ON MEASUREMENT TOOL DEVELOPMENT AND ADAPTATION STUDIES TO BE SENT TO IJATE
As of 1 April 2020, all measurement tool development and adaptation studies to be sent for review are expected to follow the rules explained below. The studies which are incompatible with the required rules will be rejected by the editors without taking the studies into review processes.
1.Manuscript should include sample items from the original form (the language in which the tool was developed).
2.The full measurement tool should be sent as a separate file to be reviewed by the reviewers.
3. The structure which the tool is aiming to measure should be defined in detail in the introduction part of the manuscript.
4. If there are other tools developed to measure similar properties in the literature, the manuscript should explain the difference, superiority and limitations of its own with reference to other tools by citing them.
5. It should be explained that whether the assumptions regarding data analyses used are met or not.
6. For the reliability proofs of the measurement tool;
a. More than one reliability proof should be given.
b. In multi-dimensional measurement tools, because of underlying assumptions, besides Cronbach Alpha coefficient for the full measurement tool, Composite Reliability coefficient should also be reported (For further reading, see Bacon, Sauer, & Young, 1995).
7. For validity proofs reported;
a. It is mandatory to include evidence of content and construct validity.
b. In order to prove structure validity, not only factor analytical techniques, but also hypothesis tests are also expected.
c. It is also recommended to test the measurement invariances of the developed scales in the process.
d. When factor analytical techniques are used for proof of construct validity;
i. First, exploratory factor analysis should be conducted and the structure measured by the measuring tool and the theoretical structure should be compared.
ii. During exploratory factor analysis, factor extraction techniques other than principal component analysis should be preferred (For detailed reading, see Tabachnick & Fidell, 2013, p.614).
iii. Correlation matrix suitable for scoring of measurement tool items should be used (e.g. Tetrachoric, etc.).
iv. The method used while deciding the number of factors should be explained (Parallel analysis is recommended).
v. Factor load and contribution criteria to common variance should be stated.
vi. Confirmatory factor analysis should be performed by applying the form formed as a result of exploratory factor analysis in another group and the structure should be proved to be checked in another group as well.
8. If the measurement tool is adapted from another culture;
a. Permission taken from the developers of the original form of the tool should be documented and uploaded to the system.
b. Evidence that the structure measured by the tool is found in the adapted culture should be included.
c. Instead of developing a new measurement tool, it should be clearly justified why it was decided to adapt.
d. Since the factor structure of the structure that the measuring tool is trying to measure is certain, confirmatory factor analysis should be performed and the original form of the scale should be adhered to. If changes are made to the scale items, permission should be obtained from the developers of the original form of the scale accordingly.
e. For the confirmatory factor analysis, assumptions should be tested and relevant evidences should be provided.
f. For the confirmatory factor analysis, which estimation method is used, whether it is modified or not should be justified and which fit indices are chosen to be used should be explained.
9. Software and their version numbers should be provided for all analyses conducted.
Bacon, D. R., Sauer, P. L., & Young, M. (1995). Composite reliability in structural equations modeling. Educational and Psychological Measurement, 55(3), 394-406.
Tabachnick, B.G., Fidell, L.S. (2013). Using Multivariate Statistics, 6th Ed. Pearson.