2448 5420 1 PB
2448 5420 1 PB
2448 5420 1 PB
Abstract
This study aims to determine the extent of the role of happiness at work in improving employee performance in the head office
of BKKBN, both directly and indirectly. The approach of research uses quantitative methods. Primary data were collected by
distributing questionnaires. Analysis of data processed by descriptive analysis and inferential analysis using the structural equation
model to 200 respondents data. The results showed that happiness at work had not directly affect employee performance, but when
the relationship was mediated by work motivation there was a positive effect between happiness at work on employee performance.
Keywords: happiness at work, work motivation, employee performance
ried out which aims to test the feasibility of the research ques- 4. ANALYSIS AND DISCUSSION
tionnaire. The pre-test was carried out by distributing initial
questionnaires to 30 civil servants in the head office BKKBN Descriptive analysis in this study describes the characteris-
environment. The validity test of the pre-test questionnaire was tics of respondents who are seen based on the frequency and
carried out using the factor analysis method. Factor analysis percentage of age, gender, latest education, duration of work as
used to test the validity of research instruments is by using the a civil servant in BKKBN, position, and work unit. To map the
value of the Kaiser-Meyer-Olkin Measures of Sampling Ade- characteristics further, the researcher cross-tabulated the data
quate (KMO-MSA), Barlett’s Test of Sphericity (BTS) in the on the characteristics/identity of the respondents with latent vari-
KMO and Barlett’s Test Table, and Anti-Image Correlation in ables. Meanwhile, to interpret the attitude of the respondents to
the Anti-Image Matrices Table. The variable indicator is de- the measured variable, the mean score is used. The average
clared valid if the KMO-MSA value is¿ 0.6, the BTS value is value is then grouped into an interval scale to see the tendency
¡0.05, and anti-image correlation value is¿ 0.5. Meanwhile, to of the respondent’s attitude towards the statement.
determine the reliability of the research instrument was carried The interval scale in this study consists of five classes with
out using the Cronbach-α value. An indicator is declared valid a scale range of 0.8. Work Motivation and Employee Perfor-
if the value of Cronbach-α ≥ 0.6 (Hair, 2014). The results of mance Variables have interpretations ranging from ”Very Low”,
the validity test showed that the indicators stated in the ques- ”Low”, ”Medium”, ”High”, to ”Very High”. Meanwhile, the
tionnaire are all valid, so there is no need to reduce them, as Happiness at Work variable has interpretations of ”Unhappy”,
stated in Table 2. ”Less happy”, ”Fairly Happy”, ”Happy”, and ”Very Happy” for
the five classes from the lowest to the highest level. The aver-
age value of respondents’ perceptions on each variable can be
Table 2: Validity Test Results
seen in Table 4.
Variables KMO BTS Anti Im- Exp.
MSA age Cor-
Table 4: Mean Score Test Results
relation
Variables Mean Score Category
Happiness at 0.713 0 >0.5 Valid
Happiness at Work 3.774 Happy
Work
Work Motivation 3.949 High
Work Motivation 0.817 0 >0.5 Valid
Employee Performance 4.131 High
Employee Per- 0.731 0 >0.5 Valid
formance Source: Data processed
Source: Data processed
The results of the mean score per indicator test show mixed
results. However, most employees have a positive perception,
so the mean score on the work motivation variable and perfor-
Table 3: Reliability Test Results mance is in the high category, and in the happy category for the
Variables Cronbach- α Explanation happiness at work variable.
(≥0.60) However, when cross-tabulation was carried out between
Happiness at Work 0.902 Reliable the characteristics of age and length of work at the BKKBN
Work Motivation 0.913 Reliable against each of the latent variables as stated in Table 5, it was
Employee Perfor- 0.884 Reliable found that there were respondents who gave the lowest score.
mance Happiness at Work is the variable with the lowest score. This
certainly affects the achievement of the mean score on this vari-
Source: Data processed able, so that happiness at work has the smallest mean score
compared to other variables.
Based on the data collected through the pre-test, it can be Cross-tabulation of the age level category of work motiva-
seen if all of these variables are included in the reliable category tion shows that employees aged 31-50 years are those who have
with all Cronbach-α values above 0.60. Therefore, all indica- high motivation to work. For the cross-tabulation category for
tors as outlined in the questionnaire were declared eligible for a long time working at BKKBN, those who have long worked
use in further research. at BKKBN have high intensity, direction, and persistence.
Data analysis was performed using the structural equation Employee performance is the variable that gets the high-
modeling approach with the partial least square (SEM-PLS) est mean score, as 4.131. Because this variable gets the high-
method. Data processing using SEM is based on a research est value from the respondents, as 73 respondents (36.5%) and
model that has relationships between variables that occur simul- the least gets the lowest score. It is inversely proportional to
taneously, where one dependent variable will become the inde- the happiness at work variable which only received the highest
pendent variable in the next relationship. This analysis model score out of 42 employees (21%) and the most score was the
applies to this study, in the work motivation variable which is lowest.
the dependent variable of happiness at work and is an indepen- The measurement model is a model that represents the rela-
dent variable for the employee performance variable. tionship between latent variables and their indicators. Accord-
Agustien and Soeling / Hasanuddin Economics and Business Review 04:2 (2020) 58–64 62
Table 5: Crosstabulation of Age and Length of Work in BKKBN with Latent Variables
Happiness at work Work Motivation Employee Performance
Category I II III IV V I II III IV V I II III IV V
21-30 1 1 5 5 1 6 7 1 10 2
31-40 3 2 21 86 20 2 16 75 39 2 7 76 47
Age (year) 41-50 2 1 7 15 11 2 3 18 13 1 2 19 14
51-60 9 9 10 8 9 9
> 60 1 1 1
Total 6 4 33 115 42 2 2 25 110 61 1 2 10 114 73
Duration of 5-Jan 1 1 6 11 1 5 14 1 1 18 1
work 10-May 1 1 17 42 17 1 9 49 19 5 47 26
(year) >10 4 2 10 62 24 2 1 11 47 41 1 2 4 49 46
Total 6 4 33 115 42 2 2 25 110 61 1 2 10 114 73
Source: Data processed
ing to Hulland (1999), a good measurement model must meet was carried out by calculating the value of R² and the Goodness
three criteria, namely reliability, convergent validity, and dis- of Fit Index (GoF).
criminant validity.
Testing the convergent validity of latent variables is done by Table 7: R² Test Results
looking at the Average Variance Extracted (AVE) value. The
Construct R Square R Square Adjusted
measurement model is stated to have convergent validity if it
has an AVE value ≥ of 0.5, meaning that if AVE is equal to 0.5 Work Motivation 0.624 0.622
or more, it means that the model is good (Wong, 2013). Employee Performance 0.498 0.492
Meanwhile, to measure the internal consistency of the mea- Source: Data processed
suring instrument, a reliability test was carried out by looking at
the composite reliability value. The model has declared reliable R2 test results of 0.624 on the construct of work motiva-
if it has a composite reliability value above 0.7 (Hair, 2014) and tion indicate the influence of happiness at work variable on
Cronbach-α ≥ 0.6. work motivation by 62.4%, the remaining 37.6% is explained
Figure 2 shows the structural model of the study, which by other variables outside of it. While the employee perfor-
shows that all indicators in each construct have a factor load- mance construct has an R² value of 0.498, which means that
ing above 0.5. The results of cross-loading between indicators the work motivation variable acts as an exogenous latent vari-
of different constructs are not greater than the loading factor on able and affects employee performance by 49.8%, the remain-
the indicators of variables HW, MK, and KP. So that this model ing 50.2% is influenced by other variables outside of work mo-
is declared to meet the requirements of discriminant validity. tivation.
The next test is done by calculating the GoF value. GoF
Table 6: Convergent Validity and Reliability Test Results analysis is carried out to validate the overall model. The GoF
value is obtained from the square root of the average AVE mul-
Construct AVE Comp. Cron-
Relia- bach-α tiplied by the R2 (Tenenhaus et al., 2004).
bility
Happiness at Work 0.624 0.948 0.94 AV E = (HW + MK + KP)/3
Work Motivation 0.609 0.933 0.919 R2 = (R2 MK + R2 KP)/2
Employee Performance 0.696 0.948 0.936 q
Source: Data processed GoF = AV E.R2
= 0.600
Based on Table 6, it can be concluded that the outer model
of this study also meets the requirements of convergent validity A GoF value of 0.600 indicates that the model is fit or has a
and reliability. Because both the AVE value, composite relia- high level of conformity.
bility, and Cronbach-α meet the minimum value requirements. The hypothesis testing is done by looking at the path co-
The structural model analysis was conducted to examine the efficient value, p-value, and t-statistic. The following are the
relationship between exogenous latent variables and endoge- path coefficient values, p-value, and t-statistic in both the di-
nous latent variables, as well as the role of mediation in the rect relationship and the indirect relationship which includes
model stated in the hypothesis. Inner model testing in this study the mediating variables.
Agustien and Soeling / Hasanuddin Economics and Business Review 04:2 (2020) 58–64 63
ple of 631 employees. However, the participation rate was only URL: http://www.ccsenet.org/journal/index.php/ibr/
207 employees, with 7 (seven) of whom did not match the qual- article/download/0/0/38180/38688
Bungin, B., 2005. Metode Penelitian Kuantitatif. Prenadamedia, Jakarta.
ifications of the research sample because the working period at Byars, L., Rue, L., 2011. Human Resource Management. McGraw-Hill, New
BKKBN was less than one year. So that data processing is only York.
carried out on 200 incoming primary data. Carr, J. Z., Schmidt, A. M., Ford, J. K., DeShon, R. P., 09 2003. Climate Per-
ceptions Matter: A Meta-Analytic Path Analysis Relating Molar Climate,
Cognitive and Affective States, and Individual Level Work Outcomes. Jour-
5. CONCLUSION nal of Applied Psychology 88, 605–619.
URL: https://doi.org/10.1037/0021-9010.88.4.605
Cooper, D. R., Schindler, P. S., 2014. Business Research Methods, 12th Edition.
Based on the descriptive analysis of each latent variable McGraw-Hill, New York.
through measuring the mean score, it can be concluded that Efendi, R., Rifa’i, M. N., Bahrun, K., Milla, H., 2020. The Mediation of Work
in general, employees have a good perception of happiness at Motivation on the Effects of Work Discipline and Compensation on Per-
work, work motivation, and employee performance with all scores formance Batik MSMEs Employees in Yogyakarta City, Indonesia. Interna-
tional Journal of Multicultural and Multireligious Understanding 7 (1).
in the high / happy category. However, there are still respon- URL: https://ijmmu.com/index.php/ijmmu/article/view/1375
dents who give answers in the lowest class range. Even though Garcia, C. R., 2019. Happiness at work: Measurement Scale Validation. Journal
the number is very small, this should be a concern, especially of Business Management 59 (5), 327–340.
URL: https://doi.org/10.1590/s0034-759020190503
in the happiness at work variable, which gets the lowest score
Gibson, 1995. Organisasi dan Manajemen.
and has the smallest mean score compared to work motivation. Hair, J. F., e., 2014. Multivariate Data Analysis, Seventh Edition. En: Pearson
From the results of the inferential analysis, it can be seen Education Limited.
that the outer model in this study has met the requirements of re- Hulland, J., 1999. Use of Partial Least Squares (PLS) in Strategic Management
Research: A Review of Four Recent Studies. Strategic Management Journal
liability, convergent validity, and discriminant validity. Whereas 20 (2), 195–204.
in the inner model, it is also declared as fit or has a high level URL: https://doi.org/10.1002/(SICI)1097-0266(199902)20:2%
of conformity. 3C195::AID-SMJ13%3E3.0.CO;2-7
Hypothesis test results indicate that the variable happiness KemenPPPA, 2018. The Ministry of Women’s Empowerment and Child Pro-
tection and The Central Bureau of Statistics.
at work does not have direct effect on employee performance. Lee, C., Keyes, C., Hysom, S., Lupo, K., 2000. The positive organization:
While in the direct relationship between happiness at work on Leadership legitimacy, employee well-being, and the bottom line. The
work motivation, there is a positive effect. Likewise, in the di- Psychologist-Manager Journal 4, 143–153.
rect relationship between work motivation which acts as an ex- URL: https://doi.org/10.1037.h0095888
Lin, Y.-C., Yu, C., Yin, C.-C., 2014. The effects of positive affect, person-job
ogenous variable to the endogenous variable of employee per- fit, and well-being on job performance. Social Behavior and Personality: an
formance, there is a positive relationship with the endogenous International Journal 42 (9), 1537–1547.
variable of employee performance. URL: https://doi.org/10.2224/sbp.2014.42.9.1537
If in the direct relationship between happiness at work on Moore, M. G., 2007. The Theory of Transactional Distance.
Pryce-Jones, J., 2010. Happiness at work: Maximizing Your Psychological
employee performance there is no effect, but when mediated Capital For Success. Wiley-Blackwell.
by work motivation, happiness at work has a positive effect on Rego, A., Cunha, M. P., 2008. Authentizotic Climates and Employee Happi-
employee performance. So that the role of happiness at work in ness: Pathways to Individual Performance? Journal of Business Research
61, 739–752.
influencing employee performance cannot be done directly, but
URL: https://doi.org/10.1016/j.jbusres.2007.08.003
must involve work motivation as a mediating variable. Robbins, S. P., Judge, T. A., 2012. Organizational Behavior, 15th Edition. Pear-
It can also be interpreted that high happiness at work does son.
not necessarily improve employee performance, but high hap- Schaffer, B., 2015. A legifjabb titánok. Book Kiadó, Hungary.
Spector, P. E., 1997. Job Satisfaction: Application, Assessment, Causes, and
piness at work can increase work motivation, which will also Consequences. Sage Publication.
have an impact on improving employee performance. Tenenhaus, M., Amato, S., Vinzi, V., 2004. A Global Goodness-of-Fit Index for
PLS Structural Equation Modelling.
URL: https://www.scirp.org/(S(351jmbntvnsjt1aadkposzje))/
References reference/ReferencesPapers.aspx?ReferenceID=1960154
Ulrich, D., 1998. A New Mandate for Human Resources. Harvard Business
Achor, S., 2010. The Happiness Advantage: The Seven Principles of Posi- Review.
tive Psychology That Fuel Success and Performance at Work. Crown Busi- URL: https://hbr.org/1998/01/a-new-mandate-for-human-
ness/Random House. resources
Aguinis, H., 2009. Performance management (2nd ed.). Prentice-Hall/Pearson Wong, K., 2013. Partial least square structural equation modeling (pls-sem)
Education. techniques using smartpls. Marketing Bulletin 24 (1).
Amabile, T. M., Hill, K. G., Hennessey, B. A., Tighe, E. M., 1994. The Work URL: https://www.researchgate.net/profile/Arumugam_
Preference Inventory: Assessing Intrinsic and Extrinsic Motivational Orien- Raman/post/What_is_the_interpretation_of_SmartPLs/
tations. Journal of Personality and Social Psychology 66 (5), 950–967. attachment/59d646c2c49f478072eae9ca/AS:273837248188417@
URL: https://psycnet.apa.org/buy/1994-35880-001 1442299297149/download/Smartpls.pdf
Aprianto, C., 2016. Pengaruh Motivasi Kerja Terhadap Kepuasan Kerja serta Çetin, F., Askun Celik, D., 2018. The Effect of Occupational Self-Efficacy on
Dampaknya Terhadap Kinerja Karyawan di PT.X. Jurnal Ilmu Manajemen Work Performance Through Intrinsic Work Motivation. Management Re-
4 (2). search Review 41 (3).
URL: https://jurnalmahasiswa.unesa.ac.id/index.php/jim/ URL: https://doi.org/10.1108/MRR-03-2017-0062
article/view/17516
Bataineh, K. A., 2019. Impact of Work-Life Balance, Happiness at work, on
Employee Performance. International Business Research 12 (2).