In general, are item specific (IS) response options better than agree/disagree (A/D) Likert scale response option in terms of reliability, validity and quality? I am looking at it from the perspective of medication adherence questionnaire. Reading few papers, it appears to me that A/D is cognitively difficult to comprehend, indirect, burdensome for respondents and prone to acquiescence bias compared to IS. But this is relatively new topic for me so looking to hear from people with experience and expertise in this area. Many thanks in advance.
Robert Trevethan, you have been a great source of learning for me. Whenever I read your input, I find myself enlightened. Thanks for your input on this particular issue. I find it greatly useful.
Asta R Prajapati, there has been a fairly large amount of literature written about the kind of thing you're concerned about. I think a very good place to start exploring is the following:
Krosnick, J., & Presser, S. (2010). Question and questionnaire design. In P. V. Marsden & J. D. Wright (Eds.), Handbook of survey research (2nd ed., pp. 263–313). Bingley, UK: Emerald.
From my own experience, I suggest you aim for 5 to 7 response options and that, in addition, you include options such as don't know and not applicable.
Robert Trevethan, you have been a great source of learning for me. Whenever I read your input, I find myself enlightened. Thanks for your input on this particular issue. I find it greatly useful.
Muhammad Mumtaz Khan, thanks, and you're very welcome.
In case it's of any use, here are some more references for the topic at hand. I'm sorry I didn't provide them in my earlier post, but I was a bit preoccupied and overlooked doing so.
Dolnicar, S. (2013). Asking good survey questions. Journal of Travel Research, 52, 5, 551–574.
Fink, A. G. (2003). The survey handbook. (2nd ed.). Thousand Oaks, CA: Sage.
Fink, A. G. (2017). How to conduct surveys: A step-by-step guide (6th ed). Thousand Oaks, CA: Sage.
Fowler, F. J., Jr. (1995). Improving survey questions: Design and evaluation. Thousand Oaks, CA: Sage.
Kelley, K., Clark, B., Brown, V., & Sitzia, J. (2003). Good practice in the conduct and reporting of survey research. International Journal for Quality in Health Care, 15, 261–266.
Kitchenham, B. A., & Pfleeger, S. L. (2002). Principles of survey research, Part 3: Constructing a survey instrument. ACM SIGSOFT Software Engineering Notes, 27(2), 20–24.
Platek, R. (1985) Some important issues in questionnaire development, Journal of Official Statistics, 1(2), 119–136.
Punch, K. F. (2003). Survey research: The basics. London: Sage.
Rattray, J., & Jones, M. C. (2007). Essential elements of questionnaire design and development. Journal of Clinical Nursing, 16, 234–243.
I should acknowledge that someone else provided these references in a post to an RG thread about a month ago, and at the time I simply copied them across to a document among my files for my personal use. Since then, however, I have gone through them, cleaned them up a bit, and tried to present them in the format of APA 6th.
Robert Treventhen, I am really pleased with your new input. It is greatly useful.
Thank you Robert Trevethan for references. I was particularly keen to see if item specific response option (1) were more reliable, valid and has better quality than agree/disagree response option (2). For example:
1. How often do you forget to take your medication?
Never Rarely Sometimes Often Always
2. I forget to take my medications.
Strongly disagree disagree Neither agree nor disagree Agree Strongly agree.
The evidence I have seen so far suggests that 1 is better than 2 (for the reasons I mentioned above) but looking to see what people thought.
Asta R Prajapati, I'm glad that what I've suggested has been helpful.
I think, apart from that, it's really important to use common sense when creating response options. So, with the two possibilities in your post up above (and without going into details), I think the first is much better.
One useful thing to do is to "pilot test" different versions on some people and ask them what they prefer, and why.
I completely agree with Robert Sir. Having 5 options and two direct response will help the respondents to select their choice better.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire