I’m currently required to make a questionnaire for my research. Is it okay to use only specific questions from previously published questionnaires as these questions perfectly apply to my research context? Will that be considered a plagiarism if I cited these questions?
I don't see this as plagiarism. Plagiarism is passing something off as your own words (or thoughts perhaps) without attribution. If you publish an article under your name as author which contains unattributed text copied from elsewhere, that is plagiarism. But here you are using a questionnaire as a tool to collect information. You are not claiming you wrote those questions nor would it be reasonable for someone to assume that. In fact, it would be unexpected and bizarre to cite sources for the questions in a questionnaire. Things could be different if you make other use of the questionnaire. For example, if you publish it as an appendix to an article you probably should include some attribution to sources.
"Is it okay to use only specific questions from previously published questionnaires as these questions perfectly apply to my research context?"
Ans: Using specific questions from previously published questionnaires may be acceptable if they are relevant and appropriate for your research context and objectives. However, you should not rely solely on existing questions (to cherrypick) as they may limit the scope and depth of your inquiry. Instead, you should design your own questionnaire based on a thorough literature review and a clear research gap. This will ensure that your questionnaire is novel, robust, and tailored to your specific case.
Will that be considered a plagiarism if I cited these questions?
Ans: Questions should only be cited if they are relevant to the same context. In research questions, context is crucial. Different fields of research may pose similar or identical questions but with different meanings and implications. For instance: “Did the proposed model improve the performance?” This question could appear in both an Economics paper and a Software Engineering paper. However, these papers should not cite each other’s questions, as it is too common of a question and belongs to different contexts.
If the context is partially overlapping, it is advisable to address this issue of question repetition in the Related Works section, where the differences in context for the same questions can be further discussed and clarified.
An example can illustrate this point: suppose I am proposing model X in my paper P for a certain task. The current state-of-the-art (SOTA) approach, model Y, was proposed in the paper P’ for the same task. I want to include the following question in the questionnaire for the user study of my paper: “On a scale of 0 to 10 (0 being completely disagreed and 10 being completely agreed), do you think the proposed model performs better with respect to the SOTA approach?”. I have already explained the ‘proposed model’ and ‘SOTA approach’ (model Y) to the user group. However, when I examine paper P’, I find the exact same question (verbatim) used by the authors of P’ to validate the superiority of their model Y. I need to use the same question for my model X to demonstrate its superiority over model Y. Therefore, I acknowledge this issue in the Related Section, where I cite paper P’ and explain that I have followed their work and asked similar questions in my user study for the same purpose.
First, you need to understand the difference between plagiarism and copyright infringement. If you cite your source, it's not plagiarism, but if you copy the questionaire without permission, then it is copyright infringement.
Second, the questions in a questionaire are rarely random. Very often a questionaire contains questions where the interviewer has no interest in the answer whatsoever, but these questions are added so the interviewee cannot figure out what the interviewer wants and cannot give the answers that they think the interviewer wants to hear. So if you remove these "pointless" questions, then the answers to the other questions can and will change.