The most prominent conceptual framework describing students’ adjustment during their transition is that of Baker and Siryk (1984, 1986): the student adaptation to college (SAC) model. This framework delineates that the college experience is multifaceted and comprises various “demands” that require a variety of “adjustments” or “coping responses” from the individual (Baker and Siryk, 1986, p. 32). According to the authors, two sets of such demands arise in the academic and social sphere of the HE institution, and the extent to which students effectively adapt to these demands of the new HE environment, is referred to as academic and social adjustment, respectively. The conceptual model continues that both academic and social adjustment each comprise several sub-facets of student experiences, as is detailed in Table 1.

Baker and Siryk’s model is operationalized in a self-report measure: the Student Adaptation to College Questionnaire (SACQ; Baker and Siryk, 1984). In this instrument, academic and social adjustment are measured using several statements (items), each referring to the abovementioned sub-facets of the first-year experience. Respondents are then asked to assess how well they are dealing with that aspect of adjustment.

Today, the SACQ still is the most commonly used instrument to map out first-year student’s academic and social adjustment (e.g., van Rooij et al., 2018b; Watson and Lenz, 2020), and the instrument has proven to be valuable in predicting first-year students’ academic achievement (e.g., review study by Credé and Niehorster, 2012). Nonetheless, questions have been raised about the conceptual foundation of the instrument, as the authors have only published limited information hereupon (see Taylor and Pastor, 2007; Watson and Lenz, 2020). We corroborate this idea that Baker and Siryk (1984, 1986) do not treat in great detail the rationale and theories on which they based their conceptual framework. It is thus unclear where the sub-facets of academic and social adjustment (see Table 1) were derived from, and to what extent they (still) provide a comprehensive representation of the first-year experience nowadays. Taylor and Pastor (2007), who critiqued the SACQ in their work, recommend that future research should first “pursue further theoretical development of the [adjustment] construct,” and second, “either revisit the structure of this instrument or create a new instrument to measure adjustment to college” (p. 1016).

Research on first-year students’ academic and social integration was first outlined in Tinto’s Student Attrition framework (1975). The original works of Tinto (1975, 1993), however, have been critiqued of not having clearly defined “integration” (e.g., Hurtado and Carter, 1997; Braxton, 2000), which has led to various interpretations of the construct, and consequently, a great variety of operational definitions (Hurtado and Carter, 1997). It is clear that, up till today, scholars do not agree on which sub-facets should be incorporated in measures of academic and social integration (Davidson and Wilson, 2013; Lee et al., 2018). To illustrate this latter point, we have summarized in Table 1 the various sub-facets that are considered in different studies, when conceptualizing and operationalizing academic and social integration.

Due to the vague theoretical grounding of the adjustment construct, and the multitude of interpretations of the integration construct, the overlap of both constructs is ambiguous. Both constructs have been described to be strongly related and are often used interchangeably (e.g., van Rooij et al., 2018a; Larose et al., 2019; Fematt et al., 2021). A few years ago, Tinto defined integration as students’ “sense of belonging” during their transition to HE, explaining it is a “state of being” based on the students’ perceptions of compatibility with their HE environment (Wolf-Wendel et al., 2009). Building further on this exposition, a sense of belonging, in transition literature, is defined as the psychological sense that one is a valued member of the HE institution community (Hausmann et al., 2007), and has been described to refer to students’ feelings of fitting in, acceptance, and support from the campus community (Strayhorn, 2012). This perspective on integration is in line with the earlier assertion of Braxton (2000) that “social and academic integration can be viewed as the psychological consequence of interactions with the institutions’ systems.” (p. 63).

Following this rationale of Braxton (2000) and Tinto [as cited by Wolf-Wendel et al. (2009)], we define adjustment as a process of adaptation of behavior and attitudes that may or may not enable a student to effectively meet the various academic and social demands students encounter in the first semester of FYHE. Integration, on the other hand, is delineated as the psychological outcome of the adjustment process at a certain point in time. This state of being is based on students’ perception of experiences within the academic and social sphere, reflecting their perceived fit with the new HE environment, and comprising components such as feeling supported, feeling competent, feeling prepared, or feeling related to the chosen study (Braxton, 2000; Wolf-Wendel et al., 2009).

The above described conceptual differentiation between adjustment and integration proved to be valuable for our previous qualitative work (Willems et al., 2021a,b), as it provided two complementary theoretical lenses on the first-year academic and social experience. This difference, however, is less meaningful when operationalizing both constructs in a Likert-type self-report questionnaire that maps out how students are dealing with their social and academic transition at a certain point in the first semester. Indeed, mapping out a student’s level (perceived quality) of academic and social adjustment (i.e., the active adaptation of behavior and attitudes as demanded by HE) at a certain point in FYHE, will in effect result in the measurement of this student’s academic and social integration (i.e., the psychological state of being that results from an individual’s perception of fit). Let us illustrate this with an example: one facet to which many first-year students will have to actively adjust is the increasing quantity of academic work in HE (Willems et al., 2021a). Measuring where students stand in the process of adapting to a large amount of work, by means of a self-report questionnaire, entails using items that de facto elicit students level of perceived competence that they can handle this facet at the time of data collection, such as: “The amount of work required for the university college is quite manageable.” In sum, a questionnaire will inevitably elicit the individual’s perceived level of adjustment, and thus – according to our theoretical framework – its perceived fit with the new HE environment or integration.

In order to identify the most important student experiences of first-year students in professional HE, in two previous studies, qualitative data were collected from 104 purposively selected FYHE students, enrolled in various study programs of a Flemish university college (see Willems et al., 2021a,b). Drawing on the critical incident technique, participants in these studies engaged in the completion of “reflective logs” at the start of the second semester of FYHE. The purpose of these logs was to encourage participants to reflect upon three critical academic experiences and three critical social experiences they encountered during their first semester in HE. Content analyses of the collected narratives unveiled nine main themes of academic integration, and five main themes of social integration (each containing several sub-themes) that university college students perceived to be critical, and which can be considered central constructs that are at play in the multifaceted academic and social transition process in professional HE contexts. Based on the perceived importance of the reported (sub)themes of experiences, we selected 16 (sub)themes (10 academic and 6 social experience themes),¹ for which we developed items in order to include them in our questionnaire. Table 2 provides a brief description of these 10 academic and 6 social experience themes, as well as an item example that we created to measure each of the themes.

For the construction of the items, we mainly used students’ phrasings as present in their reflective logs. In addition, for the formulation of the items of two of these scales, we also drew inspiration from existing items; four of the items of the ‘Feeling prepared’ scale were based on the work of Torenbeek et al. (2010), and five of the ‘Committing to the study’ scale, were based on items from the SACQ (Baker and Siryk, 1984, 1986). In order to further enhance face and construct validity (Cohen et al., 2011), this list of preliminary items was reviewed by members of the university college faculty and by four second-year university college students. After making several adjustments to both question phrasing and formatting, this resulted in the final questionnaire, counting 115 survey questions in total (Academic experience: 72 items; Social experience: 43 items; see Appendices A and B). All item are scored on a five-point Likert scale ranging from ‘Completely disagree’ to ‘Completely agree’. An overview of the steps in the development process of the FIT questionnaire is provided in Figure 1.

This first version of the FIT was administered in November (approximately two months after the start of the first semester) of academic year 2019–2020 in a Flemish university college. In total, 2,583 students completed the questionnaire. From this initial dataset, we omitted students who had indicated that they had already studied at a university or university college (N = 840), who did not gave their informed consent (N = 113), and students who did not answer correctly on any of the four control items as indicator of careless responding (N = 63), which brought about our final dataset of 1,567 respondents. These students were enrolled in 17 different professional bachelor programs (see Appendix A). To examine the validity of the questionnaire, we randomly split the sample in a calibration (N = 783) and a validation sample (N = 784).

In order to examine the criterion-related validity of the FIT scales (RQ1d), we also collected data regarding the respondents’ mental well-being and academic achievement. Firstly, two weeks after administering the initial version of the FIT, we administered the validated Warwick-Edinburgh Mental Well-being Scale (WEMWBS; Tennant et al., 2007). This 14 items scale is scored on a 5-point Likert scale ranging from ‘none of the time’ to ‘all of the time’ (e.g., “I’ve been feeling optimistic about the future”; α = 0.89). The development of the WEMWBS scale was driven by the objective of capturing a comprehensive understanding of well-being, incorporating affective-emotional aspects, cognitive-evaluative dimensions, and psychological functioning, which corroborates with our definition of mental well-being. Drawing upon previous scales, the WEMWBS was designed in a concise format to facilitate its effective utilization in population-level surveys (Tennant et al., 2007). It should be noted that only 134 respondents completed the well-being scale, 52 of which could be linked to the 1,567 students that were included in the final FIT dataset.²

Secondly, students’ academic achievement data were collected through the administration database of the university college involved. Academic achievement was conceptualized as early Grade Point Average (GPA): the weighted average score across all examinations after the first semester.

With the intention of reducing the complexity of the analyses, we examined the two sets of items pertaining to academic and social experiences separately. To address our first research question (RQ1a) regarding the factor structure of the survey, we initially examined each individual scale. Starting with the calibration sample, we conducted an exploratory factor analysis (EFA) with oblique rotation for each scale to explore their dimensionalities. This allowed us to determine whether each initially developed scale indeed measured a single component. To validate our findings, we performed confirmatory factor analysis (CFA) on each scale using the calibration sample. Subsequently, we conducted exploratory and confirmatory factor analyses on the calibration sample again, but this time we examined the two comprehensive models: (1) the set of 10 academic experience scales and (2) the set of 6 social experience scales (see Table 2).

In this phase of the analyses, modifications were made to the structure of the factor models and malfunctioning or inferior items were deleted. For the deletion of items, we simultaneously used several criteria: (1) in the factor analysis the item loads to another factor than was theoretically assumed, (2) the item has a relatively low factor loading; (3) the removal of an item enhances the Cronbach’s alpha index with minimum 0.01, (4) the item has a corrected item-total correlation of under 0.30 (Nunnally and Bernstein, 1994). For every decision to remove an item, theoretical arguments and goodness-of fit indices of the CFA’s were considered as well.

To compare the fit of different CFA models, we utilized the Akaike Information Criterion (AIC). A lower AIC value indicates a better fit compared to a model with a higher AIC value (Kline, 2016). Additional fit indices used to assess model fit included the comparative fit index (CFI), root mean square error of approximation (RMSEA), and standard root mean square residual (SRMR) (MacCallum et al., 1996; Hu and Bentler, 1999).

In the next step, we estimated the resulting model on the validation sample to ensure that the modifications were not based on chance, as they were made using a specific calibration sample (MacCallum et al., 1992). We first examined each scale separately and then estimated the full CFA models with all scales included.

Following the previous analyses, we proceeded to calculate Cronbach’s alphas for the resulting scales in both the calibration and validation samples. This step aimed to assess the internal consistency of the individual scales within the FIT, addressing our second research question (RQ1b). To determine the convergent and discriminant validity of the scales (RQ1c), we examined the correlations between the scales under investigation. These correlation analyses were conducted using the entire sample consisting of 1,567 participants. Finally, in order to examine the criterion-related validity of the different first-year integration scales (RQ1d), we calculated Pearson’s moment correlations between every academic and social integration variable on the one hand, and mental well-being and academic achievement on the other hand. Once again, this analysis was performed using the entire sample.

Several alterations were carried out at the scale level to obtain parsimonious models with good model fit, of which the different scales have good internal consistencies. Firstly, in the academic integration scales, a total of 32 items were deleted following the criteria and procedure as outlined in the analysis section above. Furthermore, EFA analyses made clear that item PSR11 (“I have not been very efficient in using my study time recently”), which was initially included under the “Problems with self-regulation” scale, is actually perceived by students to belong to “Committing to the study” When subjected to factor analysis in the calibration sample, this item exhibited a substantial factor loading of 0.77 on the “Committing to the study” factor. In contrast, its factor loading on the “Problems with self-regulation” factor was notably low, measuring only −0.02. This reclassification was also theoretically justifiable. Next, EFA’s on both the calibration and validation sample showed that the scale “Making sense of expectations” comprised two sub-scales: one sub-scale (“Knowledge & skills” in Figure 2) refers to the extent to which students have an idea of the knowledge and skills that is expected of them, the other (“Exam expectations” in Figure 2) refers to the extent to which students know what questions they can expect on examinations. Figure 2 shows the resulting CFA model for the academic section of the FIT (Calibration sample: CFI = 0.946, RMSEA = 0.041, SRMR = 0.046; Validation sample: CFI = 0.944, RMSEA = 0.041, SRMR = 0.046). All latent factors in the model were allowed to correlate, but for reasons of clarity, these intercorrelations were omitted from the figure. Manifest correlations between all academic constructs, however, can be found in Table 3. The final set of academic FIT items are presented in Appendix B.

Figure 3, then, shows the resulting CFA model for the social integration scales of the FIT (Calibration sample: CFI = 0.958, RMSEA = 0.050, SRMR = 0.038; Validation sample: CFI = 0.960, RMSEA = 0.052, SRMR = 0.033). Again, the intercorrelations between factors were omitted from the figure, and manifest correlations between constructs are displayed in Table 3. In total, 20 items were appointed to be inferior, and thus were deleted. The final set of social FIT items are presented in Appendix C.

Finally, we also fitted all resulting social (6) and academic (10) scales together in one CFA model. This full model had good fit in both the calibration (CFI = 0.940; RMSEA = 0.034; SRMR = 0.039) and validation (CFI = 0.945; RMSEA = 0.033; SRMR = 0.039) sample. In summary, subsequent to the removal of inadequate items, the FIT exhibits robust construct validity, thereby confirming its ability to effectively measure the constructs outlined in the original theoretical framework.

The alpha coefficients in Table 4 indicate that every scale in both calibration and validation samples is reliable to highly reliable, ranging from α = 0.71 to α = 0.90 (Cohen et al., 2011).

The correlations between the variables under study, as displayed in Table 3, all have the anticipated directions, which adds to the convergent and discriminant validity of the scales. For the academic scales, for instance, Problems with self-regulation is, as expected, significantly related to all other academic integration scales. In this light, Feeling prepared was weakly (Cohen, 1988) correlated with problems with self-regulation (r = −0.11). Furthermore and as theoretically expected, moderate correlations were found between Problems with self-regulation and dealing with a mounting Quantity of work (r = −0.36), General planning semester (r = −0.38), Making sense of expectations (r = −0.34), Problems following class (r = 0.42), and Taking notes (r = −0.36). This is also the case for the strong relationships between Problems with self-regulation and the remaining academic scales: Committing to study (r = −0.60), Processing learning content (r = −0.58), and Feeling competent (r = −0.51).

With regard to the social integration scales, all scales are significantly related to each other in meaningful ways. For example, Loneliness is negatively correlated to all other social integration scales; a weak relationship was found between loneliness and Feelings at start-off (r = −0.24), while moderate to strong relationships emerged between loneliness and Social self-beliefs (r = −0.45), Establishing first connection (r = −0.56), Establishing deep connection (r = −0.61), and Feeling supported by peers (r = −0.53).

To assess the criterion-related validity of the academic and social integration scales, we calculated Pearson’s moment correlations between each of these variables and mental well-being and GPA after the first semester of FYHE (Table 5).

Even though a small sample of students completed the mental well-being scale (WEBWBS), several significant relationships with the FIT scales were found. Firstly, all social integration scales have moderate to strong (Cohen, 1988) relationships with mental well-being: dealing with unknown (r = 0.39), social self-beliefs (r = 0.60), establishing first connection (r = 0.44), establishing deep connection (r = 0.36), feeling supported (r = 0.42), loneliness (r = −0.51). With regard to the academic integration scales, then, mental well-being was moderately to strongly correlated to general planning semester (r = 0.48), making sense of expectations – knowledge & skills (r = 0.50), problems with self-regulation (r = −0.46), processing learning content (r = 0.32), and feeling competent (0.58). A weak significant correlation was found between mental well-being and feeling prepared (r = 0.28).

Further, most academic integration scales were significantly related to students’ early academic achievement. Weak correlations were found between early GPA and general planning semester (r = 0.12), making sense of expectations – Exam expectations (r = 0.07), problems with self-regulation (r = −0.20), committing to study (r = 0.23), problems following class (r = −0.14), taking notes (r = 0.09), processing learning content (r = 0.24), feeling prepared (r = 0.13). Feeling competent was moderately correlated with early GPA (r = 0.33). The only significant association between early GPA and the social integration scales was the weak correlation with establishing first connection (r = 0.13).

A first important limitation is that, although multiple study programs were involved and a substantial sample of students participated, this validation study was carried out in only one university college. More work, that deploys the FIT in other HE contexts, will need to be done to determine whether the instrument is also transferable to other professional HE education settings. Additionally, it would also be valuable to explore the potential applicability of the FIT scales across academic HE contexts. The aim of this study was to develop a widely applicable questionnaire, capable of comprehensively capturing the most important aspects of first-year students’ academic and social integration in professional HE contexts. To achieve this, our prior qualitative research took a broader, more remote perspective, leading to the identification and subsequent operationalization of rather abstract and general themes within the FIT. Thus, the FIT scales do not specifically focus only on the professional aspects of HE programs, which might suggest that the instrument’s psychometric value may extend to university contexts as well.

Further, it is also important to bear in mind that we were only able to examine the relationship between academic and social experience variables and fist-year students’ mental well-being (measured two weeks after FIT scales), using data of a very small group (N = 52) of respondents. This low sample size notwithstanding, we did find several significant relationships, giving us first indications that the variables under study indeed have considerable predictive value with regard to students’ mental well-being in the first semester of FYHE. Future research, however, should further examine these relationships, using a more substantive data set.

Thirdly, it should be mentioned that the validation study was carried out cross-sectionally, at only one point in the first semester (in November, approximately two months after the start of the first semester). Taking into account the tumultuous nature of the first semester (e.g., Coertjens et al., 2017), the question could be raised whether the factor structure and relationships found in this study, would remain stable throughout the complete FYHE (and beyond). To enhance the instrument's robustness and deepen our understanding of its impact, future research designs should incorporate multiple measurement moments throughout the FYHE experience.

Despite the limitations mentioned above, we believe that the FIT in its current form, counting a feasible set of 63 items (Academic: 40 items, social: 23 items), is valuable not only for research but also for practice. To begin with, its design based on actual first-year students’ voices, the FIT provides scholars with an empirically-based and comprehensive research instrument to assess the most important integration experiences that are central to professional FYHE students’ academic and social transition process, in a valid and reliable fashion. This, for instance, allows for large-scale measurement of the concepts that were previously unveiled in our qualitative work.

The FIT might also be a valuable tool in the quality assurance system of (professional) HE institutions, when used to monitor critical aspects of the first-year students’ academic and social experience (Parpala and Lindblom-Ylänne, 2012). Indeed, certain aspects of integration could be more challenging for students in specific study programs. Examining these kind of data, thus, can provide important evidence for assessing the quality of the first-year student experience in the HE institution. In light of this, it is important to note that the FIT should not be employed as a definitive measure to assess the overall quality of education in an institution, since it is based solely on students’ perceptions of integration. Rather, the FIT serves as a valuable tool for monitoring the first-year student population and identifying potential problematic situations. Finally, the development of the FIT might answer to the call of developing instruments for student feedback generation that aim to increase first-year students’ self-knowledge with regard to essential facets of their transition to HE (Kyndt et al., 2017). Questionnaire-based feedback, when implemented in an institutional student guidance system that offers systematic opportunities for student counseling, can be an important vehicle for student reflection, and might ultimately facilitate students’ transition to HE (Ruohoniemi et al., 2017).

As a final remark, scholars and practitioners should be aware of a potential drawback, induced by the FIT’s current item count, namely the possibility of inducing respondent fatigue. With 63 items to complete, students may experience fatigue and a decrease in their engagement and attention when providing responses. This could impact the accuracy and quality of the data collected (Cohen et al., 2011). On the upside, however, the inclusion of a larger number of items allowed for a more comprehensive and qualitative measurement of the complex concept of integration, that encompasses various sub-concepts. The thoroughness and depth of the instrument in its current form might offer valuable insights that can inform both research and practical interventions aimed at supporting students’ successful integration into higher education.