DA which is a pedagogical approach and a development-based activity (Poehner, 2005) is theoretically framed within the works of Vygotsky and contends that unlike traditional testing methods instruction and assessment are dialectically integrated. Some key concepts are laid at the heart of the notion of DA. The first important concept is mediation; the process by which other-regulated activities are transformed into self-regulated ones (Lantolf & Thorne, 2006). In the same vein, Aljaafreh and Lantolf (1994) proposed a model of mediation from other-regulation to self-regulation in learners which included five transitional levels starting from the most implicit or indirect to the most explicit or direct. These levels are, in fact, indicative of three stages: 1. object-regulation, 2. other-regulation, and 3. self-regulation. As the theoretical underpinnings of DA, mediation and regulation of high importance in understanding which type of mediation should be offered to whom, at what extent, and when. Practically, this is a tremendous task to do in educational contexts in case the interactionist DA is utilized, as “the levels [are] not determined in advance of the study. (Aljaafreh & Lantolf, 1994)” To make it practically feasible, this study made use of C-DA which utilizes the interventionist approach to DA.
According to Lantolf (2000), one of the forms of mediation is the concept of regulation. Frawley and Lantolf (1985) defined regulation as how an individual sees a task and also their ability to successfully complete that task. It is one form of mediation and goes through three stages to complete its process. The stages, respectively, are as follows:
In the first stage or object-regulation, individuals use objects in their environment in order to think. That is, an object tells us to do something; a persuasive advertisement, for instance. In this regard, Poehner (2008, p.27) commented that “At the level of object regulation, psychological functioning is controlled by the environment rather than by the individual, and so in response to hunger the individual eats what is immediately available or goes in search of food.”
In the second stage or other-regulation, individuals’ performance is primarily controlled by someone else (Lantolf & Poehner, 2011). That is, it includes implicit and explicit mediation by parents, peers, teachers, so forth. Here someone tells us to do something; for instance a mother tells her child to do his homework.
In the third stage or self-regulation, minimal or no external assistance is required from the individuals’ side to accomplish activities. In other words, individuals establish control over their own performance (Lantolf & Poehner, 2011). We tell ourselves to do something; for instance ‘I need to finish my MA thesis before Ramadan.’ In fact, self-regulation enables us to control our responses in order not to merely act instinctively but instead choose from among possible alternatives intentionally (Poehner, 2008). Preferring not to eat anything in an effort to lose weight while being invited by a friend of yours is an example of this kind.
In this regard, Vygotsky (1978) argued that moving from other- or object- regulation to self-regulation is the primary way in which humans develop higher order thinking skills. In other words, a learner has to pass from being object-regulated to being self-regulated for development to occur (Summers, 2008) and this movement is termed Internalization; a process through which higher mental functions are created.
The importance of the type of mediation or interaction which is provided for learners is reflected in Vygotsky’s belief who stated that learning occurs as the result of interaction, but not any kind of interaction, i.e. it only emerges as the result of interaction within the ZPD; the theoretical underpinning of DA (Kozulin & Gindis, 2007) implying that potential development differs from actual development (Poehner & Lantolf, 2005). That is to say, what the individual is able to do one day with assistance, s/he is able to do tomorrow alone. This means that depending on an individual’s ZPD, the mediator should match the provided interaction to that person’s potential for better results. Out of what was just stated, it can be understood that People’s ZPD is not fixed but instead it is a malleable and open-ended trait of them which can become apparent through interaction and consequently develop the potential for learning, of course, if suitable opportunities are provided (Wells, 1998).
As it is clear from these discussions, almost all ZPD-based studies cannot be conducted without relevant help or assistance. Any assistance from the mediator’s side should have two important properties (mechanisms) to be effective: First, it should be gradual and second, it should be contingent. Different researchers have used different terms to refer to these two properties. Summers (2008) referred to these mechanisms as ‘quality mediation’, for instance. Any help which has these two properties is referred to as ZPD-based help (Tajeddin & Tayebipour, 2012) or negotiated help (Nassaji & Swain, 2000) and if it does not have these mechanisms, it is called, according to the just-mentioned studies, random help; that is there is no attempt to adjust the level of assistance to the learner’s responsiveness. Aljaafreh and Lantolf (1994) defined the former, i.e. graduation, as help which moves from highly implicit level through concrete and more and more concrete levels until the appropriate level is reached. Of course, the assistance from the mediator’s side should not be too explicit to let him/her take over more of the activity than is necessary. They (p.468) also defined the latter, i.e. contingency, as help which “should be offered only when it is needed, and withdrawn as soon as the novice shows signs of self-control and ability to function independently.” In another definition, Gibbons (2003, p. 267) stated that contingency consists of the “assistance required by the learner on the basis of moment-to-moment understanding.” Tajeddin and Tayebipour (2012) called these two mechanisms as the building blocks of DA and claimed that many academic disciplines have utilized them.
Dynamic Assessment vs. Dynamic Testing
The difference between dynamic assessment and dynamic testing is reflected in Sternberg and Grigorenko’s (2002) distinction. They remarked, “In essence the goal of dynamic assessment is to intervene and to change. The goal of dynamic testing, however, is much more modest – it is to see whether and how the participant will change if an opportunity is provided” (ibid., p.30). Based on Sternberg and Grigorenko (2002), dynamic testing occurs if two components, i.e. assessment and pedagogical intervention, are combined. Therefore, it can be concluded that dynamic testing is providing prefabricated mediation for students to find out how much they will or will not change when offered pre-determined assistance.
Although Sternberg and Grigorenko were determined to highlight the differences between these two terms, “dynamic assessment and dynamic testing should not be thought of as separate enterprises. (Poehner, 2008, p.17)” By the same token, and without considering the differences between these two terms, having chosen Vygotsky’s discussion of microgenesis which deals “with the issue of development occurring very quickly (ibid. p.18)” in his study [the term microgenesis was first made by Wertsch (1985)], the present researchers adopted C-DA to be used throughout this study to refer to those sessions which aim at unifying assessment–instruction as the basis of the DA procedures.
Computerized Dynamic Assessment (C-DA)
In congruence with the concept of DA, the computerized dynamic assessment (C-DA) is grounded in Vygotsky’s theoretical framework (1978). Some studies have been conducted in the field of education on C-DA.
For instance, Tzuriel and Shamir (2002) conducted a study in the area of C-DA and tailored mediation to learners. They attempted to assess kindergarten children’s seriational thinking abilities because they believed that these abilities were central to success in learning mathematics. The prompts have been prefabricated and arranged from implicit (‘try again’) to explicit (providing more relevant information about the item in question). As it is clear, it follows an interventionist approach to DA because the prompts are prefabricated but since teachers are also allowed to take part in the administration of the test actively, i.e. provide supplemental support for learners who fail to answer the questions correctly, just like interactionist DA. The authors stated that more in-depth diagnoses of learner abilities is provided through this procedure when teachers are present in comparison to the time when the mediation is only provided by computer.
Another study conducted on this domain was the one by Pishghadam and Barabadi (2012). Underscoring the increasing importance of DA in second language and reading comprehension through citing some seminal papers on this area, the researchers magnified the shortcomings of DA and paved their own way for introducing their own developed software called CDRT to examine L2 reading comprehension through C-DA. To justify what they have done, they cited some interactionist studies which based on the authors followed a sandwich format, though such a claim cannot be supported based on the seminal work done by Sternberg and Grigorenko (2002) and many other renowned researchers such as Poehner and Lantolf. They claimed that the problem of interactionist studies is that the number of their participants is low and that in sandwich format studies, mediation phase and assessment session are administered separately from each other. In other words, instruction and assessment are not fully integrated in interactionist studies which follow a sandwich format. They also contended that in addition to these shortcomings, interactionist DA does not take the psychometric properties of testing into consideration.
To solve these problems, they used C-DA which is interventionist and follows a cake format. Though not an unbreakable principle, the general consensus is that the interventionist studies tend to follow the sandwich format because of their assessment-instruction-assessment type. But the reason why Pishghadam and Barabadi (2012) claimed C-DA, though being interventionist, follows the cake format is the mediation which the CDRT software provides for learners whenever problems arise during the administration of the assessment. Their study can also be regarded as a study in which sandwich format has been used because in addition to the mediation provided in the pretest for any individual items of the reading comprehension questions, a mediation was also provided to students based on their pretest performance. This mediation was provided for students in a separate way from assessment. That is, while having no assessment session, the students were mediated to be more prepared for the posttest. Three advantages of C-DA were mentioned in Pishghadam and Barabadi (2012, p.79) as well: “1) reliability and validity are taken into account; 2) many students can be assessed dynamically, and 3) mediation is given at the time of assessment not in a separate session.” The test and software construction procedures were detailed too. The test was piloted and by asking a number of ELT experts the content validity of the test was substantiated. In fact, they piloted the test to (ibid. p.84): “1. Make some modifications in the content of test including both the items and hints,” and “Make some modifications in the software package.”
The two most prominent figures of DA, i.e. Poehner and Lantolf, carried out a study on the domain of C-DA in 2013 as well to show its importance of in applying to larger classes. Focusing on the significance of the instructional quality of mediation, they referred to a phenomenon called ‘microgenesis’ which Wertsch (1985) considered it as a process that provided opportunities for development simultaneously even during a single session. While microgenesis which primarily deals with a context in which learners and mediators have a moment-to-moment interaction (Lantolf & Poehner, 2011), their study explores the principles of mediation into a computerized approach to DA. In fact, their study explored microgenesis in the context of C-DA taking L2 Chinese, French, and Russian listening and reading comprehension into consideration. They designed some tests and aimed to differentiate between the learners’ independent and mediated performance, to foresee the difference between their mediated and non-mediated performance (learning potential), and finally to reassure evidence of learning by applying the concept of transcendence into the tests. Similar to Poehner (2005), they have also taken the number of semesters the participants had spent studying in university (here intermediate level because they had studied four semesters) as a way to determine the participants’ proficiency level.
In that study, Poehner and Lantolf took two skills (reading comprehension and listening comprehension) into account for the learners of two languages (Chinese and French) and got the actual and mediated scores for Chinese listening, Chinese reading, and French listening. The number of learners for each one was respectively 62, 80, and 21. They have also calculated the gain score or Learning Potential Score (LPS) for any one of these skills and also the reliability coefficient of the tests for the aforementioned skills.
As it is clear, C-DA has several advantages including being administered simultaneously to large numbers of learners; providing the opportunity for learners to reassess as many times as they would like; and informing the test takers of their performance in the test automatically after they respond the exam. But this does not mean that C-DA is flawless. Though it overcomes some of the shortcomings of other approaches to DA, it faces the same challenge as all other interventionist approaches such as Group Dynamic Assessment (G-DA): we cannot claim and know how learners’ performance would differ if they were provided with other forms of mediation. This study was guided by the following question:
How useful are the scores generated by the computerized dynamic assessment to planning future teaching programs?