Measuring the Sense of Belonging of Students With and Without a Migration Background in Dutch Higher Education: Design and Psychometric Quality of the UB-ARCS Scale

Abstract

Student populations in higher education have diversified internationally. Enhancing students’ sense of belonging is linked to educational success; yet, the applicability of existing measures in the diversified context is questioned. We developed and validated a sense of belonging measure in a population of students with and without a migration background. Study 1 involved creating the measure and testing its factor structure, reliability, and measurement invariance with 374 students from a large urban university. Study 2 confirmed the factor structure and examined construct validity with 151 students from a technical university. The “University Belonging – Acceptance, Recognition, Commonality and Support” (UB-ARCS) scale comprises 16 items across four dimensions. Metric invariance was found between students with and without a migration background. Strong correlations with “belongingness” and “academic efficacy” demonstrated convergent validity, while divergent validity was demonstrated with “conscientiousness.” Therefore, we found preliminary evidence of the validity and reliability of the UB-ARCS scale. Additional research is necessary for refining the suitability of the UB-ARCS for minoritized student groups and revealing the stories behind the numbers, offering deeper insights into the complex nature of belonging and marginalization in HE.

Keywords

sense of belonging university migration background measurement invariance validation

In higher education (HE), sense of belonging plays an important role in students’ decision to drop-out as it relates to students’ intention to persist and reduces student dropout likelihood in the first year (Gehringer et al., 2022; Ishitani & Kamer, 2024). Sense of belonging in HE has previously been described as feeling accepted, supported, and valued by the HE community alongside a feeling of connectedness to this community (Dias-Broens et al., 2024; Hausmann et al., 2007; Hurtado & Carter, 1997; Pittman & Richmond, 2008). Differences were found in experienced sense of belonging and educational pathways between minoritized students and students from historically privileged social identity groups (Luciano-Wong & Crowe, 2019; Stebleton et al., 2014; Vaccaro & Newman, 2016). Minoritized students experience less sense of belonging (Gopalan & Brady, 2020; Hausmann et al., 2009; Husbands et al., 2025) and have lower completion rates in certain countries such as the Netherlands (OECD, 2024). “Minoritized” refers to social identity groups that have historically been a numerical minority in HE and experience ongoing marginalization or exclusion resulting from historical and contemporary discrimination or racism. In comparison, “minority” only refers to groups that historically are a numerical minority in the student population (Chase et al., 2014; Gillborn, 2005).

Since early 2000, sense of belonging in HE has been measured with a large number of instruments (Dias-Broens et al., 2024). Responding to reoccurring calls for re-operationalization (Ribera et al., 2017), Slaten et al. (2018) developed the University Belonging Questionnaire (UBQ), validated with predominantly White undergraduates in the United States of America (USA). While the UBQ meets earlier demands for instruments applicable to first-year bachelor students (Hausmann et al., 2009; Kane et al., 2014), Slaten et al. (2018) emphasized the need for further research involving minoritized students. The 24-item UBQ, while valid and reliable, was tailored to the US context, including a “University affiliation” factor that does not fit non-US settings like the Netherlands, particularly regarding university pride (e.g., “I take pride in wearing my university’s colors”). In the Dutch context, university pride in terms of university colors or merchandise is not something Dutch-speaking students are particularly concerned with or that is meaningful to them, nor is attending university sports events part of the culture.

In the Netherlands, “migration background” is a central demographic indicator used in research, policy, and national statistics to describe population diversity and monitor social inequality. The term refers to individuals with at least one parent born outside the Netherlands (Statistics Netherlands, n.d). Whereas it differs from race- or ethnicity-based classifications used in other countries, it identifies the groups that experience minoritization in the Dutch context. The Netherlands has a long history of migration and a significant influx followed World War II, especially by migrants from Indonesia in the late 1940s and 1950s. The 1960s and 1970s brought labor migrants from Southern Europe, Turkey, and Morocco, alongside a significant wave from Suriname around its independence in 1975. In the 1980s asylum seekers and refugees came from Vietnam, Iran, and the former Yugoslavia. During that time, large groups of migrants also came to the Netherlands as a result of reunification. Since the 2000s, migration from Eastern Europe, notably Poland, has risen following EU expansion in 2004. Recent years have seen diverse migration, including highly skilled workers, international students, and refugees from Syria and Afghanistan (Statistics Netherlands, 2025).

In current Dutch society, groups such as those with Moroccan, Turkish, or Surinamese migration backgrounds (i.e., three of the five largest migrant-origin communities in the Netherlands; Statistics Netherlands, 2024b) are frequently positioned as culturally “other” and face disproportionate structural disadvantages in society, including differential treatment based on ethnicity, appearance, and religion (Andriessen et al., 2020; Statistics Netherlands, 2024a). Even groups that may appear more similar to the majority population, such as Polish migrants from Eastern Europe, experience discrimination relatively often (Van Den Broek, 2021). Discrimination and inequality occur in education, the labor market, sports, and financial or official institutions (Statistics Netherlands, 2024a). In education, children with a migration background are disadvantaged early on due to the Dutch stratified and selective early tracking educational system. Their academic potential may not be fully realized by the time students transition to secondary education, and these inequalities persist throughout their educational careers (Van De Werfhorst, 2019). Consequently, in Dutch HE, substantial differences in completion rates exist between students with and without a migration background (OECD, 2024). These patterns illustrate how “migration background,” although not ideal due to the heterogeneity of countries of birth, operates as a socially consequential category that structures inequality in Dutch HE. Therefore, in this study, we aimed to develop and validate a new sense of belonging instrument suitable for students with and without a migration background in the Dutch HE context.

Theoretical Background

Sense of belonging is a basic human need (Baumeister & Leary, 1995). In HE, it was initially viewed as part of broader constructs like institutional fit, commitment, and habitus (Bean, 1985; Cabrera et al., 1992; Cabrera & Nora, 1996; Hausmann et al., 2009; Nora & Cabrera, 1993; Thomas, 2002). Later, studies showed that students’ sense of belonging is distinct from social aspects of students’ persistence and involvement such as social integration or mattering (Hausmann et al., 2009; Tovar & Simon, 2010) and distinct from academic aspects such as institutional commitment, intrinsic motivation, and task value (Freeman et al., 2007; Hausmann et al., 2009), affirming its distinct role in student persistence (Hurtado et al., 2007). Despite this evolving understanding, large differences in used definitions and measures of students’ sense of belonging in HE exist, alongside more fundamental differences about its nature and how it is understood, and the specific groups (e.g., ethnically diverse) whose sense of belonging is being addressed (Dias-Broens et al., 2024; Gravett & Ajjawi, 2021).

Recently, common beliefs about sense of belonging in HE as uniform for all students were critiqued in light of the diversified student population and how HE institutions and research have responded to this change (Dias-Broens et al., 2024; Gravett & Ajjawi, 2021). Through widening participation initiatives, which target the increase of student groups who are underrepresented in HE, enrollment of minoritized students has grown and student populations are diversifying (Gravett & Ajjawi, 2021). While strengthening students’ sense of belonging is often addressed in these initiatives, questions arise about the uniformity of belonging understandings or expectations of the changed student population as a result. This uncertainty is supported by findings from Dias-Broens et al. (2024). Elements of students’ belonging definitions (e.g., “experiencing diversity” or “experiencing similarities”) in recent qualitative studies were not included in the most used and longstanding definitions and measures of sense of belonging in quantitative studies (e.g., Bollen & Hoyle, 1990; Goodenow, 1993; Strayhorn, 2019). These definitions were often from perspectives of minoritized students (Dias-Broens et al., 2024).

Current definitions of belonging in HE can be excluding for minoritized students as they are often the result of power dynamics in HE (Gravett & Ajjawi, 2021; Thomas, 2015). Sense of belonging definitions are often based on outdated experiences of students with historically privileged identities such as middle or upper class and white. When longstanding definitions and aligning measures are re-used and remain unchallenged, the “norm” of a typical or privileged student in HE is reproduced (Gravett & Ajjawi, 2021; Thomas, 2015). Minoritized students such as ethnically minoritized students or students with a migration background can experience incongruence between their understanding of what it means to belong and the understanding of privileged groups. For example, Stebleton and Aleixo (2016) found that African American students saw belonging in a predominantly white institution as the presence of diversity or seeing other minoritized students. This understanding of belonging, alongside other conceptualizations of belonging by ethnically minoritized students (such as “feeling safe” and “experience similarities”), was not found in used definitions and operationalizations of sense of belonging in quantitative studies. Some of the most used conceptualizations in quantitative studies were acceptance, connectedness, support, valuation, and respect (Dias-Broens et al., 2024). To test whether different student identity groups interpret a construct in a similar way, an analysis of measurement invariance (MI) can be used (Byrne & Watkins, 2003). MI has only recently been tested across students’ race for one of the multidimensional sense of belonging measures used in HE in the last 20 years (Dias-Broens et al., 2024; Maghsoodi et al., 2023). Maghsoodi et al. (2023) found strict MI for the use of the four-factor sense of social fit scale (Walton & Cohen, 2007) with Asian American/Pacific Islander and White students in the USA. MI across students’ migration background has not been tested internationally thus far.

Moreover, Gravett and Ajjawi (2021) plead for a situated and more nuanced understanding of belonging. For instance, studies show the relevance of specifying social groups (e.g., peers or teachers) and the HE environment (e.g., a major or the university campus) as they relate to different students’ experiences of belonging (Dias-Broens et al., 2024; Parsaiyan & Karimi, 2023; Ribera et al., 2017). We think the call for a more nuanced understanding of belonging (Gravett & Ajjawi, 2021) should include further research into the possible multidimensional structure of this construct. While the majority of existing quantitative studies on sense of belonging used a unidimensional approach (Dias-Broens et al., 2024), several authors, who also reported on minoritized students’ sense of belonging, used a multidimensional approach (O’Brien et al., 2020; Ribera et al., 2017). Furthermore, a large variety of combinations of underlying dimensions such as acceptance, connectedness, and support was used to define belonging (Dias-Broens et al., 2024), which suggests multidimensionality.

The Present Study

After reviewing the literature, we could not find a multidimensional measure resulting from extensive psychometric analyses, that includes items that sufficiently specify both the social groups and the HE environment, where experiences of ethnically diverse students or students with and without a migration background are considered. Using such an instrument can lead to more valid insights in the needs of all students, and eventually, these results can then be used to strengthen students’ educational pathways (Nuñez, 2009; Vaccaro & Newman, 2016). Therefore, we have developed and examined a sense of belonging in HE measure in two studies. We aimed to answer the following research questions:

(1) What constitutes a valid measure of sense of belonging in higher education across a diverse first-year student population (study 1)?

(2) What items show evidence for addressing this concept among students with and without a migration background (study 1)?

(3) To what degree does the resulting instrument provide initial evidence of a replicable structure, reliability, and convergent and divergent validity in a different student population (study 2)?

Method Study 1: Scale Development, Exploratory Testing, and MI

To answer research questions one and two, we (a) developed a sense of belonging measure with underlying dimensions and accompanying items, (b) explored the measure’s most suitable factor structure, (c) tested the measure’s reliability, and (d) tested the MI across students’ migration background.

Procedure and Participants

First-year, full-time Dutch-speaking bachelor students in law, social sciences, and business administration at a large urban research university in the Netherlands during the 2022–2023 academic year were invited to participate. Participants received a QR code or survey link via faculty emails, classroom distribution, or a research participation hour system (for social sciences students). This system requires students to engage in research for course credits, allowing them to choose projects. A follow-up email was sent a week after initial distribution. As an incentive, participants could enter their email in the survey for a chance to win a €50 e-voucher.

Out of the 429 participants who started the survey, 55 did not complete the necessary 60% of the survey that included all sense of belonging items and were excluded. This means that 374 first-year, full-time Dutch-speaking bachelor students completed all sense of belonging items. Seven of these 374 students did not fill in (any) demographic information and therefore the sample for MI testing comprised 367 students. Participants were enrolled in the School of Law (65%, n = 239), Social Sciences (31%, n = 116), Business Administration (2%, n = 9), or multiple schools (2%, n = 6). To assess MI, we collected data on students’ migration background. One-third of the participants who reported this information (33%, n = 120) had a migration background (see Table 1 for students’ migration background and country of origin), consistent with the university’s student demographics, where 33–39% in these schools had a migration background during the data collection period (Data Competence Hub, 2025).

Table 1.

Study 1: Information regarding students’ migration background

	No.	Percent
Migration background
Without migration background	247	66
With migration background	120	32.1
Morocco	18	4.8
Turkey	17	4.5
Suriname	13	3.5
Iraq	6	1.6
Afghanistan	5	1.3
China	3	0.8
Other^a	58	15.5
Total	367	98.1
Missing	7	1.9

Note. ^aIn this category, one or two students per country were included with the following countries of origin: Armenia, Aruba, Austria, Bosnia Herzegovina, Brazil, Cape Verde, Chile, Colombia, Croatia, Curacao, Czech Republic, Dominican Republic, England, Ethiopia, Georgia, Germany, Ghana, Great Britain, Greece, Hong Kong, Hungary, Indonesia, Iran, Israel, Italy, Japan, Jordan, Kazakhstan, Pakistan, Peru, Philippines, Poland, Russia, Saint Martin, Slovakia, South Korea, Spain, Sri Lanka, Sudan, Taiwan, Tunisia, Uruguay, United States of America, Venezuela, and Vietnam.

Scale Development

We started our scale development process by selecting items from existing measures identified in a literature review (Dias-Broens et al., 2024). To ensure content validity (Worthington & Whittaker, 2006), we conducted focus group interviews with 26 bachelor students with and without migration backgrounds and across different years, and we extracted underlying dimensions of the descriptions of sense of belonging given by these students. We then held brainstorming sessions with a selection of these students and with teachers to explore these dimensions further. Next, we matched items from existing measures to the focus group dimensions, identifying key aspects of sense of belonging in HE (i.e., acceptance, respect, acknowledgment, inclusion, value, support, connectedness, contribution, diversity experience, and peer similarities). Additional items were formulated for dimensions with few existing items, resulting in 84 items in total with at least three items per dimension. The majority of the pool of items within each key aspect consisted of program and university-level references in combination with references to teachers and fellow students. Also, if possible, we included multiple versions of an item with varying social groups or contexts. Details of this process are available from the first author upon request.

Before administering the sense of belonging in HE measure, a cognitive pretest was carried out with seven students and this strategy was used to pilot the items for the purpose of gathering information on student perceptions of the comprehensibility of the items (Lenzner et al., 2016). Accordingly, 13 items were deleted due to ambiguous or unintended interpretation. Consequently, the first version of the sense of belonging in HE measure that was administered to students for factor analyses consisted of 10 dimensions operationalized by 71 items.

Results of Study 1

Exploratory Factor Analysis

To explore the structure of the initial sense of belonging measure, we conducted an exploratory factor analysis (EFA) using principal factor analysis with maximum likelihood in SPSS, applying oblique rotation (direct oblimin) due to anticipated intercorrelated dimensions. The data was suitable for factor analysis, indicated by a high KMO value (0.948) and significant Barlett’s Test of Sphericity (p < .001; Field, 2013). Factors were retained based on Cattell’s scree test (Field, 2013; Floyd & Widaman, 1995; Worthington & Whittaker, 2006), which showed shifts between factors 4 and 5, and 7 and 8, while parallel analysis suggested a five-factor solution (Horn, 1965). Following Costello and Osborne’s (2005) strategy, we tested factor structures ranging from three to eight factors. We assessed item loading tables from the pattern matrix across 3–8 factors to identify the structure that (a) aligned with our theoretical expectations, (b) minimized items with loadings below 0.298 (based on our sample size; Field, 2013; Stevens, 2002), and (c) reduced cross-loadings of 0.298 or higher on multiple factors (Floyd & Widaman, 1995; Worthington & Whittaker, 2006). Both four-factor and five-factor structures were viable. Items failing to meet loading criteria were removed, resulting in the exclusion of 16 items for the four-factor structure and 19 for the five-factor structure. Further analysis led to additional removals, ensuring no problematic loadings remained. After thorough comparison and discussion, the four-factor structure was chosen for its stronger theoretical coherence, as it grouped items by underlying theoretical dimensions rather than contextual ones, unlike the five-factor structure which included a university-level factor.

Finally, due to the exploratory nature of study 1, we applied the principle of parsimony, to refine the initial item pool by removing items with high inter-item correlations within factors. Items were deleted if they correlated at 0.50 or higher with another item in the same factor (Clark & Watson, 1995), prioritizing those with the most correlations first. This process continued until no inter-item correlations of 0.50 or higher remained, resulting in the exclusion of eight items. Items from existing sense of belonging measures were excluded from this process. Consequently, the final measure comprised four dimensions with a total of 32 items. We named the measure: “University Belonging – Acceptance, Recognition, Commonality and Support (UB-ARCS)” (α = 0.93), which together explained 50.13% of the variance in the observed item responses. It includes four dimensions: (1) acceptance (11 items; α = 0.84), (2) recognition (12 items; α = 0.88), (3) commonality (5 items; α = 0.86), and (4) peer support (4 items; α = 0.80; see Table 2) with a 5-point rating scale of 1 (strongly disagree) to 5 (strongly agree) for each item. The “Commonality” dimension solely consists of items from the initial “similarities with peers” dimension identified through literature and focus group interviews, while the other dimensions consist of a combination of items from several initial dimensions.

Table 2.

Study 1: Dimensions and items of the 32-, 17- and final 16-item version of the UB-ARCS (N = 374)

#	Item	M	SD	α	λ	Factor loading
#	Item	M	SD	α	λ	1	2	3	4
Subscale: Acceptance				0.84	2.17
1^a	I am treated with as much respect by my teachers as my fellow students.	4.52	0.69			0.67
2	Fellow students from my study program are friendly to me.	4.27	0.69			0.53
3	I am accepted for who I am by fellow students from my study program.	4.16	0.77			0.51
4^c	My teachers have prejudices about me.	4.2	0.86			0.59
5^b	As a student in my study program I am treated as a person.	4.26	0.82			0.49
6^b	I can say what I want in my study program.	4.04	0.77			0.43
7	I am respected by my teachers.	4.25	0.68			0.59
8^a	Fellow students from my study program have a judgement about me right away.	3.9	0.86			0.48
9^b	I am engaged by my teachers during class.	4.16	0.66			0.41
10	I feel excluded by fellow students from my study program.	4.36	0.82			0.54
11^a	I feel equal to fellow students from my study program.	4.13	0.8			0.30
Subscale: Recognition				0.88	10.16
12	Fellow students from my study program are considerate of me.	3.68	0.77				0.43
13	Fellow students from my study program see who I am.	3.36	0.82				0.46
14^a	I feel welcomed to join activities that are organized for all students from the university.	3.49	1				0.38
15^b	In my study program my opinion matters.	3.62	0.93				0.59
16^a	My teachers notice when I am good at something (Knekta et al., 2020).	3.64	0.95				0.61
17^b	My opinion is valued by fellow students in my study program (Knekta et al., 2020).	3.71	0.82				0.72
18	My contributions in class are appreciated by fellow students.	3.75	0.81				0.67
19^b	Fellow students mention it when I make a good contribution in class.	3.18	0.97				0.64
20	In my study program I am encouraged to be myself.	3.59	0.87				0.55
21^b	I want to commit myself to things I find important at university.	3.1	1.04				0.51
22	I can apply the knowledge and skills that I bring from my background to educational activities within my study program.	3.62	0.88				0.45
23^b	Fellow students from my study program pay attention to the knowledge and skills that I bring from my background.	3.26	0.95				0.67
Subscale: Commonality				0.86	2.43
24	I have similarities with fellow students at university.	3.78	0.87					0.72
25	At university there are fellow students who look like me.	3.59	0.96					0.81
26	At university there are fellow students who have the same background as me.	3.73	0.95					0.62
27	In my study program there are fellow students who have the same norms and values as me.	3.93	0.82					0.58
28^a	I recognize myself in fellow students at university.	3.51	0.92					0.75
Subscale: Peer Support				0.80	1.29
29	If I have a study-related question I reach out to a fellow student from my study program.	3.96	1.02						0.62
30	If I’ve missed a class, I ask a fellow student in my study program for the notes (Hoffman et al., 2002).	3.64	1.17						0.58
31^a	I meet with fellow students from my study program outside of class to study for an exam (Hoffman et al., 2002).	2.81	1.43						0.69
32	I have developed personal bonds with fellow students in my study program (Hoffman et al., 2002).	3.53	1.17						0.77

Note. Translation procedures differed by item origin. Items adapted from existing English measures were translated into Dutch for the survey and back-translated into English by external translators. Items originally developed in Dutch were translated into English for this manuscript and back-translated into Dutch by external translators. Review of the back-translations indicated preservation of the core meaning of all items. Differences were limited to minor linguistic features (e.g., prepositions or word order) and did not alter item content. Where such differences were identified, wording was adjusted following discussion among the authors. Notably, for items originating from existing English measures (i.e., item 16, 17, 30, 31, and 32), only incidental and negligible differences were observed.

^aAs discussed in an earlier paragraph, these items were excluded during model refinement in Amos and therefore do not appear in the 25-item version of the UB-ARCS.

^bThese items were excluded from the 25-item version based on our final conceptual and empirical considerations and therefore do not appear in the 17-item version of the UB-ARCS.

^cThis item was excluded from the 17-item version to achieve metric invariance and therefore does not appear in the final 16-item version of the UB-ARCS.

“Acceptance” was defined as being embraced and treated in a respectful, attentive, and equitable way by peers and teachers. We defined “recognition” as feeling seen, heard, and understood within an educational setting together with the feeling that one’s contributions and background are validated. “Commonality” was defined as experiencing comparability with peers in terms of experience, background, and appearance. Finally, “peer support” was defined as the experience of being able to turn to a peer for personal or academic needs. Resulting from the four dimensions of sense of belonging that we extracted from our data and their meaning, we now define sense of belonging in HE as the extent to which a student feels respected and treated in an attentive and equitable way, feels supported by peers, and experiences comparability with others in a study program and/or on campus.

Model Refinement, Multidimensionality, and MI Testing Across Migration Background

Before assessing whether students with and without a migration background conceptually interpreted the sense of belonging items similarly through MI testing, it was necessary to achieve an acceptable model fit for the configural model (Byrne, 2010). Consequently, we engaged in a process of model refinement for the 32-item version of the UB-ARCS using Amos (Arbuckle, 2021), which involved steps similar to confirmatory factor analysis (CFA).¹ We then compared the four-factor model with a one-factor model to evaluate multidimensionality. Model fit was assessed using several indices: (1) chi-square/df ratio, (2) Root Mean Square Error of Approximation (RMSEA; Browne & Cudek, 1993), (3) Comparative Fit Index (CFI; Bentler, 1990), (4) Tucker-Lewis Index (TLI; Tucker & Lewis, 1973), and (5) Akaike Information Criterion (AIC). A chi-square/df ratio below 3.00 indicates good model fit (Kline, 1998), RMSEA values between 0.05 and 0.08 indicate acceptable fit, and values below 0.05 indicate good fit (Byrne, 2010). CFI and TLI values of 0.90 or higher are considered acceptable (Byrne, 2010). The AIC was used solely to compare the four-factor model with a one-factor model, where smaller values suggest a better fit (Hu & Bentler, 1995).

Tables 3 and 4 detail the steps taken to assess and enhance the model fit of the UB-ARCS, along with corresponding fit indices. Initially, the four-factor model with 32 items showed acceptable fit based on the chi-square/df ratio and RMSEA, but the CFI and TLI were inadequate. Item deletion was conducted in two phases. In step 1, modification indices in Amos identified items with cross-loadings on other factors, leading to the sequential removal of seven items. After each deletion, we reassessed the chi-square/df ratio, RMSEA, CFI, and TLI, ensuring that the overall fit improved and the deletion was theoretically justified before officially removing the item from the model. Step 2 focused on achieving a concise measure with acceptable fit by identifying items that were redundant, misaligned with the factor’s meaning, or vaguely formulated. Eight items were identified as either redundant, misaligned, or lacking specificity in educational context or social entity (see Tables 2 –4). The refined 17-item model demonstrated good reliability (α = 0.88) across four dimensions (α = 0.73; α = 0.82; α = 0.75; α = 0.77), with all fit indices showing acceptable results (see Table 3). Additionally, the one-factor model showed poor fit compared to the four-factor model, confirming the UB-ARCS’s multidimensional structure (see Table 3; Byrne, 2010).

Table 3.

Study 1: Model fit statistics for model refinement of the UB-ARCS (N = 374)

Model	χ²/df	RMSEA	CFI	TLI	AIC
Four-factor models
32-Item model	2.178	0.056	0.885	0.876	1202.057
17-Item model	1.661	0.042	0.964	0.958	301.044
16-Item model	1.488	0.036	0.976	0.971	252.789
One-factor models
17-Item model	5.728	0.113	0.734	0.698	787.413
16-Item model	5.912	0.115	0.746	0.709	714.731

As the four-factor model with 17 items had acceptable fit, the configural invariance was tested with a multigroup analysis to test if the imposed pattern of the four factors was equal across migration background (Byrne, 2004; Vandenberg & Lance, 2000). The unconstrained configural model, with a similar factor structure but freely estimated parameters (i.e., item loadings and intercepts) across students with at least one parent born outside the Netherlands (n = 120) and students with parents born in the Netherlands (n = 247), showed acceptable fit (see Table 4). Next, metric invariance was used to determine if the factor loadings of the 17 items were equal across students with and without at least one parent born outside the Netherlands (i.e., if the items loaded on the four latent factors equally, and if these four dimensions loaded on the higher-order sense of belonging construct equally; Vandenberg & Lance, 2000). At first, the metric model did not adequately fit the data, and there was a significant difference between the unconstrained model and the constrained measurement weights model with p < 0.05 (see Table 4). When we closely inspected the standardized estimates of the regression weights for each item, we saw a large difference for item 4 (see Table 2) between the sample of students with and without at least one parent born outside the Netherlands (i.e., β = 0.30 in the sample of students with at least one parent born outside the Netherlands, and β = 0.59 in the sample of students with parents born in the Netherlands, see Appendix A).

Table 4.

Study 1: UB-ARCS MI testing across migration background (N = 367)

Model	χ ² /df	RMSEA	CFI	TLI	p	Δχ²/df	ΔRMSEA	ΔCFI	ΔTLI
17-item four-factor model
Configural invariance	1.477	0.036	0.949	0.940	-	-	-	-	-
Metric invariance	1.498	0.037	0.944	0.937	0.029	+0.022	+0.001	−0.005	−0.003
16-item four-factor model
Configural invariance	1.422	0.034	0.959	0.951	-	-	-	-	-
Metric invariance	1.414	0.034	0.957	0.952	0.222	−0.008	0	−0.002	+0.001

After removing item 4, as presented in Table 2, from the factor acceptance, sufficient reliability remained (a = 0.73). In hindsight, after deleting item 4 (see Table 2) we believe the remaining items are even more closely in line with our description of acceptance, as item 4 (see Table 2) is more ambiguous than the other items in the scale because it remains unclear if this statement relates to directly observable behavior by others or an estimation that the participant makes or idea that they have. When we compared the one-factor 16-item model with the four-factor 16-item model, the former showed a poor fit to the data (see Table 3). Contrary to the 17-item model, for the 16-item model, metric MI could be established (see Table 4 and see Appendix A for the path models). This means that each item loaded onto one of the four latent factors in a similar way and with similar strength for students with and without a migration background (see Appendix B for the sense of belonging scores in each group). This indicates that the 16-item version of the UB-ARCS was interpreted in a conceptually similar manner by students with and without a migration background (Byrne, 2004).

Method Study 2: Confirmatory Factor Analysis and Construct Validity Evaluation

To answer research question three, we conducted a CFA on the four-factor, 16-item UB-ARCS structure (acceptance, recognition, commonality, peer support) from study 1 using a sample from a technical university. Our first hypothesis was that this four-factor model would fit the data better compared to the one-factor model (Hypothesis 1). To assess construct validity, we examined convergent and divergent validity using measures conceptually related and unrelated to the UB-ARCS. For convergent validity, we used the 6-item belongingness scale by Godwin et al. (2018), chosen for its prior use with engineering students and inclusion of university- and class-level items. Despite having only six items, it shares constructs like support and course integration with UB-ARCS. We also used the Academic Efficacy scale by Blom and Severiens (2008) as self-efficacy in HE is conceptually linked to sense of belonging. Self-efficacy, the belief in one’s ability to achieve goals (Bandura, 1997), can enhance confidence and academic capability when students feel they belong (Holloway-Friesen, 2021). Previous studies found correlations between self-/academic efficacy and sense of belonging (Graham et al., 2023; Lytle & Shin, 2020; Saroughi & Kitsantas, 2021). Thus, we anticipated positive relationships between the UB-ARCS, the belongingness scale, and academic efficacy (Hypothesis 2).

To assess divergent validity, we used the conscientiousness scale from the Big Five Inventory (BFI; John & Srivastava, 1999), which includes competence, order, dutifulness, achievement striving, self-discipline, and deliberation. In HE, conscientiousness relates to individual academic achievement, while sense of belonging involves interactions between external (social) contexts and the (affective state of an) individual (Hagerty et al., 1992). We view these as distinct aspects of the student experience. Among Big Five traits, extraversion and agreeableness consistently show positive links to sense of belonging in HE (Anistranski & Brown, 2023; Stubblebine et al., 2024). However, findings on conscientiousness vary: a positive link at the end of the first year in HE (Stubblebine et al., 2024), a weak significant correlation with general belongingness outside education (Checa & Oberst, 2022), and a significant correlation between general belongingness (not in education) and conscientiousness that did not contribute uniquely in a simultaneous regression with other Big Five personality traits (Malone et al., 2012) were found. Altogether, we anticipated no relationship between sense of belonging in HE and conscientiousness (Hypothesis 3).

Procedure and Participants

Our study involved Dutch first-year full-time bachelor students enrolled in a STEM program at a large technical university in the Netherlands. Recruitment occurred via a QR code on a PowerPoint slide shown during a lecture on May 3, 2023, with students given in-lecture time to participate. They could enter their email for a chance to win a €50 e-voucher as an incentive. Two follow-up messages were posted on the program’s electronic learning environment, and the survey closed on May 27. A total of 169 students consented and began the survey; 18 were removed for not completing the required 23% of the survey that included all UB-ARCS items but not other scales or demographic data. Of the remaining 151 students, 15 did not provide demographic information. Among those who did, 29% (n = 38) had a migration background (see Table 5 for students’ migration background and country of origin). The university in study 2 does not collect migration background data, so we cannot determine if this sample is representative of the entire student population.

Table 5.

Study 2: Information regarding students’ migration background

Migration background	No.	Percent
Without migration background	91	60.3
With migration background	38	25.2
Belgium	6	4.0
Curacao	3	2.0
Turkey	3	2.0
Indonesia	3	2.0
China	2	1.3
Morocco	2	1.3
Suriname	2	1.3
United States of America	2	1.3
Other^a	15	9.9
Total	128	84.8
Missing	22	14.6

Note. ^aIn this category, one or two students per country were included with the following countries of origin: Austria, Bosnia Herzegovina, Canada, Czech Republic, Denmark, former Yugoslavia, France, Germany, Great Britain, Hungary, Mexico, Mozambique, Panama, Portugal, Romania, South Africa, Soviet-Union, Spain, and Vietnam.

Measures

Sense of belonging in HE was measured with the 16-item UB-ARCS (see Table 2 for included items in English and Appendix C for the original Dutch items), consisting of the subscales acceptance (5 items; α = 0.68), recognition (4 items; α = 0.72), commonality (4 items; α = 0.73), and support (3 items; α = 0.75). The Cronbach’s alpha for the complete measure in study 2 was 0.85 (N = 151).

Belongingness was measured with the 6-item belongingness scale (α = 0.88 in study 2) by Godwin et al. (2018). This measure was slightly adapted, “engineering class” was replaced by “study program” or “the classes in my study program.” The first author translated the items to Dutch and then the remaining three authors critically assessed these translations and necessary adaptations were made. Furthermore, external translators did a back-translation of these items from Dutch to English to ensure the measure can be used in a similar and reliable manner in both English and Dutch.² An example item is “I feel that I am part of my study program.” Respondents indicated to what extent they agreed with each statement on a 5-point rating scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Academic efficacy was measured with an adapted version of the Dutch Self-Efficacy for Learning and Performance scale of the Motivated Strategies for Learning Questionnaire (Blom & Severiens, 2008; Kickert et al., 2019; Pintrich et al., 1993) (α = 0.91 in study 2). In consultation with the program director of the study program, the items were slightly adapted to fit the specific context of the program by adding the equivalent of “this semester” if necessary. Also, the Dutch items of this measure were back-translated to English by external translators in a similar manner as the belongingness scale². An example item is “I expect that I’ll easily pass the courses I’m taking this quarter.” Respondents indicated to what extent they agreed with each statement on a 5-point rating scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Conscientiousness was measured with the 9-item (α = 0.74 in study 2) scale from the Big Five Inventory (John & Srivastava, 1999) that was translated into Dutch and validated by Denissen et al. (2008) across different age groups. An example item from conscientiousness is “I am someone who is easily distracted.” Respondents indicated to what extent they agreed with each statement regarding their perceptions of themselves on a 5-point rating scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Results of Study 2

Confirmatory Factor Analysis

In study 2, a CFA was conducted in Amos (Arbuckle, 2021). Model fit was evaluated using chi-square/df ratio, RMSEA (Browne & Cudek, 1993), CFI (Bentler, 1990), the TLI (Tucker & Lewis, 1973), and the AIC (Hu & Bentler, 1995), as shown in Table 6. First, we compared the four-factor and one-factor 16-item models to test multidimensionality. The four-factor model demonstrated acceptable fit across all indices, while the one-factor model showed poorer fit (see Table 6) confirming the model’s multidimensional structure. All items loaded significantly on the four factors, with standardized loadings ranging from 0.49 to 0.72 for recognition, 0.46 to 0.76 for commonality, 0.40 to 0.80 for acceptance, and 0.64 to 0.77 for peer support. Hypothesis 1 was supported, as the four-factor model fitted the data better than the one-factor model. See Appendix B for the sense of belonging scores of students with and without a migration background in study 2.

Table 6.

Study 2: Model fit statistics for confirmatory factor analysis (N = 151)

Model (UB-ARCS)	χ ² /df	RMSEA	CFI	TLI	AIC
Four-factor model	1.477	0.056	0.930	0.917	251.677
One-factor model	2.790	0.109	0.726	0.687	386.976

Construct Validity

For convergent validity, the anticipated positive correlations between the UB-ARCS scale and both an alternative belongingness scale and academic efficacy were found (r = 0.71 and r = 0.36; see Table 7). Two of the four dimensions of the UB-ARCS had a significant correlation with academic efficacy (r = 0.42, r = 0.15, r = 0.40, r = 0.14; see Table 7). For divergent validity, the anticipated non-significant correlation between the UB-ARCS and the conscientiousness scale (r = 0.07) was found. These results confirm Hypotheses 2 and 3; this means that there is preliminary evidence of the convergent and divergent validity of the UB-ARCS.

Table 7.

Study 2: Bivariate correlations (r) between the 16-item UB-ARCS and (un)related variables

Variable	1	1A	1B	1C	1D	2	3	4
1 UB-ARCS Total	-
1A Recognition	0.80**	-
1B Commonality	0.79**	0.47**	-
1C Acceptance	0.72**	0.52**	0.46**	-
1D Peer support	0.70**	0.39**	0.41**	0.28**	-
2 Belongingness in engineering	0.71**	0.72**	0.44**	0.59**	0.40**	-
3 Academic efficacy	0.36**	0.42**	0.15	0.40**	0.14	0.45**	-
4 Conscientiousness	0.07	0.12	−0.03	0.09	0.04	0.12	0.40**	-

*p < .05. **p < .01.

Discussion

Because students spend a large amount of time in HE, HE institutions have an essential role in students’ development and wellbeing. Given the important role that sense of belonging plays in students’ first year in HE, and the differences in belonging experienced among various social identity groups (Gopalan & Brady, 2020; Hausmann et al., 2009; van der Zanden et al., 2018), it is necessary to use a valid, reliable, and appropriate measure to assess the sense of belonging of all students in order to appropriately support the sense of belonging of those in need. Therefore, we aimed to develop and examine a valid and reliable instrument to measure the sense of belonging in a student population of students with and without a migration background in HE.

In study 1, we combined input from qualitative sessions with students and teachers and a literature review to develop a preliminary version of the UB-ARCS. After performing an EFA and subsequent model refinement, the final 16-item version of the UB-ARCS consisted of four dimensions (i.e., acceptance, recognition, commonality, and peer support). Configural and metric MI were established across students’ migration background. In study 2, a CFA conducted in a new sample demonstrated acceptable fit of the UB-ARCS and provided support for the multidimensional structure of the UB-ARCS, thereby confirming Hypothesis 1. Positive correlations between the UB-ARCS and both an alternative belongingness scale and academic efficacy indicated initial evidence of convergent validity, confirming Hypothesis 2. Finally, with no significant correlation between the UB-ARCS and conscientiousness, initial evidence of divergent validity was found, and Hypothesis 3 was confirmed.

Remarkably, although the UB-ARCS had a positive significant correlation with academic efficacy, two of the four underlying dimensions (i.e., commonality and peer support) did not correlate significantly with academic efficacy. Both dimensions are exclusively focused on peers, either in the extent to which a student experiences compatibility with peers in terms of experience, background, and appearance or their supportive relationship with peers. Finding a solid and thorough explanation for this result goes beyond the scope of our research. Therefore, we recommend further investigation of this result. We recommend studying correlations of the UB-ARCS altogether and its underlying dimensions, with concepts that distinguish between different variations of (academic) efficacy or underlying dimensions, or concepts in the academic environment that are slightly different from academic efficacy such as academic satisfaction or college gratitude (all indicators of college-specific wellbeing behavior; Renshaw, 2018).

The results of our studies uncovered the multidimensionality of sense of belonging (e.g., from focus groups output and model fit analyses) and as such gave further supporting evidence of insights from previous research that were used in the instrument development (Dias-Broens et al., 2024). The multidimensionality remains one of the complicating factors of studying sense of belonging in HE, as was also indicated by its “messiness” and the multitude of very similar yet slightly differing definitions and underlying concepts that have been used to describe it (Dias-Broens et al., 2024). Through our studies, we hope to have reduced some of this messiness by providing initial evidence for a valid and reliable multidimensional measure that captures this essential aspect of first-year students’ experiences in HE in the Netherlands.

Moreover, what resulted from our studies is that careful attention needs to be paid to what different aspects of the HE context and the different dimensions of sense of belonging are relevant when studying this construct in a diversified student population. Our results further substantiate the need to re-examine longstanding definitions and measures of sense of belonging in light of the diversified student population in HE (Dias-Broens et al., 2024; Gravett & Ajjawi, 2021). One finding that signals this reconsideration is the dimension “commonality” of the UB-ARCS, which was not found in existing sense of belonging measures but arose from qualitative input from students. Our findings provide initial support for the validity and reliability of this dimension of sense of belonging in our sample of students with and without a migration background in the Netherlands.

Strengths and Limitations

A strength of this study is that through the combination of input from a literature review, qualitative sessions with students, and cognitive pretesting, it became clear that an added value of a new sense of belonging in HE measure would be to make item-wording concrete in terms of educational context and social groups. Consequently, it becomes easier to translate results to future actions such as interventions or policy. In most items of the UB-ARCS, the educational context and social groups are specified, such as “In my study program there are fellow students who have the same norms and values as me,” but not too specific so that the measure is only useable in the Dutch HE context. We do advise future researchers to critically examine the useability, reliability, and validity of the UB-ARCS in contexts different from Dutch HE.

A different strength of this study was the embeddedness of students’ experiences and evaluations of the gathered input. Initially, we mainly aimed to gather input from students about their definitions of sense of belonging through focus group interviews. However, as our research unfolded, we realized more steps were necessary to understand this input. Therefore, we organized more follow-up discussions with students (and staff) to do their input justice and to transform it into valid dimensions or items for distribution. We think the extensive involvement of students in the analysis of the data strengthens our study and honors the role we think students should have in the development of a measure that is intended for use in a population they are part of.

The primary limitation of our study regards the use of migration background as a classification for minoritized and privileged groups in HE. While this approach helps explore diverse experiences of belonging, it may overlook the nuanced differences within the group due to its heterogeneity. Although classifications like ethnicity, culture, or race might seem more appropriate for classification of student groups in relation to marginalization, they are less applicable in the Dutch context given its unique history and migration dynamics. The classification of migration background serves as an important initial step in better understanding students’ sense of belonging experiences in HE and assessing the suitability of current measures. However, it should be supplemented with further MI tests involving other relevant categorizations, such as generation status, religion, and sexual preference, alongside qualitative studies. These additional approaches are crucial for refining the suitability of belonging measures for minoritized students groups and revealing the stories behind the numbers, offering deeper insights into the complex nature of belonging and marginalization in HE.

The second limitation of our study concerns constraints in our validation design, including sample reuse in study 1 and the small CFA sample in study 2. Initially, we intended to conduct a CFA and MI testing across students’ migration background with the sample from study 2. However, due to the importance of the MI testing for migration background, we needed large enough groups of students with and without a migration background, and this was only the case in study 1. We have now situated the steps necessary for model fit in study 1 as “model refinement” although our strategy was similar to conducting a CFA, and we position the CFA in study 2 as the “true” CFA. We find it promising that the UB-ARCS’s 16-item, 4-factor structure was confirmed in study 2. Nevertheless, the relatively small sample size in study 2 limits the stability and generalizability of the CFA results, underscoring the need for replication with larger samples. Furthermore, although we selected conscientiousness because it is part of the well-established and validated BFI (John & Srivastava, 1999) and a known predictor of academic success (Verbree et al., 2023), this choice did not meet the ideal standard for discriminant validity testing, as conscientiousness is not conceptually similar yet empirically dissimilar to sense of belonging (Shaffer et al., 2016). Altogether, we recommend that future research using the UB-ARCS with a larger sample conduct a CFA for further discriminant and predictive validation, include reliability testing given that one subscale in study 2 showed only marginally sufficient internal consistency, and also examine the five-factor structure suggested by the parallel analysis (Horn, 1965).

The third limitation we want to discuss concerns the replicability of our research. Despite our best intentions to describe the research process in as much detail as possible, we think it might be challenging to replicate some of our findings due to many (small) steps that were taken to come to our results and the complexity of the process. Parts of the transformation of output from students into useable information for our instrument development resemble analysis processes used in qualitative studies. This process is relative by nature and based on the interaction between the researcher(s) and the research output (Bailey, 1997) and cannot be captured completely. Therefore, to ensure methodological integrity, we used different strategies such as consensus building among team members and member checking. Although we recognize the complexity and opaqueness of this process, we do believe that every step that was taken was necessary and worthwhile to come to an optimal sense of belonging measure for a student population of students with and without a migration background.

Implications for Research and Educational Practice

Our studies have several theoretical implications for the field of HE research. Importantly, we think it is necessary to critically assess the suitability of a prospective definition and measure of sense of belonging in HE in light of the university context and student population one wishes to study. Across both the development phase and the empirical results, it became evident that including a diverse student population with respect to migration background was essential, as the input and subsequent analyses differed from earlier definitions and operationalizations of sense of belonging in HE. Also, we encourage concrete references to the HE context and social groups in items when operationalizing sense of belonging. For example, in the UB-ARCS the dimension “commonality” mostly concerns university-wide experiences, and the other dimensions concern the study program. When this is not mentioned, it remains unclear if students relate the content of an item to a teacher or a peer. Consequently, when the HE context and the social groups are mentioned in items, a more accurate picture can be drawn concerning the results on sense of belonging experiences. Furthermore, our research has shown the importance of conducting a multigroup analysis in discovering if and how a measure can be used and understood in a similar way for different social identity groups. As MI has not been determined for many relevant background indicators of minoritized groups in HE research such as LGBTQIA + status or first-generation in HE status, we encourage future researchers to continue the exploration of the useability of sense of belonging measures for these and other relevant background indicators.

Concerning implications for HE practice, we encourage HE institutions to continue the involvement of a representative student body when further steps are taken by HE institutions in response to research results. When the operationalization of a multidimensional construct aligns with how students understand this construct, the results will be representative and therefore more likely to be valid and useable. Institutions should be critical towards the influence of the relations of power in further actions and policy concerning the translation of results into practice (Hermansen & Mausethagen, 2023). To further strengthen the useability of sense of belonging results, we encourage HE institutions to pay detailed attention to how and where this then translates into practice. If an institution wishes to study sense of belonging with the UB-ARCS with the goal of positively impacting students’ sense of belonging, close attention needs to be paid to the results on the different dimensions and what contexts it entails. We found that representation matters (i.e., through the “commonality” items) and that most sense of belonging items concern peers (and not faculty or other staff). These could already be two concrete focus points in further translation of our results into practice. Importantly, as we did not establish predictive validity of the UB-ARCS in this study, this should first be addressed before informing HE practice.

Conclusion

In summary, through analyzing input from the literature and focus groups, we found preliminary evidence of the validity and reliability of the UB-ARCS scale, consisting of the dimensions acceptance, recognition, commonality, and peer support, to measure students’ sense of belonging in an HE population of students with and without a migration background. Paying attention to the dimensionality of the measure, concreteness in terms of context and social groups, and suitability for a diverse student sample regarding migration background was found to be important during the developmental process. By using the UB-ARCS, studying the sense of belonging needs of all students can be improved, and results from such studies can then be used to design interventions that could strengthen all students’ educational pathways.

Footnotes

Acknowledgments

The first author has used the Generative AI tool of their university (Erudite) as an inspiration for reducing the word count to meet the word limit guideline of the Journal of Psychoeducational Assessment after the first draft of the paper was written.

ORCID iDs

Aike S. Dias-Broens

Marieke Meeuwisse

Lonneke A. L. de Meijer

Ethical Considerations

On 22 February 2021, the Ethics Review Committee DPECS gave ethical approval for the activities that were described in our ethics review application (#21-018). Also, written informed consent was given by all participants whose data has been used in this study.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Notes

Appendix

References

Andriessen

Dijkhof

J. H.

Torre

Putters

(2020). Ervaren discriminatie in Nederland II. Sociaal en Cultureel Planbureau (SCP).

Anistranski

J. A.

Brown

B. B.

(2023). A little help from their friends? How social factors relate to students’ sense of belonging at a large public university. Journal of College Student Retention: Research, Theory & Practice, 25(2), 305–325. https://doi.org/10.1177/1521025120985107

Arbuckle

J. L.

(2021). Amos. (Version 28) [SPSS].

Bailey

P. H.

(1997). Finding your way around qualitative methods in nursing research. Journal of Advanced Nursing, 25(1), 18–22. https://doi.org/10.1046/j.1365-2648.1997.1997025018.x

Bandura

(1997). Self-efficacy: The exercise of control. W. H. Freeman.

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://doi.org/10.1037/0033-2909.117.3.497

Bean

J. P.

(1985). Interaction affects based on class level in an explanatory model of college student dropout syndrome. American Educational Research Journal, 22(1), 35–64. https://doi.org/10.3102/00028312022001035

Bentler

P. M.

(1990). Comparative fit indices in structural models. Psychological Bulletin, 107(2), 238–246. https://doi.org/10.1037/0033-2909.107.2.238

Blom

Severiens

(2008). Engagement in self-regulated deep learning of successful immigrant and non-immigrant students in inner city schools. European Journal of Psychology of Education, 23(1), 41–58. https://doi.org/10.1007/BF03173139

10.

Bollen

K. A.

Hoyle

R. H.

(1990). Perceived cohesion: A conceptual and empirical examination. Social Forces, 69(2), 479–504. https://doi.org/10.2307/2579670

11.

Browne

M. W.

Cudek

(1993). Alternative ways of assessing model fit. In Personnel selection in organizations (pp. 35–70). Jossey-Bass.

12.

Byrne

B. M.

(2004). Testing for multigroup invariance using AMOS graphics: A road less traveled. Structural Equation Modeling: A Multidisciplinary Journal, 11(2), 272–300. https://doi.org/10.1207/s15328007sem1102_8

13.

Byrne

B. M.

(2010). Structural equation modeling with AMOS: Basic concepts, applications, and programming (2nd ed.). Routledge.

14.

Byrne

B. M.

Watkins

(2003). The issue of measurement invariance revisited. Journal of Cross-Cultural Psychology, 34(2), 155–175. https://doi.org/10.1177/0022022102250225

15.

Cabrera

A. F.

Castañeda

M. B.

Nora

Hengstler

(1992). The convergence between two theories of college persistence. The Journal of Higher Education, 63(2), 143–164. https://doi.org/10.1080/00221546.1992.11778347

16.

Cabrera

A. F.

Nora

(1996). The role of perceptions of prejudice and discrimination on the adjustment of minority students to college. The Journal of Higher Education, 67(2), 119–148. https://doi.org/10.1080/00221546.1996.11780253

17.

Chase

M. M.

Dowd

A. C.

Pazich

L. B.

Bensimon

E. M.

(2014). Transfer equity for “minoritized” students: A critical policy analysis of seven states. Educational Policy, 28(5), 669–717. https://doi.org/10.1177/0895904812468227

18.

Checa

Oberst

(2022). Measuring belongingness: Validation and invariance of the general belongingness scale in Spanish adults. Current Psychology, 41(12), 8490–8498. https://doi.org/10.1007/s12144-020-01279-x

19.

Clark

L. A.

Watson

(1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. https://doi.org/10.1037/1040-3590.7.3.309

20.

Costello

A. B.

Osborne

(2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7), 1–9. https://doi.org/10.7275/JYJ1-4868

21.

Data Competence Hub . (2025). Requested student data from EER database. Erasmus University.

22.

Denissen

J. J. A.

Geenen

van Aken

M. A. G.

Gosling

S. D.

Potter

(2008). Development and validation of a Dutch translation of the Big Five Inventory (BFI). Journal of Personality Assessment, 90(2), 152–157. https://doi.org/10.1080/00223890701845229

23.

Dias-Broens

A. S.

Meeuwisse

Severiens

S. E.

(2024). The definition and measurement of sense of belonging in higher education: A systematic literature review with a special focus on students’ ethnicity and generation status in higher education. Educational Research Review, 45(2), 100622. https://doi.org/10.1016/j.edurev.2024.100622

24.

Field

A. P.

(2013). Discovering statistics using IBM SPSS statistics: And sex and drugs and rock “n” roll (4th edition). Sage.

25.

Floyd

F. J.

Widaman

K. F.

(1995). Factor analysis in the development and refinement of clinical assessment instruments. Psychological Assessment, 7(7), 286–299. https://doi.org/10.1037/1040-3590.7.3.286

26.

Freeman

T. M.

Anderman

L. H.

Jensen

J. M.

(2007). Sense of belonging in college freshmen at the classroom and campus levels. The Journal of Experimental Education, 75(3), 203–220. https://doi.org/10.3200/JEXE.75.3.203-220

27.

Gehringer

T. A.

Folberg

A. M.

Ryan

C. S.

(2022). The relationships of belonging and task socialization to GPA and intentions to re-enroll as a function of race/ethnicity and first-generation college student status. Journal of Diversity in Higher Education, 15(6), 744–754. https://doi.org/10.1037/dhe0000306

28.

Gillborn

(2005). Education policy as an act of white supremacy: Whiteness, critical race theory and education reform. Journal of Education Policy, 20(4), 485–505. https://doi.org/10.1080/02680930500132346

29.

Godwin

Verdín

Kirn

Satterfield

(2018). The intersection of gender and race: Exploring chemical engineering students’ attitudes. Chemical Engineering Education, 52(2), 9. https://doi.org/10.18260/2-1-370.660-105853

30.

Goodenow

(1993). The psychological sense of school membership among adolescents: Scale development and educational correlates. Psychology in the Schools, 30(1), 79–90. https://doi.org/10.1002/1520-6807(199301)30:1<79::aid-pits2310300113>3.0.co;2-x

31.

Gopalan

Brady

S. T.

(2020). College students’ sense of belonging: A national perspective. Educational Researcher, 49(2), 134–137. https://doi.org/10.3102/0013189X19897622

32.

Graham

M. C.

Jacobson

Husman

Prince

Finelli

Andrews

M. E.

Borrego

(2023). The relations between students’ belongingness, self-efficacy, and response to active learning in science, math, and engineering classes. International Journal of Science Education, 45(15), 1241–1261. https://doi.org/10.1080/09500693.2023.2196643

33.

Gravett

Ajjawi

(2021). Belonging as situated practice. Studies in Higher Education, 47(7), 1386–1396. https://doi.org/10.1080/03075079.2021.1894118

34.

Hagerty

B. M. K.

Lynch-Sauer

Patusky

K. L.

Bouwsema

Collier

(1992). Sense of belonging: A vital mental health concept. Archives of Psychiatric Nursing, 6(3), 172–177. https://doi.org/10.1016/0883-9417(92)90028-H

35.

Hausmann

L. R. M.

Schofield

J. W.

Woods

R. L.

(2007). Sense of belonging as a predictor of intentions to persist among African American and white first-year college students. Research in Higher Education, 48(7), 803–839. https://doi.org/10.1007/s11162-007-9052-9

36.

Hausmann

L. R. M.

Schofield

J. W.

Woods

R. L.

(2009). Sense of belonging and persistence in white and African American first-year students. Research in Higher Education, 50(7), 649–669. https://doi.org/10.1007/s11162-009-9137-8

37.

Hermansen

Mausethagen

(2023). Beyond the research–practice gap: Constructing epistemic relations in teacher education. International Journal of Educational Research, 119(3), Article 102171. https://doi.org/10.1016/j.ijer.2023.102171

38.

Hoffman

Richmond

Morrow

Salomone

(2002). Investigating “sense of belonging” in first-year college students. Journal of College Student Retention: Research, Theory & Practice, 4(3), 227–256. https://doi.org/10.2190/DRYC-CXQ9-JQ8V-HT4V

39.

Holloway-Friesen

(2021). The role of mentoring on Hispanic graduate students’ sense of belonging and academic self-efficacy. Journal of Hispanic Higher Education, 20(1), 46–58. https://doi.org/10.1177/1538192718823716

40.

Horn

J. L.

(1965). A rationale and test for the number of factors in factor analysis. Psychometrika, 30(2), 179–185. https://doi.org/10.1007/BF02289447

41.

L.-T.

Bentler

P. M.

(1995). Evaluating model fit. In Structural equation modeling: Concepts, issues, and applications. In Hoyle

R. H.

(Ed.) (pp. 76–99). Sage.

42.

Hurtado

Carter

D. F.

(1997). Effects of college transition and perceptions of the campus racial climate on Latino college students’ sense of belonging. Sociology of Education, 70(4), 324–345. https://doi.org/10.2307/2673270

43.

Hurtado

Han

J. C.

Sáenz

V. B.

Espinosa

L. L.

Cabrera

N. L.

Cerna

O. S.

(2007). Predicting transition and adjustment to college: Biomedical and behavioral science aspirants’ and minority students’ first year of college. Research in Higher Education, 48(7), 841–887. https://doi.org/10.1007/s11162-007-9051-x

44.

Husbands

Linceviciute

Yetkili

(2025). The impostor phenomenon among racially minoritised university students: ‘who knows how to get rid of this? Race Ethnicity and Education, 28(5), 670–688. https://doi.org/10.1080/13613324.2024.2386949

45.

Ishitani

T. T.

Kamer

J. A.

(2024). Exploring first-generation students’ collegiate outcomes: Longitudinal comparison of dropout and transfer behaviors. Journal of College Student Retention: Research, Theory & Practice, 26(3), 900–922. https://doi.org/10.1177/15210251221127934

46.

John

O. P.

Srivastava

(1999). The Big-Five trait taxonomy: History, measurement, and theoretical perspectives. In Handbook of personality: Theory and research (pp. 102–138). Guilford.

47.

Kane

Chalcraft

Volpe

(2014). Notions of belonging: First year, first semester higher education students enrolled on business or economics degree programmes. The International Journal of Management Education, 12(2), 193–201. https://doi.org/10.1016/j.ijme.2014.04.001

48.

Kickert

Meeuwisse

Stegers-Jager

M. K.

Koppenol-Gonzalez

G. R.

Arends

Prinzie

(2019). Assessment policies and academic performance within a single course: The role of motivation and self-regulation. Assessment & Evaluation in Higher Education, 44(8), 1177–1190. https://doi.org/10.1080/02602938.2019.1580674

49.

Kline

R. B.

(1998). Principles and practice of structural equation modeling. Guilford Press.

50.

Knekta

Chatzikyriakidou

McCartney

(2020). Evaluation of a questionnaire measuring university students’ sense of belonging to and involvement in a biology department. CBE—Life Sciences Education, 19(3), ar27. https://doi.org/10.1187/cbe.19-09-0166

51.

Lenzner

Neuert

Otto

(2016). Cognitive pretesting. In GESIS survey guidelines. https://doi.org/10.15465/GESIS-SG_EN_010

52.

Luciano-Wong

Crowe

(2019). Persistence and engagement among first-year Hispanic students. Journal for Multicultural Education, 13(2), 169–183. https://doi.org/10.1108/JME-12-2017-0072

53.

Lytle

Shin

J. E.

(2020). Incremental beliefs, STEM efficacy and STEM interest among first-year undergraduate students. Journal of Science Education and Technology, 29(2), 272–281. https://doi.org/10.1007/s10956-020-09813-z

54.

Maghsoodi

A. H.

Ruedas-Gracia

Jiang

(2023). Measuring college belongingness: Structure and measurement of the Sense of Social Fit Scale. Journal of Counseling Psychology, 70(4), 424–435. https://doi.org/10.1037/cou0000668

55.

Malone

G. P.

Pillow

D. R.

Osman

(2012). The General Belongingness Scale (GBS): Assessing achieved belongingness. Personality and Individual Differences, 52(3), 311–316. https://doi.org/10.1016/j.paid.2011.10.027

56.

Nora

Cabrera

A. F.

(1993). The construct validity of institutional commitment: A confirmatory factor analysis. Research in Higher Education, 34(2), 243–262. https://doi.org/10.1007/BF00992164

57.

Nuñez

A.-M.

(2009). A critical paradox? Predictors of Latino students’ sense of belonging in college. Journal of Diversity in Higher Education, 2(1), 46–61. https://doi.org/10.1037/a0014099

58.

O’Brien

L. T.

Bart

H. L.

Garcia

D. M.

(2020). Why are there so few ethnic minorities in ecology and evolutionary biology? Challenges to inclusion and the role of sense of belonging. Social Psychology of Education, 23(2), 449–477. https://doi.org/10.1007/s11218-019-09538-x

59.

OECD . (2024). Education at a Glance 2024: OECD indicators. OECD Publishing. https://doi.org/10.1787/c00cad36-en

60.

Parsaiyan

S. F.

Karimi

(2023). English-major students’ construction of academic sense of belonging: A metaphoric exploration. International Journal of Educational Research, 122, Article 102236. https://doi.org/10.1016/j.ijer.2023.102236

61.

Pintrich

P. R.

Smith

D. A. F.

Garcia

Mckeachie

W. J.

(1993). Reliability and predictive validity of the Motivated Strategies for Learning Questionnaire (MSLQ). Educational and Psychological Measurement, 53(3), 801–813. https://doi.org/10.1177/0013164493053003024

62.

Pittman

L. D.

Richmond

(2008). University belonging, friendship quality, and psychological adjustment during the transition to college. The Journal of Experimental Education, 76(4), 343–362. https://doi.org/10.3200/JEXE.76.4.343-362

63.

Renshaw

T. L.

(2018). Psychometrics of the revised college student subjective wellbeing questionnaire. Canadian Journal of School Psychology, 33(2), 136–149. https://doi.org/10.1177/0829573516678704

64.

Ribera

A. K.

Miller

A. L.

Dumford

A. D.

(2017). Sense of peer belonging and institutional acceptance in the first year: The role of high-impact practices. Journal of College Student Development, 58(4), 545–563. https://doi.org/10.1353/csd.2017.0042

65.

Saroughi

Kitsantas

(2021). Examining relationships among contextual, motivational and wellbeing variables of immigrant language-minority college students. Innovative Higher Education, 46(1), 1–19. https://doi.org/10.1007/s10755-020-09520-y

66.

Shaffer

J. A.

DeGeest

(2016). Tackling the problem of construct proliferation: A guide to assessing the discriminant validity of conceptually related constructs. Organizational Research Methods, 19(1), 80–110. https://doi.org/10.1177/1094428115598239

67.

Statistics Netherlands . (nd). Migratieachtergrond. Retrieved October 2, 2024, from. https://www.cbs.nl/nl-nl/onze-diensten/methoden/begrippen/migratieachtergrond

68.

Slaten

C. D.

Elison

Z. M.

Deemer

E. D.

Hughes

H. A.

Shemwell

D. A.

(2018). The development and validation of the university belonging questionnaire. The Journal of Experimental Education, 86(4), 633–651. https://doi.org/10.1080/00220973.2017.1339009

69.

Statistics Netherlands . (2024a). Ervaren discriminatie in Nederland. https://www.cbs.nl/nl-nl/longread/rapportages/2024/ervaren-discriminatie-in-nederland?onepage=true

70.

Statistics Netherlands . (2024b). Integratie en samenleven 2024. https://longreads.cbs.nl/integratie-en-samenleven-2024/bevolking/

71.

Statistics Netherlands . (2025). Tijdlijn migratie, 1849–2025. https://www.cbs.nl/nl-nl/longread/diversen/2025/tijdlijn-migratie-1849-2025

72.

Stebleton

M. J.

Aleixo

M. B.

(2016). Black African immigrant college students’ perceptions of belonging at a predominately white institution. Journal of the First-Year Experience & Students in Transition, 28(1), 89–107. https://doi.org/10.65005/154230716818452724

73.

Stebleton

M. J.

Soria

K. M.

Huesman

R. L.

(2014). First-generation students’ sense of belonging, mental health, and use of counseling services at public research universities. Journal of College Counseling, 17(1), 6–20. https://doi.org/10.1002/j.2161-1882.2014.00044.x

74.

Stevens

(2002). Applied multivariate statistics for the social sciences (4th ed.). Lawrence Erlbaum Associates.

75.

Strayhorn

T. L.

(2019). College students’ sense of belonging: A key to educational success for all students (2nd ed.). Routledge.

76.

Stubblebine

A. M.

Gopalan

Brady

S. T.

(2024). Who feels like they belong? Personality and belonging in college. PLOS ONE, 19(1), Article e0295436. https://doi.org/10.1371/journal.pone.0295436

77.

Thomas

(2015). Rethinking belonging through Bourdieu, Diaspora and the spatial. Widening Participation and Lifelong Learning, 17(1), 37–49. https://doi.org/10.5456/WPLL.17.1.37

78.

Thomas

(2002). Student retention in higher education: The role of institutional habitus. Journal of Education Policy, 17(4), 423–442. https://doi.org/10.1080/02680930210140257

79.

Tovar

Simon

M. A.

(2010). Factorial structure and invariance analysis of the sense of belonging scales. Measurement and Evaluation in Counseling and Development, 43(3), 199–217. https://doi.org/10.1177/0748175610384811

80.

Tucker

L. R.

Lewis

(1973). A reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38(1), 1–10. https://doi.org/10.1007/BF02291170

81.

Vaccaro

Newman

B. M.

(2016). Development of a sense of belonging for privileged and minoritized students: An emergent model. Journal of College Student Development, 57(8), 925–942. https://doi.org/10.1353/csd.2016.0091

82.

Vandenberg

R. J.

Lance

C. E.

(2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3(1), 4–70. https://doi.org/10.1177/109442810031002

83.

Van Den Broek

(2021). Length of stay, acculturation and transnational medical travel among Polish migrants in the Netherlands. International Journal of Intercultural Relations, 84, 210–219. https://doi.org/10.1016/j.ijintrel.2021.08.002

84.

van der Zanden

P. J. A. C.

Denessen

Cillessen

A. H. N.

Meijer

P. C.

(2018). Domains and predictors of first-year student success: A systematic review. Educational Research Review, 23, 57–77. https://doi.org/10.1016/j.edurev.2018.01.001

85.

Van De Werfhorst

H. G.

(2019). Early tracking and social inequality in educational attainment: Educational reforms in 21 European countries. American Journal of Education, 126(1), 65–99. https://doi.org/10.1086/705500

86.

Verbree

A.-R.

Hornstra

Maas

Wijngaards-de Meij

(2023). Conscientiousness as a predictor of the gender gap in academic achievement. Research in Higher Education, 64(3), 451–472. https://doi.org/10.1007/s11162-022-09716-5

87.

Walton

G. M.

Cohen

G. L.

(2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92(1), 82–96. https://doi.org/10.1037/0022-3514.92.1.82

88.

Worthington

R. L.

Whittaker

T. A.

(2006). Scale development research: A content analysis and recommendations for best practices. The Counseling Psychologist, 34(6), 806–838. https://doi.org/10.1177/0011000006288127