Professional Documents
Culture Documents
Chapter - 4
Chapter - 4
4.1 RELIABILITY
Before measuring reliability it is important to know that the concept of reliability refers to
consistent outcomes when repeated measurements are taken. In order to establish the
reliability of the measuring instrument several measures of reliability can be taken. The
most important measure of reliability is to measure internal consistency. It is measured
using reliability coefficient Cronbach’s alpha (Cronbach, 1951). For the purpose of
measuring reliability of all factors Cronbach’s α reliability was calculated using SPSS.
Reliability values of all the six variables are summed up in the following table
Higher alpha shows high reliability although there is no standard agreed cut off figure but a
Cronbach’s alpha of 0.7 and above is acceptable (Nunnally & Bernstein, 1994). As we can
see from the above table all variables have Cronbach’s alpha more than 0.70 even the least
Cronbach alpha among six variables is 0.766 thus can be seen from the table that all six
variables showing higher reliabilities.
4.2 VALIDITY
Validity is the property by which it is ensured that the scale is measuring as desired.
Validity is of different types and different methods needs to be adapted to measure different
type of validity. One of such validity is face validity face validity as per Kalpan and
Sucuzzo` (1993) face validity is determines by just appearance it is very subjective validity
but cannot be said as not useful as most of the time researcher relies on subjective judgment
in this research all the instruments is found high and hence found all valid.
Construct validity is also an important type of validity (Trochim, 2006). It finds how well is
the test to measure the degree of construct for which it is meant. Construct is the attribute,
skill or ability, based on established theories. Now construct validity is based on two
components that are convergent validity and discriminatory validity. Convergent validity
check measures whether the test we using is consistent with the other established tests
which are used to measure the construct. Whereas discriminatory validity checks whether
the test we are using is not consistent with the test which are meant to measure a construct
which is exactly opposite to the construct we are required to measure.
Another kind of validity is content validity (Bollen, 1989) in which content is validated by
checking its meaning in line with the content already available. Thus content validity is a
coverage test that the test we using covers all or maximum aspects of the construct. Content
validity is again subjective and logical rather than statistical (Kalpan and Sucuzzo, 1993).
As the questionnaires were based on standardized set of questionnaires aimed on the
variables they pass the content validity test. Next type of validity is criterion validity in
criterion validity it is seen that weather the test is consistent of measuring a particular
criterion of the construct which similar tests can measure
Here for the purpose of our research we have made use of discriminate validity to measure
the effectiveness of the tests employed. The calculation has been explained in the ‘Results
& Discussion’ chapter.
4.2.1 Item Total Correlation of Organizational Justice
First of all Item to Total Correlation was measured for organizational justice and the results
are shown in table. In table out of 12 statements 7 items were used in first factor which are
12, 11, 10, 8, 7, 9 & 4. Second factor included 3, 1, 5. And third factor included only 6. The
statement number 2 was inconsistence so it was dropped.
Secondly, item to total correlation was applied for employee engagement and correlations
of all the 16 items were measured.
In this case2, 3, 5,11,15,16 items were found to be inconsistent. As the correlation value
was negative, hence lowering the overall individual’s score. So, this item was dropped and
was not considered for further analysis.
19 Before decisions about my job are made, all of my .602 Consistent Accepted
concerns are heard.
20 The organization shows very little concern for me. .240 Inconsistent Dropped
21 This organization cares about my general satisfaction .579 Inconsistent Dropped
at work.
In this case 3, 7, 8, 12, 16, 17, 20, and 21 items were found to be inconsistent. As the
correlation value was negative, hence lowering the overall individual’s score. So, these
items were dropped and were not considered for further analysis.
In this case 2, 4, 10, 11, 12 items were found inconsistent. As the correlation value was
negative, hence lowering the overall individual’s score. So, these items were dropped and
were not considered for further analysis.
4.2.1.5 Total Correlation for Employee Job Satisfaction
Thirdly, item to total correlation was applied for of employee job satisfaction all the12
items were measured.
In this case Third, Five, Six, Seven and twelve item was found inconsistent. As the
correlation value was negative, hence lowering the overall individual’s score. So, this item
was dropped and was not considered for further analysis.
Lastly, the total correlation of all the ten items was calculated for dependent variable
Organizational Commitment
In this case first and third items were found inconsistent. As the correlation value was
negative, hence lowering the overall individual’s score. So, these items were dropped and
were not considered for further analysis.
Exploratory factor analysis is used to calculate the underlying factor structure of a set of
data or a construct when one has obtained calculation on a number of variables and want to
identify the number and nature of underlying factors. It helps to observe the
interrelationships among the items of a scale that are used to reveal the clusters of items that
have enough common variation to justify their grouping together as a factor. This process
condenses a group of items in to a smaller set of composite factors with a minimum loss of
information. The varimax rotation with (KMO) Keiser-Meyer-Olkin and Bartlett’s Test was
applied to sampling adequacy for the all six questionnaires. The scree test plots were used
to determine the levels of values and a rotated component matrix is measured to identify the
loading.
All 12 items of the Abbas Ali Rastgar, Nina Pourebrahimi (2013) were subjected to factor
analysis developed three size namely; Interactional Justice, Distributive Justice and
Procedural Justice .According to Abbas Ali seven items to Interactional Justice and
Procedural Justice, three item to Distributive Justice and last only one item to Procedural
Justice.
The KMO index and Bartlett’s test were calculated value of .850 and Chi-square value of
1238.983 at p value .000. It is indicated that high value of factor analysis shown as in table
8 of KMO test. It is proof those exploratory factor analysis (EFA) good responses of
organizational justice (as shown in table 4.9).
Figure 2
TABLE: 9- SHOWING FACTOR ANALYSIS OF ORGANIZATIONAL JUSTICE
The scale was based on sixteen items developed by Aligned (2008). KMO Bartlett Test
showed value sampling adequacy .889 and chi square value 1709.442 at p- value is.000 it is
indicated in table 10 and the value of KMO is good.
TABLE 10: SHOWING KMO AND BARTLETT’S TEST FOR EMPLOYEE
ENGAGEMENT
In table three factors were emerged out of sixteen questions. The first factor
includes 4, 15,5,14,3,9,10,16 statements. Second factor included 2, 1, 8, 6. In third
7, 13, 12 were included. Only 11th statement was found to be inconsistent so it was
dropped.
TABLE 11: SHOWING FACTOR ANALYSIS OF EMPLOYEEENGAGEMENT
The 21-items Psychological scale developed by Serge Gagnon, Maxime Paquet, François
Courcy, and Christopher P. Parker (2009) was put to factor analysis. The KMO and Bartlett
test showed sampling Adequacy value .898 and Chi-Square 2579.667 and p value .000, the
value of KMO higher than 0.7 is considered as good (Table 12).
The results show that out of 21 statements 5 items were extracted in first factor which
include item number 12,14,15,13 and 16. Second factor included items 7, 6,8,5,3, and 4. In
third factor items 19, 18 and 21 were converged. In fourth factor included items 2, 1, and 10
and last five factor included 11, 20, 9.
Only 17th statement was not included as the statement was inconsistent so the statement was
dropped.
For measuring job involvement a scale developed by Daneshwar Doobree (2009) was
used with 15-items. Apply KMO and Bartlett’s tests Sampling Adequacy value .843
higher than 0.7 value to be good and Chi-square value 1348.742 and p-value is .000 it is
also considered to be good. (Table .14)
TABLE 14: SHOWING KMO AND BARTLETT’S TEST FOR JOB
INVOLVEMENT
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .843
Bartlett's Test of Sphericity Approx. Chi-Square 1348.742
Df 105
Sig. .000
All four factors were extracted. Item number 5, 4,6,15, and 7 were extracted in first factor.
Item number 2, 3, and 1 were extracted in second factor. Item number 14, 9, 8, and 13 were
extracted in third factor. Item number 11, 12 and 10 were extracted in fourth factor.
TABLE 15: SHOWING FACTOR ANALYSIS OF JOB INVOLVMENT
TABLE 16: SHOWING KMO AND BARTLETT’S TEST FOR EMPLOYEE JOB
SATISFACTION
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .903
Bartlett's Test of Sphericity Approx. Chi-Square 1739.709
Df 66
Sig. .000
In this table two factors were extracted out of twelve questions. Item numbers 4, 6, 2, 5, 3,
7, and 1 were extracted in first factor. Item number 12,11,10,8,9 were extracted in second.
TABLE 17: SHOWING FACTOR ANALYSIS OF EMPLOYEE JOB
SATISFACTION
There are 10 questions which were emerged in three factors. The first factor included 6, 5
and 7. Second factor included statement number 3, 2, and 4 and third factor included items
8,9,10.
Role(Conflict), 1.7 PC9. There are too many people telling me what to do. 0.536
Role(Workload) 00 PC11. The amount of work I am assigned keeps me
from doing a good job. 0.795
JI5. I give myself 100% to the job. 0.771
JI6. For me, the best form of relaxation is doing my
Complete
2.9 work. 0.738
involvement,
26 JI7. I enjoy my job activities more than my leisure
work itself
activities. 0.584
JI15. Not able to complete my job worries me a lot. 0.633
Strong involvement, 2.2 JI1.I feels miserable when I have less work to do. 0.697
Job satisfaction 44 0.823 JI3. I get depressed when I am not working. 0.798 10
JI8. If a job-connected problem is not solved by the
time I go home, I keep thinking about it. 0.637
JI9. I come early or stay late when I have more work to
Detachment 2.1 do. 0.664
Involvement 63 JI13. I am so much interested in my job that I have
little time for my friends. 0.621
JI14. While away on leave, I keep on worrying that my
work may be suffering. 0.686
OC5 Too much of my life would be disrupted if I leave
my organization. 0.713
Continuance,
2.0 OC6 I believe I have too few options to consider
Normative
89 leaving this organization. 0.787
Commitment
OC7 Even if it were to my advantage, I do not feel it
would be right to leave. 0.704
OC2 I really feel as if this organization’s problems are
Affective, 0.721 8
1.9 my own. 0.689
Continuance
22 OC4 It would be very hard for me to leave my job at
Commitment
this organization right now even if I wanted to. 0.640
OC8 I would feel guilty if I left this organization now. 0.774
Normative 1.6
Commitment 89 OC9 This organization deserves my loyalty. 0.727
OC10 I owe a great deal to this organization. 0.646
The descriptive statistics contained show in table reported levels of all the six variables. All
the variables were positively skewed.
Firstly Histogram for Organization Justice was created. The histogram in figure shows that
the data is again positively skewed indicating responses on higher side. The histogram in
figure 8 shows that the data is again positively skewed indicating responses on higher side.
Organization Justice
Figure 9: Showing Histogram of Employee Engagement
Figure 10: Showing Histogram Diagram of Psychological Climate
Figure 11: Showing Histogram diagram of Job Involvement
Figure 12: Showing Histogram of Employee Job Satisfaction
Figure 13: Showing Histogram diagram of Organization Commitment
The figure is also showing that the data was positively skewed.
TABLE 22: Coding of Variable’s Name
NO. Name of Factor Coding
1. Organization Justice OJ
2. Employee Engagement EE
3. Psychological Climate PC
4. Job Involvement JI
5. Employee Job Satisfaction EJS
6. Organization Commitment OC