Find validated items in the original article that developed the scale (the method section or appendix lists the items), then confirm them in later studies that used the scale; adapt the wording only minimally, and re-test reliability on your own data against a cited threshold such as Cronbach's alpha ≥ .70 (Nunnally, 1978). Using a published scale saves you from developing and validating a new one, a long process with its own stages (Boateng et al., 2018), and gives examiners confidence that you are measuring what you claim.
Step 1: define the construct precisely
Before searching, write a one-sentence definition of each construct. "Trust" in an online retailer, "trust" in a manager and "trust" in science are measured with different scales. Your definition decides which scale fits, and the examiner will compare the two.
Step 2: search in the right places
- Highly cited studies in your field. Find a recent study testing a similar model, look at its measurement table, and trace each construct to the original source it cites.
- The original development paper. Always read it. Items are usually in a table or appendix, with the response format and any reversed items.
- Scale handbooks and databases in your discipline (marketing, organisational behaviour, psychology and health each have compilations of measures).
- A curated scale library. A library that lists original wording, reversed items, the source and the licence saves time, but always check against the original article.
Step 3: check the evidence and the licence
A scale is "validated" only in the sense that someone reported evidence for it. Look for: reliability (alpha or composite reliability of .70 or higher; Nunnally, 1978; Hair et al., 2019), factor analysis results, and use in several samples. Check the licence as well. Many academic scales are free for research, but some clinical and commercial instruments require permission or a fee. If the article or the author's website states terms, follow them.
Step 4: adapt with care
- Change only the context, for example "this system" becomes "this banking app". Keep the structure of each item.
- Keep all items where possible. Dropping items before data collection reduces content coverage.
- Keep the response format unless you have a reason; if you change a 7-point scale to 5 points, say so.
- Translate properly if your respondents use another language: forward translation, back-translation and reconciliation by bilingual experts (Brislin, 1970).
- Pilot test with a small group from your population to catch confusing wording.
Worked example
A student studying mobile banking adoption needs a measure of Perceived Usefulness. Searching recent technology-acceptance studies, she finds that most cite Davis (1989). In the original article she finds the items, which refer to using "the system" in "my job". She adapts four of them to her context, for example "Using this banking app would enable me to accomplish tasks more quickly", keeping the 7-point response format.
After data collection (N = 230), she checks reliability: Cronbach's alpha = .89, above .70 (Nunnally, 1978), and all standardised loadings exceed .708 (Hair et al., 2019). Her measures section reads:
Perceived Usefulness was measured with four items adapted from Davis (1989), rated on a 7-point scale from 1 (strongly disagree) to 7 (strongly agree). The items were reworded to refer to mobile banking. Reliability in the present sample was good, α = .89.
She lists the full adapted wording, with the original source of each item, in an appendix table. That table is the first thing many examiners look for.
Common mistakes
- Citing a recent study that used the scale instead of the original source.
- Mixing items from different scales without saying so.
- Reporting the original paper's alpha as if it were yours.
- Forgetting to recode reversed items before analysis.
Run it free
Browse the validated scale library for original English items, reversed items, sources and licences. When your data are in, check reliability with the free Cronbach's alpha calculator, and format the source with the APA citation generator.
FAQ
Can I change the wording of validated items?
Minimally, to fit your context. Report that the items were adapted and show the final wording.
Do I need permission to use a published scale?
Many research scales are free to use with citation, but some instruments are copyrighted and licensed. Check the source.
What if no validated scale exists for my construct?
Look for the closest construct with an established scale first. Developing a new scale requires item generation, expert review and factor analysis (Boateng et al., 2018), which is usually beyond a master's dissertation.
Should I still run reliability tests on a validated scale?
Yes. Reliability is a property of scores in your sample, so it must be reported for your data.
References
- Boateng, G. O., Neilands, T. B., Frongillo, E. A., Melgar-Quiñonez, H. R., & Young, S. L. (2018). Best practices for developing and validating scales for health, social, and behavioral research: A primer. Frontiers in Public Health, 6, Article 149. https://doi.org/10.3389/fpubh.2018.00149
- Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1(3), 185–216. https://doi.org/10.1177/135910457000100301
- Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008
- Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. European Business Review, 31(1), 2–24. https://doi.org/10.1108/EBR-11-2018-0203
- Nunnally, J. C. (1978). Psychometric theory (2nd ed.). McGraw-Hill.