Q: How to develop a research instrument?

Thank you for your question. The choice of a research instrument depends on the context and objectives of a study, and the instrument should always align with the former. When choosing your research instrument, you may consider the following:

• Is the instrument reliable? Are the results likely to be robust? For example, if your research instrument is a questionnaire, are the questions too specific for the data to be generalizable to other settings?
• Is the instrument based on a theoretical framework? This would underpin its validity (and therefore its acceptance as a reliable instrument by readers of your research).
• Is it an instrument that has been used in previous studies? This ties in with the previous point; there are many preexisting scales that can be used in multiple settings, eliminating the need to reinvent the wheel. However, each study is unique, and a scale or measure may need to be tweaked or adapted to the current study’s settings.
• Based on the previous point, does the instrument tie in with your (and the coauthors’, if there are any) perception of how the factors involved in your study relate to each other?
• Will the instrument be able to collect sufficient and pertinent data for the study, which can then be used for analysis and conduct tests to meet the study’s objectives?
• Will the instrument minimize (as complete elimination isn’t possible) bias?

When developing a research instrument, considerations should include making sure the instrument itself isn’t affecting the results. For instance, the questions in surveys (questionnaires or interviews) shouldn’t condition participants or encourage them to answer in a certain way, as this would introduce bias.

The response rate and accessibility as well as ease of use of the survey instrument play a critical role in its effectiveness. A low response rate and incomplete responses may defeat the purpose of conducting the survey, and a beneficial practice would be to conduct a pilot survey among a smaller group of participants to test if any tweaks are required to facilitate comprehension, for instance.

Another factor to consider is the amount of structure you’d like to introduce in your survey instrument. If responses to open-ended questions (in questionnaires or interviews) are likely difficult to quantify post survey, you may consider adding metrics that will facilitate the evaluation of the collected data at a later stage. Focusing on the weaknesses of a research instrument helps researchers determine ways to address them. Focus group discussions, for example, may not be as structured as interviews, and views may affect each other; the facilitator can conduct the discussion keeping this in mind to circumvent or minimize its effect.

A final consideration is ethical concerns and whether the instrument conforms with protocols. For example, randomized controlled trials may not be deemed suitable if the intervention that a particular group is deprived of is urgently required by both groups. Likewise, non-participant observations, given their nature, do not involve participant consent, and may be considered (depending on the journal/publisher) as disempowering for respondents.

If you have any other queries related to developing a research instrument, you should consider Editage’s Experimental Design service. Our academic experts will be happy to help!

All the best with your research!