Introduction

Principles of Dynamic Assessment (DA) have recently been used to instruct and assess learners' language ability, especially in classroom contexts. DA is an approach to assessment and instruction that has derived its principles and procedures from Vygotsky's theory of the Zone of Proximal Development (ZPD). Dynamic assessment of second language learning (L2 DA) envisions that since learners' language ability is always approaching the potential zone (that is why it is called dynamic), the conventional Non- Dynamic Assessment (N-DA) tools are not able to tap the learners' real abilities. In other words, the advocates of DA claim that Static Assessment (SA) can only measure the learners' actual level of performance (what they can perform independently) but cannot assess their potential level of ability (what they can perform with assistance). As Poehner and Lantolf (2005) express "the principle underlying DA is that a full picture of what an individual or group is capable of does not emerge unless and until the ability is not only observed in independent performance but is also pushed forward through specific forms of intervention and/or social interaction between learners and assessors" (p. 261).

In the last two decades, we have fortunately witnessed a growing interest among language pedagogy specialists to add DA assumptions and principles to their profession. Lantolf and Poehner are two people who have pioneered the application of DA principles in L2 education in the last ten years (Poehner & Lantolf, 2010; Lantolf & Poehner, 2010; Poehner, 2008; Poehner, 2007; Poehner, 2005; Lantolf & Thorne, 2006; Poehner and Lantolf, 2005; Lantolf and Poehner, 2004; Lantolf, 2000). A few other people have also contributed to the expansion of the field. Poehner (2005), for example, made a comprehensive research on the dynamic assessment of advanced L2 learners of French; Ableeva (2008) investigated the effects of dynamic assessment on L2 listening; and Kozulin & Garb (2002) studied the effect of dynamic assessment on the EFL learners' text comprehension.

Statement of the problem and purpose of the study

Mainstream non-dynamic approaches to assessment suffer from some major problems. A pragmatic problem is that the decisions test givers make about test takers are merely based on how the testees perform on test items (their actual not potential performance). Secondly, due to intercultural and interethnic differences of learners and examiners, conventional standardized tests are not able to measure testees’ real abilities. Some cultures impose limits on the testees in responding to specific kinds of test formats; therefore, their incorrect responses to test items may be due to some non-ability factors like their unfamiliarity with the test format rather than to their lack of knowledge. The third problem with conventional tests is that instruction and assessment are kept as two different activities. Usually the testers wait for the instruction to be completed, and then they start to measure what learners have gained from instruction. Testees receive no feedback on their performance after the test is administered. The only feedback they receive is the scores reported by the testers to the school administrators.

The problems with N-DA approaches listed above are convincing enough to modify the structure of mainstream psychometric assessment procedures being used presently and to design new dynamic assessment procedures. However, from the practicality point of view, existing DA models cannot be used in every context as easily as N-DA procedures and this is one, and probably the main, reason DA is not warmly welcomed by conventional testers.

A common practical problem with both interactionist and interventionist approaches of DA is that they are time consuming. A specific problem with the interactionist approach is that especially trained mediators (teachers) are needed to help learners within DA framework. Both sandwich and cake formats of interventionist approach may be suitable for classroom settings where learners are at hand for a long period of time, but not for non-classroom settings. Thus, we need a new format within the interventionist approach of DA that does not possess the stated problems of sandwich and cake formats. We name it

sauceformat to follow the tradition of using terms from nutritional science.

In view of the theoretical problems with N-DA and practical problems with current approaches of DA, this study pursued the following two purposes:

1) to explore whether sauce format mediation could help EFL testers and teachers to observe the testees and learners' potential level of reading development, and

2) to propose a new format of mediation by which EFL learners' both actual and potential proficiency of reading comprehension skills and strategies could be assessed easily.

Significance and justification of the study

The significance of this study lies in the fact that it attempts to explore the feasibility of developing a new dynamic reading test that will help teachers and testers to understand and develop their learners reading strategies. By introducing sauce format mediation, dynamic tests can be used in non-classroom settings as well as in classroom contexts where subjects are not at full access of testers.

Research questions

The specific goal of this research was to propose sauce-format mediated reading test that would assess Iranian EFL learners' both actual and potential proficiency of reading comprehension skills and strategies. For this purpose, it was hypothesized that if reading comprehension tests were accompanied by mediation boxes, testees' potential, as well as actual, level of performance would be assessed. To test the stated hypothesis the following questions were intended to be answered:

1. Does sauce-format mediation affect Iranian EFL learners' performance on reading comprehension tests?

2. Can Iranian EFL learners transcend their assisted performance on a sauce-format mediated reading test to independent non-mediated reading tests?