ABSTRACT
The purpose of this chapter is two-fold. First, I wish to share the insights that I gained about test development while working in the Languages Group of the Test Development department at the Educational Testing Service (ETS) from 1984 to 1987. Second, I wish to describe the creation of a particular reading comprehension test I prepared for a TOEFL (Test of English as a Foreign Language) administration in 1986. I will begin the chapter with a brief description of the TOEFL as a whole, introduce readers to some basic terminology used in psychometric testing, and discuss some of the procedures I followed in the development of a TOEFL reading comprehension test. Thereafter, I shall use a passage I assembled, reviewed, and pre-tested for a TOEFL reading test to illustrate some of the debates that arise in the process of test development and to show the kind of information that is provided by a statistical analysis of individual items in a reading comprehension test. Additional analysis of the passage and items can be found in Peirce (1992).
