Belonging can rise in first year, but survey comparisons across time can mislead

Updated Sep 07, 2026

A repeated survey needs to measure a comparable construct before its scores can be interpreted as a trend. Aike S. Dias-Broens and colleagues examine this issue for first-year student belonging.

What the study reports

Students in psychology and education programmes at one Dutch university were surveyed in September 2023, January 2024 and June 2024, with 372, 273 and 254 responses respectively. The study examined the University Belonging – Acceptance, Recognition, Commonality and Support scale.

The detailed discussion reports partial longitudinal measurement invariance during the first semester. Comparability across the full year was not established, and group comparability by migration background also failed at the final timepoint. Simple full-year mean comparisons therefore need caution.

First-semester modelling indicated increasing belonging with similar changes across the groups examined. Students with a migration background reported lower belonging; generation status showed no significant difference in these analyses. These results concern this instrument and setting, not every belonging scale.

Implications for reporting

As editorial recommendations, document the scale, timing and respondent groups behind a reported trend. Obtain appropriate measurement expertise when using comparisons to evaluate an intervention or allocate resources.

Open comments can help explore what students mean by acceptance or support, but text analysis does not replace measurement-invariance testing. Keep explanations for score changes separate from what the analysis has established.

Reference and scope

Aike S. Dias-Broens, Marieke Meeuwisse, Marleen H. M. de Moor and Sabine E. Severiens. “First-year students’ sense of belonging in higher education: examining measurement invariance and longitudinal development across migration background and generation status.” Studies in Higher Education. DOI: 10.1080/03075079.2026.2631047.

Reviewed 7 September 2026 using accessible publisher methods, results and discussion excerpts and the abstract. This limited review narrows full-year comparisons and removes claims that comment analysis can determine why scale scores change.