Page 248 - rise 2017

Measure Items Frequencies (%)
Chinese 4 4.0
India 7 7.0
Marital Status Single 51 51.0
Married 48 48.0
Divorced 1 1.0
Widowed 0 0
Education Level SPM 24 24.0
Diploma 24 24.0
Bachelor’s Degree 43 43.0
Others 9 9.0
Current Position Administrative Officer 11 11.0
Executive 15 15.0
Non-Executive / 61 61.0
Admin Assistant
Technical Executive 7 7.0
Non-Technical 4 4.0
Executive
Secretary 2 2.0
Job Tenure Less than a year 14 14.0
1 – 3 years 44 44.0
4 – 6 years 19 19.0
More than six years 23 23.0
Working Experience Less than a year 5 5.0
1 – 3 years 43 43.0
4 – 6 years 27 27.0
More than six years 25 25.0

Validity and reliability of measurement instrument
The reliability for each measure was examined by computing its Cronbach’s alpha. The reliability of a
measure indicates the stability and consistency of the instrument in measuring a concept and helps to
assess the goodness of a measure (Sekaran, 2000). Sekaran (2000) suggested that the minimum
reliability be set at 0.50. This recommendation by Sekaran (2000) was used in this study as indicators of
internal consistency of the scale. Before the items were submitted for reliability analyses, all negatively
worded items in the questionnaire were first reversed-coded. This was to ensure that all the items
measuring a variable were in the same direction. Table II shows the reliability coefficients for the
variables in this study.

Table II Reliability coefficients for the variables in this study
Number of Items
Variables Cronbach’s Alpha
(n=100)
Job Satisfaction 10 0.92
Extrinsic Rewards 10 0.89
Intrinsic Rewards 10 0.91

The Pearson product-moment correlation coefficients (r) were computed to examine the correlations and
directions among the study variables. This analysis was also conducted to determine the interdependency
of the study variables. For the behavioral sciences, correlations coefficients of 0.10, 0.30, 0.50,
irrespective of the sign, are typically interpreted as small, medium and large coefficients, respectively
(Green, Salkind & Akey, 1997). Very high correlations exceeding 0.90 indicates the existence of
multicollinearity (Hair et al., 1998). Table III shows that job satisfaction is positively and significantly
correlated to extrinsic rewards (0.820**, p<0.01). Job satisfaction is positively correlated with intrinsic
rewards (0.840**, p<0.01) but more than extrinsic rewards. These results show that intrinsic rewards are
more related to UNITEN’s administrative staff as compared to extrinsic rewards. These results also