There is a debate about whether or not you should include neutral responses in a Likert scale. Here are three common arguments you'll find:

1) Opinion-based questions may be better without the neutral category.

2) Information-based questions may be better off with an "I don't know" response, as they may lack this information and you won't know this if they leave the question blank or try to guess.

3) Neutral responses encourage participants to select that category throughout the questionnaire.

In my own research, I use neutral responses when warranted--I consider the context and whether neutral is a meaningful response. You can always truncate those responses to agree/disagree where you're predicting agreement or disagreement. For example, if you were to predict agreement with negative attitudes about treatment on a five-point Likert scale, then you could put strongly agree and agree into one category (agreeing with the attitudes) and neutral, disagree, and strongly disagree in another (not identifying with those attitudes). Whether this is a reasonable strategy depends on your research question and how the survey question was worded.

If you're asking about confidence in ECGs, then it could be set up as, "How confident are you in interpreting ECGs?" 1) No confidence, 2) some confidence, 3) moderate confidence, 4) much confidence, 5) complete confidence. Using this strategy lets you interpret their responses and helps the respondent interpret the question.

If the respondents were giving you "neutral", then look at the remainder of their completed questionnaire. Was this question at the end and they just gave "neutral" down the form? That means they likely got tired and didn't give you a meaningful response. Are they giving you a good range of responses around that question? In that case, it means that they are either 1) unsure about what you mean/the question was nonspecific or 2) the group is is neither overly confident nor concerned about their ability to interpret ECGs. Ask a subject-matter expert (ideally, someone who completed the form) to see how they interpreted that question to understand whether they were confused or felt that it was too vague to give a meaningful response.

To avoid any confusion in interpreting LIKERT SCALE is to use the number corresponding to the verbal description and treat that as numerical values (quantitative data). Whichever corresponded to the "neutral" responses, interpret that in the numerical correspondence as the level of confidence of the ECG reading.

neutral does not choose both (agree or disagree) / do not have an opinion. Not taking sides with agreement or disagreement

Eekim Kee, I appreciate your concern about the neutral response option, and I think it might be a perennial problem particularly because so many researchers use a strongly agree to strongly disagree response continuum.

You might find the following article helpful:

Nadler, J. J., Weston, R., & Voyles, E. C. (2015). Stuck in the middle: The use and interpretation of mid-points in items on questionnaires. The Journal of General Psychology, 142(2), 71–89. https://doi.org/10.1080/00221309.2014.994590

In some research I'm currently working on with a colleague, we're looking at the extent to which the middle response option is used on a particular scale. In a number of pieces of research, we have found that more than 20% of the participants use the middle option on approximately half of the items. Inter alia, we will be examining the prospect that the "high use" items are in some way defective, e.g., by being poorly worded or irrelevant for many participants.

All the best with your research.

I agree that a mid-point on a Likert scale is controversial. However, I find myself convinced by the deliberation of Chyung et al. (2017). They argued that it is not a matter of having or not having a mid-point on a Likert scale; rather, it is when to include or omit the mid-point. Based on their recommendations, for example, you could include a mid-point if your respondents are familiar with the survey topic or you want to consider your scale as an interval one for conducting a statistical procedure. For insights to help you decide, you could go through the recommendations presented in Table 1 by Chyung et al. (2017, p. 19), fully cited below.

Chyung, S. Y. Y., Roberts, K., Swanson, I., & Hankinson, A. (2017). Evidence-based survey design: The use of a midpoint on the Likert scale. Performance Improvement, 56(10), 15–23. https://doi.org/10.1002/pfi.21727

Article Evidence-Based Survey Design: The Use of a Midpoint on the L...

Good luck,

Tq everyone appreciate it