Adaptation and Psychometric Evaluation of the Workplace Belongingness Scale in the Indonesian Context

Abstract

Workplace belongingness reflects an employee’s perception of being valued, supported, and fitting within the organizational environment. Although the Workplace Belongingness Scale (WBS) was originally conceptualized as a unidimensional construct, no validated Indonesian version has been available. This study aimed to adapt the WBS and examine its psychometric properties in an Indonesian working population. Following a forward–backward translation procedure, the 12-item instrument was administered to 292 employees from diverse organizational backgrounds. The dimensionality of the scale was evaluated using confirmatory factor analysis (CFA). The initial model indicated inadequate fit (CFI = 0.897; TLI = 0.874; RMSEA = 0.109; SRMR = 0.060). Model refinement was conducted by incorporating theoretically justified residual covariances based on priority 1 modification indices, resulting in improved but still marginally acceptable fit (CFI = 0.917; TLI = 0.897; RMSEA = 0.081; SRMR = 0.057). All standardized factor loadings were statistically significant. Internal consistency was high (ω = 0.905; α = 0.908). Overall, the findings provide preliminary support for the unidimensional structure of the Indonesian WBS following model refinement and for its internal consistency reliability. Further validation in independent samples is required.

Abstrak: Rasa memiliki di tempat kerja (workplace belongingness) merefleksikan persepsi pegawai mengenai sejauh mana mereka merasa dihargai, didukung, dan sesuai dengan lingkungan organisasi. Meskipun Workplace Belongingness Scale (WBS) pada awalnya dikonseptualisasikan sebagai konstruk unidimensional, hingga saat ini belum tersedia versi Bahasa Indonesia yang tervalidasi. Penelitian ini bertujuan untuk mengadaptasi WBS dan mengevaluasi properti psikometriknya pada populasi pekerja di Indonesia. Melalui prosedur forward–backward translation, instrumen yang terdiri dari 12 aitem diujikan kepada 292 pegawai dari berbagai latar belakang organisasi. Dimensionalitas skala dievaluasi menggunakan analisis faktor konfirmatori (CFA). Model awal menunjukkan kecocokan yang tidak memadai (CFI = 0.897; TLI = 0.874; RMSEA = 0.109; SRMR = 0.060). Penyempurnaan model dilakukan dengan memasukkan kovarians residual yang dijustifikasi secara teoretis berdasarkan modification indices prioritas 1, sehingga menghasilkan kecocokan model yang meningkat namun masih berada pada batas penerimaan (CFI = 0.917; TLI = 0.897; RMSEA = 0.081; SRMR = 0.057). Seluruh factor loadings terstandarisasi signifikan secara statistik. Reliabilitas internal tergolong tinggi (ω = 0.905; α = 0.908). Secara keseluruhan, temuan ini memberikan dukungan awal terhadap struktur unidimensional WBS versi Indonesia setelah proses penyempurnaan model, serta terhadap reliabilitas konsistensi internalnya. Validasi lebih lanjut pada sampel independen diperlukan.

Introduction

Belongingness has consistently been recognized as a fundamental human need and a central driver of social behavior. Baumeister and Leary (1995) conceptualized the need to belong as a pervasive motivation to form and maintain enduring, positive interpersonal relationships. When this need is thwarted (Cockshaw et al., 2014; Cockshaw & Shochet, 2010), individuals experience adverse psychological and health-related consequences (Somoray et al., 2017). This foundational perspective positions belongingness not merely as a desirable condition (Ali, 2019; Özsungur, 2020, 2024), but as a basic motivational requirement underlying well-being and adaptive functioning (Rizvi et al., 2022a).

In organizational contexts, belongingness assumes particular significance because the workplace constitutes a primary social environment in adult life (Ali, 2020; Caminong et al., 2023; Kar & Majumdar, 2025). Evidence suggests that relational experiences at work are linked to consequential organizational outcomes (AbdElhay et al., 2025; Katsaros, 2022; Saleh et al., 2024). For example, Harter et al., (2002) demonstrated that employee engagement indicators reflecting relational connection and workplace attachment are positively associated with productivity, profitability, and retention at the business-unit level. These findings underscore that employees’ psychological experiences of connection and inclusion are not peripheral (Edmondson & Lei, 2014; Frazier et al., 2017), but strategically relevant for organizational effectiveness (Koys, 2001; Maynes et al., 2024).

Recent theoretical developments further clarify that belongingness in work settings cannot be reduced to mere similarity or demographic fit. Drawing on Optimal Distinctiveness Theory, Shore et al. (2011) proposed that true inclusion involves the simultaneous fulfillment of belongingness and uniqueness needs. Individuals feel included when they perceive themselves as valued insiders (Leonardelli et al., 2010) while also being appreciated for their distinctive attributes (Slotter et al., 2014; Zhao & Glynn, 2022). This framework moves beyond traditional diversity approaches by emphasizing that belongingness must coexist with recognition of individual uniqueness (Ge, 2020; Newman et al., 2017) to foster positive individual and group outcomes (Paredes-Saavedra et al., 2024).

Despite conceptual overlap with several organizational constructs, workplace belongingness represents a distinct psychological experience. Person–organization fit primarily captures perceived compatibility between individual and organizational values (Kristof-Brown et al., 2005; Verquer et al., 2003), whereas workplace belongingness reflects a broader relational experience of being accepted and valued within the social environment of the organization. Organizational identification emphasizes cognitive self-definition in relation to the organization (He & Brown, 2013; Riketta, 2005), while belongingness encompasses affective and relational dimensions of inclusion. Similarly, perceived organizational support focuses on employees’ beliefs regarding organizational care and valuation (Eisenberger et al., 2020; Rhoades & Eisenberger, 2002; Shanock & Eisenberger, 2006), but does not fully capture the interpersonal and experiential aspects of feeling accepted as part of a social collective.

Thus, workplace belongingness can be understood as an integrative construct that reflects employees’ subjective experience of relational inclusion, extending beyond value congruence, identification, or perceived support alone (Oshana, 2025).

Importantly, belongingness is context-sensitive rather than uniformly distributed across life domains. Experimental evidence indicates that perceptions of belonging are shaped by situational cues and can fluctuate depending on social context (Walton & Cohen, 2007). Such findings imply that belongingness experienced in one domain does not automatically generalize to another. Consequently, general measures of belongingness may be insufficient to capture the specific relational dynamics embedded within organizational settings.

The contextual sensitivity of belongingness suggests that its meaning and manifestation may vary across cultural settings. In collectivistic cultures such as Indonesia, interpersonal harmony, social acceptance, and relational interdependence are highly emphasized. These cultural orientations may shape how employees interpret experiences of inclusion, fairness, and support in the workplace (Hofstede, 2001; Sondergaard, 1994a, 1994b). For instance, belongingness may be more closely tied to relational harmony and perceived social acceptance rather than to individual distinctiveness or value alignment alone (Ali, 2019; Hau, 2023; Mohamed et al., 2014) .

Consequently, the direct application of instruments developed in different cultural contexts may not fully capture the nuances of workplace belongingness in Indonesia. This underscores the need for careful cultural adaptation and psychometric evaluation to ensure that the construct is measured in a manner that is both conceptually equivalent and contextually meaningful.

Responding to this gap, Jena and Pradhan (2018) developed and validated a unidimensional Workplace Belongingness Scale to assess employees’ sense of alignment, support, fairness, and emotional attachment within organizational environments. The original validation study reported satisfactory psychometric properties with GFI, CFI and TLI were 0.93, 0.94 and 0.93 respectively, including strong internal consistency and a clear one-factor structure supported by factor analysis. These findings provide initial empirical support for workplace belongingness as a distinct and measurable construct.

However, despite the availability of this scale, no validated Indonesian adaptation currently exists. Cultural and linguistic factors may influence how employees interpret belongingness-related statements, thereby necessitating formal adaptation and psychometric evaluation.

Accordingly, the present study aims to adapt the Workplace Belongingness Scale into Bahasa Indonesia and examine its psychometric properties, including internal consistency and factorial validity. By providing a validated instrument for the Indonesian context, this study seeks to contribute to the growing literature on context-specific belongingness and to support future empirical research on employee experiences within Indonesian organizations.

Methods

Participant characteristics and research design

This study employed a cross-sectional survey design to examine the psychometric properties of the Indonesian adaptation of the Workplace Belongingness Scale. Data were collected through an online survey platform (Jakpat), which enabled nationwide participant recruitment across Indonesia. Eligibility criteria required participants to (1) be currently employed in any formal or informal work arrangement, (2) be adults (minimum 18 years old), and (3) have a minimum tenure of six months in their current workplace to ensure adequate exposure to the organizational environment. Individuals who were not currently employed or who had tenure below six months were excluded from participation.

Sampling procedures

Participants were recruited using a convenience sampling approach through the Jakpat online survey platform. The survey link was distributed to registered users who met the predefined eligibility criteria (currently employed, adult, minimum tenure of six months). Participation was voluntary and based on self-selection, as individuals independently chose to complete the survey after receiving the invitation within the application. Because the survey was administered through a panel-based platform, only users who fulfilled the screening criteria were allowed to proceed to the questionnaire. The study was conducted entirely online, and responses were collected nationwide across Indonesia. Data collection occurred in a naturalistic setting, as participants completed the survey remotely using their personal devices. A modest incentive was provided in the form of a lottery reward. Five participants were randomly selected to receive IDR 25,000 each as appreciation for their participation. The incentive was not contingent upon specific responses and was intended solely to encourage participation.

Ethical principles were clearly stated in the survey introduction. Participants were informed about the voluntary nature of the study, the purpose of the research, and their right to withdraw at any time without penalty. The study was conducted using an anonymous online survey, and no personally identifiable or sensitive information was collected. Given the minimal-risk nature of the study and the absence of identifiable data, formal ethical approval was not required in accordance with institutional guidelines for survey-based research. All procedures were conducted in accordance with general ethical principles for research involving human participants. No personally identifiable information was collected. Responses were anonymous, and confidentiality was maintained throughout data handling and analysis. Data were used exclusively for research purposes and stored securely to ensure participant privacy.

Sample size, power, and precision

The intended minimum sample size was determined a priori using power analysis conducted with G*Power 3.1. To ensure adequate statistical sensitivity for detecting medium-sized effects, the calculation was based on a two-tailed test with an assumed effect size of r = .30, significance level α = .05, and statistical power (1 − β) = .80. Under these assumptions, the required minimum sample size was N = 84. For a more conservative power level of .95, the required sample size increased to N = 134. In addition to statistical power considerations, the sample size was evaluated based on recommendations for confirmatory factor analysis rather than solely on correlational power considerations. Although an a priori power analysis using G*Power was initially conducted for general statistical sensitivity, the primary justification for sample adequacy in this study is based on CFA-specific recommendations. For factor analytic procedures, a commonly suggested guideline is a minimum participant-to-item ratio of 10:1 to ensure stable parameter estimation (Hair et al., 2019; Kline, 2016). Given that the instrument consisted of 12 items, the minimum recommended sample size based on this criterion was 120 participants. In addition to ratio-based guidelines, an absolute sample size of 200 or more is generally considered adequate for CFA models of moderate complexity (Kline, 2016). The present study employed a relatively simple one-factor model, further supporting the adequacy of the sample size.

No interim analyses or formal stopping rules were applied. Data collection proceeded during the predetermined survey period using the Jakpat platform and concluded as scheduled. All materials were administered electronically and no additional measurement instruments were used in this study. The final sample comprised 292 participants, exceeding both the recommended participant-to-item ratio and the commonly suggested minimum sample size for CFA. Therefore, the sample was considered sufficient to support stable parameter estimation and evaluation of the measurement model.

Measures and covariates

Data were collected using the Workplace Belongingness Scale developed by Jena and Pradhan (2018).The instrument consists of 12 positively worded items designed to assess employees’ perceived sense of belonging in the workplace context. The items capture perceptions of value congruence, emotional positivity toward work, organizational fairness, support, and recognition. Responses are rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The instrument was not developed by the present researchers but adapted from the original scale through a structured forward–backward translation procedure, following established guidelines for cross-cultural adaptation of psychological instruments (International Test Commission, 2017). The original English version was first translated into Bahasa Indonesia by a bilingual professional translator with expertise in psychological measurement. A second independent translator, who was not involved in the initial translation, then conducted a backward translation into English.

The original and back-translated versions were compared systematically to identify discrepancies in wording, meaning, and contextual appropriateness. Discrepancies were reviewed and resolved through an iterative reconciliation process with two organizational behavior experts, in which translation alternatives were evaluated based on semantic equivalence, conceptual alignment, and clarity of expression in the Indonesian workplace context.The aim was to achieve semantic and conceptual equivalence between the source and target versions. Particular attention was given to ensuring that item wording was appropriate for the Indonesian workplace context. Final translation decisions were made by prioritizing conceptual consistency with the original construct while ensuring linguistic naturalness and comprehensibility in Bahasa Indonesia. The original and final versions were then compared systematically to to maximize clarity and conceptual consistency across languages.

No formal cognitive interviews or pilot testing were conducted prior to the main data collection. The adaptation process relied on the forward–backward translation procedure and systematic reconciliation of discrepancies to support semantic and conceptual alignment. While this approach is commonly used as an initial step in cross-cultural adaptation, it may not fully capture potential nuances in item interpretation across cultural contexts. Therefore, the results of the present study should be interpreted as providing preliminary evidence, and further validation using additional qualitative procedures is recommended. No secondary psychological measures were included in this report. No covariates were included in the primary analyses. Although demographic characteristics (e.g., gender, marital status, education level, employment status, and job position) were collected, they were used solely for descriptive purposes and were not included as control variables in the psychometric analyses. No additional measures were collected but excluded from this report.

Data analysis

All statistical analyses were conducted using JASP version 0.95.4. Analyses were prespecified to evaluate descriptive properties, internal consistency reliability, and factorial validity of the Indonesian adaptation of the Workplace Belongingness Scale.

Results of Study

Participant characteristics

Table 1 shows that the final sample consisted of 292 working adults from various regions in Indonesia. Gender distribution was relatively balanced, with 48.85% identifying as male and 51.15% as female. In terms of marital status, 50.49% were unmarried, 47.87% were married, and 1.64% were previously married. Educational attainment varied: 38.36% had completed senior high school (SMA/SMK/MA), 12.13% held a diploma degree (D1–D4), 44.92% held a bachelor’s degree (S1), 1.97% held a postgraduate degree (S2–S3), and 2.62% reported other educational backgrounds. Regarding employment status, 44.59% were permanent employees, 21.64% were contract employees, 23.93% were freelancers or independent workers, and 9.84% reported other arrangements. Organizational positions ranged from staff/executive level (48.20%) to senior leadership positions.

Table 1. Demographic Characteristics of Participants (N = 292)

Variable	Category	n	%
Gender	Male	143	48.85
	Female	149	51.15
Marital Status	Unmarried	147	50.49
	Married	140	47.87
	Previously married	5	1.64
Education Level	Senior High School (SMA/SMK/MA)	112	38.36
	Diploma (D1–D4)	35	12.13
	Bachelor’s Degree (S1)	131	44.92
	Postgraduate (S2–S3)	6	1.97
	Other	8	2.62
Employment Status	Permanent employee	130	44.59
	Contract employee	63	21.64
	Freelancer/independent worker	70	23.93
	Other	29	9.84
Organizational Position	Staff/Executive	140	48.20
	Senior staff/Specialist	33	11.15
	Supervisor/Coordinator	26	8.85
	Manager	14	4.92
	Head of unit/section	8	2.62
	Head of division/department	4	1.31
	Director/General manager	1	0.33
	Other	66	22.62

Descriptive Statistics

Descriptive statistics for all 12 items are presented in Table 2. Item means ranged from 3.46 to 4.05, indicating moderate to relatively high levels of workplace belongingness across items. Standard deviations ranged from 0.72 to 1.02, suggesting adequate response variability. All items used the full response range (1–5). Skewness values were consistently negative (approximately between −0.78 and −0.32), indicating a tendency for responses to cluster toward the higher end of the scale. This pattern suggests that participants generally reported relatively favorable perceptions of workplace belongingness, which may reflect a positive response tendency or socially desirable responding in self-report data. The presence of negative skewness also raises the possibility of a mild ceiling effect for some items, which may limit the scale’s sensitivity in distinguishing among individuals with higher levels of belongingness. Although Shapiro–Wilk tests indicated statistically significant deviations from normality (p < .001), such results are common in moderately large samples. Given these distributional characteristics, robust estimation methods were applied in the CFA to account for potential non-normality.

Table 2. Descriptive Statistics for Workplace Belongingness Items (N = 292)

Item	Mean	SD	Skewness	Kurtosis	Min	Max
WB_01	3.73	0.92	-0.61	0.07	1	5
WB_02	4.05	0.72	-0.78	1.40	1	5
WB_03	3.70	0.80	-0.47	0.68	1	5
WB_04	3.63	0.92	-0.52	0.12	1	5
WB_05	3.82	0.85	-0.59	0.55	1	5
WB_06	3.55	0.92	-0.39	-0.06	1	5
WB_07	3.63	0.95	-0.55	0.06	1	5
WB_08	3.51	0.94	-0.46	0.11	1	5
WB_09	3.55	0.97	-0.44	-0.04	1	5
WB_10	3.46	0.92	-0.32	-0.09	1	5
WB_11	3.61	0.88	-0.43	0.07	1	5
WB_12	3.48	1.02	-0.41	-0.31	1	5

Inter-Item Correlations

Table 3 presents the inter-item correlation matrix for the 12-item scale. All inter-item correlations were positive and statistically significant (p < .001). Correlations ranged from .179 to .703, indicating moderate associations among items without evidence of excessive redundancy. The absence of extremely high correlations (e.g., > .80) suggests that the items are related but not duplicative, supporting the conceptual coherence of the unidimensional structure.

Table 3. Inter-Item Correlation Matrix for Workplace Belongingness Scale (N = 292)

WB_01

WB_02

WB_03

WB_04

WB_05

WB_06

WB_07

WB_08

WB_09

WB_10

WB_11

WB_12

WB_01

—

WB_02

.33***

—

WB_03

.46***

.38***

—

WB_04

.43***

.25***

.56***

—

WB_05

.36***

.38***

.53***

.50***

—

WB_06

.24***

.18***

.36***

.32***

.41***

—

WB_07

.40***

.25***

.66***

.53***

.62***

.36***

—

WB_08

.33***

.23***

.56***

.40***

.50***

.33***

.57***

—

WB_09

.32***

.27***

.57***

.31***

.41***

.32***

.55***

.66***

—

WB_10

.25***

.22***

.54***

.42***

.50***

.36***

.63***

.65***

.68***

—

WB_11

.35***

.26***

.54***

.44***

.50***

.29***

.58***

.58***

.56***

.70***

—

WB_12

.32***

.20***

.58***

.43***

.51***

.29***

.68***

.65***

.62***

.67***

.68***

—

Reliability Analysis

Table 4 shows that the internal consistency of the Indonesian Workplace Belongingness Scale was high. Cronbach’s alpha was .91 (95% CI [.89, .93]), indicating strong reliability. Corrected item–total correlations exceeded the commonly recommended threshold of .30, ranging from .37 to .77. Deleting any individual item did not substantially improve the reliability coefficient (α if deleted = .89–.91), supporting the retention of all 12 items. These findings indicate a high degree of internal consistency among the items.

To complement these findings, composite reliability (CR) and average variance extracted (AVE) were calculated based on the CFA results. The CR value was .91, indicating good construct reliability. The AVE value was .47, which is slightly below the commonly recommended threshold of .50, suggesting that the convergent validity of the scale is acceptable but not optimal. Taken together, these results indicate that while the scale demonstrates strong internal consistency and adequate construct reliability, further evidence is needed to fully establish convergent validity.

Table 4. Panel A. Scale Reliability Internal Consistency and Item Analysis of the Indonesian Workplace Belongingness Scale (N = 292)

Coefficient	Estimate	SE	95% CI Lower	95% CI Upper
Cronbach’s α	0.91	0.01	0.89	0.93

Table 5. Panel B. Item-Level Statistics Internal Consistency and Item Analysis of the Indonesian Workplace Belongingness Scale (N = 292)

Item	α if Deleted	95% CI Lower	95% CI Upper	Item–Rest r	95% CI Lower	95% CI Upper
WB_01	0.91	0.89	0.93	0.47	0.38	0.56
WB_02	0.91	0.89	0.93	0.37	0.26	0.46
WB_03	0.89	0.87	0.92	0.75	0.69	0.80
WB_04	0.90	0.88	0.92	0.58	0.50	0.66
WB_05	0.90	0.87	0.92	0.67	0.61	0.73
WB_06	0.91	0.89	0.93	0.44	0.34	0.52
WB_07	0.89	0.87	0.92	0.77	0.71	0.81
WB_08	0.90	0.87	0.92	0.72	0.66	0.77
WB_09	0.90	0.87	0.92	0.69	0.62	0.75
WB_10	0.89	0.87	0.92	0.74	0.68	0.79
WB_11	0.90	0.87	0.92	0.72	0.66	0.77
WB_12	0.89	0.87	0.92	0.74	0.68	0.79

Confirmatory Factor Analysis

A one-factor model was tested using robust maximum likelihood estimation (MLR) to account for non-normality in item distributions. The model yielded χ²(54) = 240.29, p < .001. Incremental fit indices indicated inadequate fit, with CFI = .897, TLI = .874, IFI = .898, and RNI = .897. Absolute fit indices also suggested this, as reflected by RMSEA = .109 (90% CI [.095, .123]) and SRMR = .060. The goodness-of-fit index (GFI = .861) and McDonald fit index (MFI = .727) were below conventional thresholds. These results indicate that the initial one-factor model did not adequately represent the observed data. Although all standardized factor loadings were statistically significant (p < .001), ranging from .354 to .813, the model did not reach acceptable levels of fit.

To address this issue, model refinement was conducted based on modification indices, with all decisions guided by theoretical considerations. A limited number of residual covariances were added between items demonstrating conceptual and wording similarity. Specifically, a residual covariance was specified between WB_10 and WB_11. This modification was theoretically justified because both items reflect closely related aspects of organizational support for employees’ development and future orientation. WB_10 captures the extent to which the organization considers employees’ career goals, whereas WB_11 reflects efforts to make work meaningful and to provide future prospects. The conceptual overlap between these items may lead to shared variance beyond the underlying construct of workplace belongingness, resulting in correlated measurement error.

Following this theoretically justified modification, the model fit improved. The revised model yielded RMSEA = 0.081 (90% CI [0.068, 0.093]), SRMR = 0.057, CFI = 0.917, TLI = 0.897, and NFI = 0.880. The CFI exceeded the conventional threshold of 0.90, and the SRMR was below 0.08, indicating acceptable but not optimal fit. The RMSEA fell within the acceptable range, although near the upper boundary. While the TLI and NFI remained slightly below 0.90, the overall pattern of indices suggests an improvement in model fit relative to the initial model. No further modifications were performed in order to maintain theoretical integrity and avoid overfitting. The revised model was therefore retained as the final measurement model.

Table 6. Comparison of Model Fit Indices Before and After Modification

Fit Index	Initial Model	Revised Model
χ²	240.29	153.50
df	54	53
p	< .001	< .001
CFI	0.897	0.917
TLI	0.874	0.897
IFI	0.898	0.918
RNI	0.897	0.917
NFI	0.872	0.880
RFI	0.844	0.851
PNFI	0.714	0.851
RMSEA	0.109	0.081
90% CI RMSEA	[0.095, 0.123]	[0.068, 0.093]
RMSEA p-value	1.433 × 10⁻¹¹	3.914 × 10⁻⁵
SRMR	0.060	0.057
GFI	0.861	0.870

Table 7. Standardized Factor Loadings for the Final CFA Model

Item	Standardized Loading (λ)	SE	z	p	95% CI
WB_01	0.473	0.066	7.187	< .001	[0.344, 0.602]
WB_02	0.361	0.071	5.087	< .001	[0.222, 0.500]
WB_03	0.767	0.033	23.164	< .001	[0.702, 0.831]
WB_04	0.597	0.054	10.989	< .001	[0.491, 0.703]
WB_05	0.684	0.043	15.960	< .001	[0.600, 0.768]
WB_06	0.446	0.069	6.449	< .001	[0.311, 0.582]
WB_07	0.812	0.027	30.304	< .001	[0.760, 0.865]
WB_08	0.767	0.034	22.289	< .001	[0.699, 0.834]
WB_09	0.743	0.033	22.298	< .001	[0.678, 0.808]
WB_10	0.784	0.031	25.173	< .001	[0.723, 0.845]
WB_11	0.749	0.039	19.299	< .001	[0.673, 0.826]
WB_12	0.809	0.025	32.889	< .001	[0.761, 0.857]

Table 7 presents the standardized factor loadings for the final model. Inspection of the loadings indicated that most items loaded strongly on the latent construct. However, two items—WB_02 (λ = .361) and WB_06 (λ = .446)—exhibited comparatively lower loadings.

Item WB_02, which refers to the use of collective language (“we” rather than “they”) when describing the organization, may be more closely associated with organizational identification than with the experiential sense of belongingness. In the Indonesian context, the use of collective pronouns may reflect social norms of politeness or communication style rather than a deep psychological sense of inclusion, potentially reducing its alignment with the core construct.

Similarly, item WB_06, which emphasizes similarity with coworkers, may capture perceived interpersonal similarity rather than relational acceptance. In collectivistic cultural contexts, belongingness may be more strongly associated with relational harmony, support, and inclusion, rather than with perceived similarity. As a result, this item may show weaker associations with the underlying construct of workplace belongingness.

These findings suggest that workplace belongingness in the Indonesian context may be more strongly defined by relational and contextual factors, such as support, fairness, and recognition, than by similarity or symbolic identification alone.

Discussion

The present study aimed to adapt the Workplace Belongingness Scale into Bahasa Indonesia and examine its psychometric properties in an Indonesian working population. Overall, the findings provide preliminary support for the unidimensional structure of the scale, particularly after theoretically guided model refinement.The Indonesian version demonstrated strong internal consistency, indicating that the items function cohesively in capturing a common construct. However, internal consistency alone does not establish the overall quality of a measurement model. The factor analytic results suggest that, although the scale largely reflects a single underlying dimension, the structure is not entirely homogeneous.

The initial confirmatory factor analysis indicated that a simple one-factor model did not adequately represent the data. Model fit improved only after the introduction of a theoretically justified residual covariance, suggesting that certain items share variance beyond the latent construct. This finding implies that workplace belongingness, while conceptually unidimensional, may involve closely related sub-facets that are not fully independent at the measurement level. More importantly, the presence of comparatively weaker items provides insight into how workplace belongingness may be experienced in the Indonesian context. Item WB_02, which captures the use of collective language, appears to align more closely with organizational identification than with the experiential sense of belongingness. In a cultural context where collectivistic norms and politeness in communication are prevalent, the use of inclusive pronouns may reflect social convention rather than a genuine psychological sense of inclusion. Similarly, item WB_06, which emphasizes similarity with coworkers, showed weaker associations with the latent construct. This finding suggests that perceived similarity may not be a central component of belongingness in this context. Instead, belongingness may be more strongly grounded in relational and contextual factors, such as perceived support, fairness, and organizational recognition.

This interpretation is further supported by the need to correlate residuals between items reflecting organizational support for employee development (WB_10 and WB_11). The overlap between these items indicates that future-oriented organizational support may constitute a particularly salient aspect of belongingness, contributing to shared variance beyond the general factor. Taken together, these findings suggest that workplace belongingness in the Indonesian context may be best understood as a relationally anchored construct. Feelings of acceptance appear to be shaped less by similarity or symbolic identification and more by experiences of support, fairness, and meaningful inclusion within the organization. While these elements remain integrated within a single latent dimension, they may operate as closely related experiential components rather than fully interchangeable indicators. From a practical standpoint, the adapted scale provides a useful starting point for assessing employees’ perceived sense of belonging in Indonesian organizational settings. However, given that the final model was obtained through post hoc modification and that some indicators showed weaker performance, the current findings should be interpreted as preliminary rather than definitive evidence of structural validity.

Several limitations should be acknowledged. First, the use of convenience sampling through an online panel may limit the generalizability of the findings, particularly across different occupational or organizational contexts. Second, the cross-sectional design does not allow for the assessment of temporal stability. Third, the reliance on model modification to achieve acceptable fit suggests that the measurement structure may be sample-specific. Future research should therefore prioritize cross-validation in independent samples, as well as the examination of measurement invariance across demographic and occupational groups. Additional qualitative procedures, such as cognitive interviewing, may also help clarify how specific items are interpreted in the Indonesian context. Finally, further refinement of items that capture similarity and symbolic identification may be necessary to better align the scale with the relational nature of workplace belongingness observed in this study.

Conclusion and Recommendation

This study aimed to adapt the Workplace Belongingness Scale into Bahasa Indonesia and examine its psychometric properties in an Indonesian working population. The findings provide preliminary support for the use of the Indonesian version of the scale, particularly in terms of internal consistency reliability and the one-dimensional factor structure following model refinement. Although the results suggest that workplace belongingness can be represented as a coherent latent construct, the adequacy of the measurement model was achieved only after the inclusion of a theoretically justified residual covariance, and some indicators showed comparatively weaker performance. Accordingly, the present findings should be interpreted as initial evidence rather than definitive validation of the scale’s structural properties.

The adapted instrument may serve as a useful starting point for research on workplace belongingness in the Indonesian context. However, further validation is required before the scale can be recommended for broader application. In particular, future studies should prioritize cross-validation in independent samples, testing of measurement invariance across demographic and occupational groups, and the examination of temporal stability and predictive validity. Additional qualitative procedures, such as cognitive interviewing, may also be valuable in refining items that showed weaker associations with the construct. Through such efforts, a more robust and contextually grounded measurement of workplace belongingness can be established.

Acknowledgments

This study was conducted as part of a broader research initiative on workplace belongingness. The authors express their gratitude to the professionals involved in the forward and backward translation process, as well as to the organizational behavior experts who were consulted during the discrepancy resolution stage of the translation process. The authors also thank all respondents for their participation. This study did not receive external funding.

Declarations

Ethics approval and consent to participate

Given the minimal-risk nature of the study and the absence of identifiable data, formal ethical approval was not required in accordance with institutional guidelines for survey-based research.

Consent for publication

Responses were anonymous, and confidentiality was maintained throughout data handling and analysis

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request

Conflicts of interest Statement

The authors declare that they have no competing interests

Funding

This research is independently funded by the author

Artificial Intelligence-Assisted Technology

AI tools are used for supportive purposes such as improving grammar, clarity, reference formatting and content organization. Author is fully responsible for (a) scientific accuracy, (b) originality, (c) absence of plagiarism, fabrication, or falsification, and (d) verifying references.

Authors' contributions.

This research is solely made by the author.

ABOUT THE AUTHORS

Gita Widya Laksmini Soerjoatmodjo completed her Bachelor’s degree in Psychology at the University of Indonesia (1994–1999). She began her career as a journalist for Tempo Magazine and later became a London-based correspondent while pursuing a Master’s degree in Understanding and Securing Human Rights at the University of London, School of Advanced Studies, Institute of Commonwealth Studies, supported by a Chevening Award scholarship. Since 2003, Gita has managed local, national, and ASEAN-level civil society partnerships on transparency, good governance, and freedom of information while also serving as a lecturer at the Diploma of Arts program at Monash College Jakarta. After earning her second Master’s degree in Educational Psychology from the University of Indonesia (2008–2010) and qualifying as an educational psychologist, she joined Universitas Pembangunan Jaya in 2011. There, she has served as a lecturer, Head of the Liberal Arts, Sustainable Eco-Development, and Entrepreneurship Unit, Head of the Department of Psychology, and Head of Internal Quality Assurance, a role she continues to hold. Gita is currently pursuing a Doctoral degree at the Faculty of Psychology, University of Indonesia.