Reliability Tests and Reliability analysis statistics

Reliability Tests and Reliability analysis statistics

This is the degree to which measures are free from random error and hence providing consistent

data over time. It is the consistence, stability or dependability of a data set. Internal consistency

reliability was used to assess the ability to produce similar results when different samples are used

to measure a phenomenon at the same time period (McDaniel & Gates, 2010). The questionnaire

was piloted by testing it on ten members of staff at Kenya Bureau of standards Headquarters in

Kisumu. The Cronbach’s Alpha Coefficient assessed the internal consistency of the instrument

with alpha coefficients of above 0.7 implying reliability (Cronbach, 2004; Nunnaly,1978). George

and mallery (2000) provide the following rules of thumb; > 0.9 ─ Excellent, > 0.8 ─ Good, > 0.7

Serial

No.

Scale Items Cronbach Alpha ( α )

1 Employee age 3 0.829

2 Human Resource Management Practices 42 0.9114

3 Employee Competence 21 0.789

4 Employee performance 16 0.885

5 Average/mean 73 0.855

Human Resource Management Practices had the best internal consistency recording a cronbach

alpha reliability coefficient of 0.9114. It was followed by Employee performance at a cronbach

alpha reliability coefficient of 0.885, employee age 0.829 and employee competence recording

a cronbach alpha reliability coefficient of 0.815, middle age employees 0.813 and older

employees 0.861.

Reliability analysis for components of employee age

Serial

No.

Employee age Measures Items Cronbach

Alpha ( α )

1 Younger employees (18 yrs – 39 yrs) 1 0.815

2 Middle age employees (40 yrs – 49 yrs) 1 0.813

3 Older employees (50 yrs – 60 yrs) 1 0.861

3 Average/mean 3 0.829

table 4.4 giving the alpha coefficient of employee participation and empowerment, with twelve

items as 0.936, employee training and development with eleven items as 0.923, adequate and fair

compensation with seven items as 0.904, employee welfare benefits with five items as 0.863 and

performance management with seven items as 0.931. The overall average/mean of the

components of HRM practices was recorded as 0.9114 which was a very high alpha coefficient

implying a very high internal consistency.

Test results for Reliability for Components of HRM practices

Serial

No.

HRM practices measures Items Cronbach

Alpha ( α )

Average/mean 42 0.9114

Table 4.5 presents alpha coefficients for the five components of employee competence provided

for the study. A total of 26 items were presented to respondents in a pilot study and 21 of the

items were responded to and contributed to the overall reliability of the components of employee

competence. The overall cronbach alpha reliability coefficient of 0.789 was recorded, indicating a

fairly strong internal consistency among the 21 sub variables of employee competence.

Tests of Reliability of the components of employee competence adopted for the study are

summarized in table 4.5 giving the alpha coefficient of educational level with five items as 0.741,

technical skills with four items as 0.775, general skills with four items as 0.882, training level

with four items as 0.783 and experience with seven items as 0.764. The overall average/mean of

2 Technical skills 4 0.775

3 General skills 4 0.882

4 Training level 4 0.783

5 Experience 4 0.764

Table 4.6 presents alpha coefficients of the two components of employee performance provided

for the study. Task performance and contextual performance. A total of 21 items were presented

to respondents in a pilot study and 16 of the items were responded to and contributed to the

overall reliability of the components of employee performance. The overall cronbach alpha

reliability coefficient of 0.8845 was obtained, indicating a very strong internal consistency among

Test of Validity

Validity is the degree to which the researcher tries to measure efficiency of the research

instrument (McDaniel & Gates, 2010). Content Validity of the research instrument was carried

out through a pilot study to ensure that the questionnaire is effective in collecting the relevant

information. This was done by administering the research instrument to State Corporations’

employees, from each of the 5 categories. Such questionnaire pre-testing helps to identify

problems with the data collection instruments and find possible solutions. Before piloting, the

instrument was discussed with the supervisors. This helps check if questions are clear, well

structured, have a logical sequence, meaning, easy to understand and identifies any potential

misunderstanding amongst respondents to avoid ambiguity and ensure clarification of technical

checking the questionnaire structure, sequence, meaning and ambiguity. The final modified

version was used to collect data.

Demographic Characteristics of the Respondents

This section identified the demographic characteristics of the study respondents including age,

gender composition of the respondents, employment position and number of years served in the

organization, number of years in the current position and nature of employment. Demographic

of employees in this study influenced job satisfaction, competence, task and contextual

performance. Vroon (1964) posits that employee performance is a function of competence in

terms of individual abilities and characteristics. Additionally, Toker (2011) in his study found that