In order to create tests appropriately adapted for Morocco, we had to deal with a number of complex methodological problems. Testing instruments had to make sense to the Moroccan children and their parents, as well as be of relevance to the scientific community (the emic-etic dilemma discussed in chapter 1). Second, the project studied literacy in Standard Arabic and

Arabic orthography, for which the scientific literature on testing is extremely limited. Further, we worked with sample populations in Morocco who speak a varied mix of languages (Moroccan Arabic, Standard Arabic, Berber, and French), and had to prepare instruments in at least two and, in some cases, all four languages in order to gather needed data. Needless to say, problems of translation and back-translation, and pilot testing, took a great deal more time and effort than in the typical research project in the United States.

Such inherent difficulties also led to increased vigilance concerning the validity of the test instruments, and to a number of innovative approaches. For example, because of the nature of Arabic orthography and Arabic writing (such as differing forms of a given letter depending on its place in a word), we were able to develop several new ways of studying children's early prereading knowledge (see chapter 5). In addition, because we observed the style of rote memorization in the Quranic school, we were able to create cognitive tasks that could mimic, to a greater or lesser extent, the possible consequences of this practice (see appendix 1).