Assessment of second language reading comprehension has evolved from a relatively narrow conceptualization of reading as a process of mastering hierarchically ordered subskills, with the author as primary creator of meaning, to reading as an interaction among reader, author, and text. Reading assessment has several purposes: sorting students; diagnosing individual problems; and evaluating instructional effectiveness. Most measurement methods are based on a psychometric perspective, but a cognitive approach, which sees reading as a constructive process, may give more insight into why a learner is able or unable to master specific objectives. Conventional measures of reading comprehension include multiple-choice questions, short-answer questions, and cloze tests. Currently, text recall is considered the best method for inferring comprehension. Criticism of the method focuses on its inappropriateness for the English-as-a-Second-Language situation, absence of an objective weighting and analyzing system, time consumed, holistic approach, and lack of differentiation of processes and skills used. However, a constructive activity scale can be used with a recall protocol to identify the cognitive activities involved in text comprehension. Such a scale would analyze activities on four levels: prepropositional/fragmented associations, knowledge/details retelling, assimilation, and problem-solving and integration. The proposed method allows for both quantitative and qualitative assessment. (MSE)