0.51201. These data are illustrated in Appendix M. SD was found to be small that is less than 1 in all variables. This means that the data are tightly concentrated to the Mean.
4.5 Correlations Analysis
Correlation is the connection between variables. It investigates the extent to which two sets of data are linked. Stronger relationship between two sets of data is indicated by higher correlation value. If the correlation between the sets of data is 1 or -1, a perfectly linear positive or negative relationship exists. A correlation of zero will imply that there is no relationship between the two sets of data. When taking into account the correlation between Satisfaction with PAS which is the independent variable and the dependent variable which is Work Performance, if the degree of correlation is large, the stronger the linear association is going to be. This has been noted by Coetzee in 2003. The standard correlation coefficient used is the Pearson's r. This standard correlation is valid mostly to variables scattered essentially along interval or ratio scales of measurement.
From Appendix N, the researcher observed that Satisfaction with PAS is positively and highly correlated with employees' Work Performance with r=0.724 and p<0.01. From the information obtained it is apparent that the subscales for the dependent variable which is employees' Work Performance is highly inter-related and hence ascertain the strength of the subscales.
4.6 Regression Analysis
Regression is the evaluation of the relationship which exists amid two quantitative variables. Regression can only occur with interval or ratio data. If an independent variable is placed against a dependent variable on the X and Y axis of a graph with the actual data spread on the graph, it is possible to draw a line through the resulting points in such a way that it can reduce the distance between the points. The line obtained can be either a straight line or a curved line. This line is called the regression line. The value of the dependent variable which in the present case is the employees' Work performance can be calculated by multiplying the value of the independent variable which is Satisfaction with PAS by the regression coefficient which is a number that determine the gradient of the line. Regression helps to predict a variable provided that we already have a known variable.
The hypotheses previously discussed in the Literature Review and in the introduction part of Analysis and Findings, were tested using the regression analysis. The research first of all regressed the employees' Work performance which the dependent variable onto the independent variable which is Satisfaction with PAS in order to test the direct effect hypothesis. After having regressed the dependent variable onto the independent variable, the researcher entered the mediating variable which is Intrinsic Motivation. The researcher used of the Three-step procedure recommended by Baron and Kenny (1986) to estimate the mediation hypotheses.
Appendix O and P present the results obtained from the regression analysis conducted by the researcher. He first tested the direct relationship between Satisfaction with PAS and employees' Work Performance which is shown in Appendix O. It can be observed that Satisfaction with PAS is positively influenced by employees' Work Performance with a Regression value (Î²) of 0.724 and p<0.001. Therefore it can be deduced that the hypothesis H1: Satisfaction with PAS will positively influence Work Performance is accepted.
The three -step procedure which Baron and Kenny (1986) suggested to test the mediating relationship consists of a sequence of steps which needs to be followed strictly. First of all, the independent variable which is Satisfaction with PAS should be significantly related with Intrinsic Motivation which is the mediating factor. As shown in Appendix P, the first condition for the success of the procedure mentioned above has been met. The independent variable is linked with the mediating factor with a positive Regression value Î² of 0.291. Subsequently, the independent variable (Satisfaction with PAS) must be linked significantly with the dependent variable (employees' Work Performance). The linkage between the independent variable and the dependent variable has already been done and is displayed in Appendix O. Ultimately, it was noted by Baron and Kenny (1985) that after having introduced the mediator (Intrinsic Motivation) in the regression model, the relationship between the independent (Satisfaction with PAS) and dependent variable (employees' Work Performance) ought either to disappear (full mediation) or lessen considerably (partial mediation). From the information displayed in Appendix O and P, the researcher figured out that the condition mentioned above has been met for all the variables. The relationship Satisfaction with PAS and employees' Work Performance is fully mediated by Intrinsic Motivation. Therefore, H2: The relationship between Satisfaction with PAS and Work Performance is mediated by Intrinsic Motivation is accepted.
4.7 Analysis of the Structured Questions
Q1. What are your views about the Appraisal System in your organization?
The researcher compiled and worked out the information he got from the questionnaires, he found that about 43% of the respondents representing 43 persons out of 100 think that the PAS at the Companies Division is well organised in terms of feedback and also review of their performance. They also stated that they tally with the goal setting and expectations of an individual.
37 percent of the respondents (37) people think that it is important to have a formal form of Performance Appraisal. According to them, PAS helps them to understand their long-term goals and it also helps them to understand the mission, vision and goals of Companies Division. PAS helps them to better understand their strength and weaknesses and also gives them an opportunity to better understand their duties and responsibilities.
20 percent of the respondents think that PAS is very important to them because it is related to their pay and reward. It also helps them to identify training and development needs and that it also motivates them to work harder to be able to get promotion. They also believe that PAS allows them to know their actual performance and also what other people think of them.
With reference to the answers given by the respondents, the researcher noticed that the employees were satisfied with the actual PAS because it was designed for their individual growth and development in the organisation. The PAS gave them the required direction to be able to improve their work and also the vital skills which they require to perform their work well.
Q2. Please explain how this appraisal process would help you perform better?
The Performance Appraisal System Currently being used at the Companies Division is perceived as a source of improvement for 53 percent of the respondents. Since Companies Division deals mainly with the public, therefore they believe that the PA process is beneficial to them in terms of negotiations and interaction with the public.
According to the 47 percent of the respondents, the PA process helps them to identify their improvement area and their strength. This identification help them to keep up the good work which they are doing and also assist them to focus on their improvement areas.
Q3. How does the Appraisal Process motivate you to perform better?
Among the 100 respondents, a large majority of persons representing a percentage of 67 percent (67 respondents) said that PAS is a system which offers reward in terms of bonus and recognition for good and acceptable performance and behaviour. It helps them to be more motivated to work better and also be able to improve in areas needed. They also perceive PAS as a system which helps them to keep doing the good work that they are already doing.
The remaining 37 percent (37 respondents) believes that PAS positively motivates them. With the PAS, they get a clearer idea of where they are and where they should be. It also helps them to identify their strength and also on which work areas they should they should improve. PA process motivates them to either keep on or improve their actual rating by better performance.
From these answers, the researcher noted that employees were motivated to perform well in the organisation. He also found that they take extra effort to deliver outstanding performance because the respondents believe that high motivation for better performance would give them bonuses and recognition. A good performance feedback helps to boost the employees for them to effectively and efficiently utilize their abilities and skills to perform better in the future.