5 non-physical environmental questions that have
been adapted based on organizational culture, the
corporate budget, work system and vision/mission
company. The second measuring tool for measuring
"work engagement" variables is derived from
benchmarks made by Schaufeli and Bakker (2003),
under the name of 17 item questions of Utrecht
Work Engagement Scale (UWES) (Schaufeli and
Bakker, 2004).
8 DATA PROCESSING
TECHNIQUE
8.1 Variable Score Analysis
The categorization of the dimension variable
dimension scale dimension of the Work
Environment dimension is Likert scale modified to
scale continuum 1-4 withthe lowest scale 1 (very
low) up to the highest scale of 4 (very high). Results
of respondent’s data displayed the category of
employees who feel the high work engagement of 6
employees (30%) and employees who feel low
engagement as many as 14 employees (70%).
Employees who feel the low work engagement is
experienced by employees who have a working
period of 2 years or more, except the position of
office boy who just worked for nine months.
Employees working in the workshop area feel the
conditions of work engagement that are no better
than employees working in the head office. This
happens because based on observation of work
environment facility and work tool support in
workshop area, there are still things that have not
fulfilled the criteria of work comfort standard
according to (Sedarmayanti, 2011; Schultz and
Schultz, 2014).
8.2 Validity and Reliability Test
The number of workplace measuring instruments at
this stage is 13 statements, while for the work
engagement includes 17 questions. Following
applied statistical tools and analyzed it, the number
of items of Variable Environment Statement used in
this study involved six items with a total value of
alpha 0.785. Thus, based on the grain analysis using
statistical tools, the number of statements of the
Work engagement variables used in this study was
15 items with a total alpha value of 0.77.
8.3 Normality Test
The normality test in this study applied
Kolmogorov-Smirnov test (Gravetter and Wallanau,
2013). The data is normally distributed when the
significance level value is > 0.05, the normality test
results only one variables have abnormally
distributed data with a Sig value of 0.001 for the
Work Engagement variable and 0.2 for Work
Environment variables when two variables are
calculated separately. Distributed data is not normal
due to Sig value. <0.05.
9 REGRESSION TEST
Testing the first hypothesis in this study is to see the
direct effect Work environment conditions against
employee employment PT. MCD. The author uses
simple regression test with SPSS 23 program to
evaluate how many variants working ties are
influenced by the variant Work environment.
Based on the regression test results, this study
found that the value of β of 0.535 and p = 0.015 <
0.05 Thus, the work environment has a significant
positive effect on the engagement of work. In this
corridor, the hypothesis of the study obtained that
there is a significant positive influence on the
variable Work Environment conditions with work
engagement at PT. MCD. Undoubtedly, the better
Work Environment at PT. MCD, the higher
employee’s work engagement level at the company.
Based on the discussion of the results of the
analysis, hypothesis in this study, Ho is rejected and
Ha is accepted. In this vein, the Work Environment
has proven to directly affect employee engagement.
10 CONCLUSION
Based on the analysis of research data at PT. MCD,
Work Environment in PT.MCD has a significant
positive effect on employee work engagement PT.
MCD. This indicates when the Work Environment
in PT. MCD is good, employee engagement will be
good. From result of the analysis, Ha accepted by the
existence of influence between work environments to
work engagement at PT. MCD. Based on the results
of research hypotheses conducted at PT. MCD, it can
be concluded that the work environment has a share
in increasing employee engagement in PT. MCD.
After the researchers conducted a regression test on
the population samples obtained, the study found that