Measuring service quality at an online university: using PLS-SEM with archival data

Research output: Contribution to journalArticleAcademicpeer-review


The aim of this study is to analyze, evaluate and validate the NSE (National Student Enquiry) as a service quality measure helping both higher education institutions (HEIs) and students in their decision making. Every year the Dutch foundation ‘Studiekeuze123’ sends out a survey (the NSE) to collect data on service quality regarding education at HEIs in the Netherlands. We used the 2019 NSE-data from the only e-learning university in the Netherlands, the Open Universiteit (OUNL), containing a sample of 1287 students. PLS-SEM was used to analyze a conceptual model in order to understand the service quality factors that promote students’ level of satisfaction and willingness to recommend the HEI. Overall, the findings reveal that the quality of the NSE is sufficient to be used for performance analysis. Nine out of twelve service components taken into account for the OUNL are found statistically significant affecting students’ satisfaction and willingness to recommend. The results help HEIs promoting and managing students’ perceptions of the quality of education and support students in their decision making process. Since many HEIs had to make a transition from onsite to online education within a short period of time, due to the Covid-19 pandemic, service quality became a major concern for HEIs. As online learning systems are expected to stay, analyzing the service quality of the OUNL as a reputed online HEI can help other HEIs getting their online learning systems on track.
Original languageEnglish
Pages (from-to)161-185
Number of pages25
JournalTertiary Education and Management
Issue number2
Publication statusPublished - 16 Jun 2021


  • Student satisfaction
  • service quality in higher education
  • Structural equation modeling
  • E-learning
  • Archival data


Dive into the research topics of 'Measuring service quality at an online university: using PLS-SEM with archival data'. Together they form a unique fingerprint.

Cite this