Students’ Perceptions of AI Digital Assistants (AIDAs): Should Institutions Invest in Their Own AIDAs?

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper deals with students' perceptions of Artificial Intelligence Digital Assistants. In the current context of proliferating publications on the use of AI and GenIA in education, this study seems to have three notable contributions: a large sample size, a study in a sufficient time frame to see the evolution of students' perceptions, and a focus on institutional vs. public AIDAs.

The topic is certainly interesting, and the ideas presented in the manuscript are correct. However, I have some main concerns and, then, minor suggestions.

Main concerns:

• I would suggest you review the reference [8], since I think that they not exactly found higher-order thinking propensities of students using ChatGPT relative to not using ChatGPT, as you state in lines 35-37.
• Lines 56-58: you state that “This -institutional AIDA- would allow students and educators to securely use, trust, and share their data and IP (e.g., course materials, assessments, marking criteria) [16,17]”. However, I think that reference [17] does not support that, since it is not about private or institutional AIDA
• I would like to read your own paper [22] for understanding the study 2 and comparing with this study 3, but I can’t find it.
• The average age of your youngest group is 37. I would suggest you to expand your discussion by comparing your results with the other related studies with younger students. Consider, for example, the following one (students aged between 18 and 34): "A. E. Cotino Arbelo, C. S. González-González, J. Molina Gil. Youth Expectations and Perceptions of Generative Artificial Intelligence in Higher Education, International Journal of Interactive Multimedia and Artificial Intelligence, vol. 9, no. 2, pp. 84-92, 2025", http://dx.doi.org/10.9781/ijimai.2025.02.004

  Minor suggestions:

• In the first paragraph of Introduction, I would eliminate the first sentence: that describing the special issue focus (line 23).
• Line 47: do you mean “with other learners”?
• Line 62: “off-the-self” instead of “off-the-shelve”?
• Lines 80-81: Since you are author of [12], consider to say “We described….” Or at least to use “et al.” instead all the authors’ names.
• Table 1: Time of measurement seems wrong for Study 3 (86%)
• Pages 9-10: quotations are too long. I think you should present them indented (or in an Annex)

Author Response

Please find attached our revision document

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents the results of 3 studies to assess the student’s perceptions on AI Digital Assistants, comparing the use of generic systems with institutional assistants.

Overall, the article is well written and structured. In addition, the topic is highly relevant and of great interest to readers of Applied Sciences.

However, since this article is related to other previous publications, it must better clarify some details instead of “just redirecting” the reader to previous publications. Based on that, this manuscript must be self-contained and provide more details about several topics. In detail:

• Lns 89-90. Please explain the instrument for Study 2. What is this instrument? The reader needs some details about the instrument. I might then consult the reference, but as it is know, it is impossible for the reader to understand that without reading reference 22.
• Lns 111-112. Please explain the types of quantitative analysis performed. Just referring that it was done using SPSS is quite irrelevant (it could have been done in python, R, etc…). Indeed it is relevant to mention which tool was used, but much more important it to mention the types of quantitative analysis methods that were used.
• Lns 111-112. As in my previous comment, please provide details about the qualitative analysis. Indeed you must provide reference 24, but the reader must be informed about the strategy used for this qualitative analysis.

Section 3 uses the research questions to guide the presentation of results. First, the authors must correct the RQ phrasing, they are not written in the same way as they were in the introduction, lacking rigor.

Figure 3 lacks quality and it is not acceptable to provide legends like Female: 0; 1. If I understand it correctly, blue is male, red is female. Also, today is not common to classify gender in just male / female. I do not want to provide a solution for that, but please take this in consideration, as it really depends on legal / cultural aspects. I do not understand what means this dot on study 2 surrounded by 80 81 82 34.

Please reformat tables, starting with Table 2. The values for mean and standard deviation can use much less space, improving overall readability.

 

Overall, the quality of Figures must be corrected, as they do not provide adequate resolution. Also, Tables must be reformatted to better used available space and be inside the established margins.

Table 6 presenting a correlation matrix is very difficult to read. Since it has several variables, a heatmap would provide a much easier way to provide this information.

Although I understand the option to present the results based on the research questions, from a methodological perspective, I would make much more sense to present the results, independently. The authors must choose which variables and types of visualization are better to present the meaningful results. After that, they should discuss how these results provide answers to their RQ.

In fact, current Discussion section is mainly a conclusion. The discussion must discuss the results in relation to their RQ. Then, the authors should clearly conclude their work presenting a conclusions section, where they made their conclusions, research gaps, future directions, limitations, etc.

Author Response

Please find attached our revision document

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript has been sufficiently improved.

Reviewer 2 Report

Comments and Suggestions for Authors

Overall, previous comments have been adequately addressed.