Scale Construction and Development
A 2-Day Seminar Taught by Tenko Raykov, Ph.D
Researchers in a wide range of disciplines are frequently involved in the development and revision of multiple-item measuring instruments, including scales, tests, inventories, questionnaires, surveys, subscales, and testlets. Scores obtained from these measuring instruments are usually employed in subsequent analyses that address substantive research questions. To a substantial degree, the quality of these instruments determines the extent to which the analyses and modeling efforts produce trustworthy results. To construct scales of high psychometric quality, researchers must engage in many activities aimed at building initial versions of the instruments and then repeatedly revising them to improve their performance.
This two-day seminar provides a comprehensive introduction to the field of scale construction and development in the social and behavioral sciences, especially as applied to empirical settings where measuring instruments consist of multiple components or items with no guessing. Throughout the seminar, many empirical examples are utilized from the clinical, educational and social disciplines. The popular latent variable modeling software Mplus is used for all the examples, along with a detailed discussion of the command syntax and interpretation of the output.
Participants in this seminar can expect to come away with:
1. A nuanced understanding of the conceptual foundations and basic mathematical and statistical relationships underlying behavioral instrument construction and development.
2. The ability to understand, interpret and explain the output from statistical software for achieving the goals of psychometric scale construction and revision.
3. An appreciation of the advantages of a thorough study of the underlying latent structure of a tentative version of a multi-component instrument.
4. Practical tools and strategies for constructing an initial version of a scale.
5. Methods for revising a scale in order to achieve higher psychometric standards.
6. The ability to deal with issues arising in the practice of scale construction and development.
Who should attend?
To benefit from this seminar, you should have the equivalent of one or more semesters of statistics: a good introductory course with some treatment of probability and random variables as well as regression analysis. Some knowledge of multivariate statistics would also be helpful, but is not essential.
Schedule and materials
The class will meet from 9 to 4 each day with a 1-hour lunch break.
Participants receive a bound manual containing detailed lecture notes (with equations and graphics), examples of computer printout, and many other useful features. This book frees participants from the distracting task of note taking.
Registration and Lodging
The fee of $895 includes all course materials.
Lodging Reservation Instructions
A block of rooms has been reserved at the Club Quarters Hotel, 1628 Chestnut St., Philadelphia, PA at a nightly rate of $137 for a Standard room. This hotel is about a 5-minute walk from the seminar location. To make a reservation, you must call 203-905-2100 during business hours and identify yourself by giving the group code STA114. For guaranteed rate and availability, you must make your reservation by October 17, 2013.
This seminar will use Mplus for all examples, but prior knowledge of Mplus is not essential. In addition, for convenient and quick confidence interval construction, short R-functions will also be provided. However, knowledge of R is not required. You are welcome to bring your own laptop computer, and outlets will be provided at each seat. No internet service will be provided, however.
- Resources for the seminar.
- Factor analysis – a modeling basis for instrument construction and development in the social and behavioral sciences; exploratory and confirmatory analyses, models, and their fitting to data.
- Scale development with categorical items – examining latent structure for a set of categorical items and instrument improvement.
- Construction of initial scale version – how to choose items from a possible pool; point and interval estimation of item difficulty, inter-item correlations, and item-total correlations, all doing it the right way.
- Scale revision to enhance psychometric quality – coefficient alpha and reliability; point and interval estimation of reliability as well as of change in reliability due to revision, and why “Alpha if item deleted” should generally not be used.
- Essential unidimensionality of multiple component measuring instruments – when could we consider a scale practically unidimensional even though it may not be strictly homogeneous.
- Some practical issues in scale construction and development – what may be relevant in a practical setting for instrument development and revision; validity, multidimensional, and hierarchical scales.
- Scale construction and development with data from nationally representative studies.
- Optimal shortening of scales.