Moving from association to mediation: A sociocultural approach to assessment

The processes of learning and teaching (obuchenie)

Within the sociocultural theoretical framework, learning and teaching are viewed as dialectical processes that mutually influence each other. The Russian word obuchenie, which has been mostly translated as either “instruction,” or “teaching-learning process” (Van der Veer and Valsiner, 1991) captures the interdependent relationship that exists between the two-way processes very well (Wertsch and Sohmer, 1995). Many researchers (e.g., Daniels, 2001; Moll, 1990) have noted the complexity involved in translating the term obuchenie into English as there is no directly corresponding one-word that captures the essence of the Russian word. The essence of the term lies in the recognition of the dialectical view of the two phenomena (i.e., teaching and learning) and is in accordance with the general dialectical synthetic views espoused by Vygotsky. Thus, as Vygotsky noted (1978), children not only develop, but are developed (by others).

In order to understand the essential characteristics of obuchenie, one must first analyze a more fundamental relationship that exists between school instruction and cognitive development. The relationship between education and development was a recurrent theme in Vygotsky’s research and writings and is best described in the sixth chapter of Thinking and Speech (1987). In this chapter, Vygotsky outlines three general perspectives in viewing the relationship between formal education and the development of school-aged children. The first view was represented by Piaget and argued that school teaching should follow development as certain psychological functions must be in place in order for children to be able to benefit from formal instruction. The second view, in contrast, maintained that the processes of teaching and learning serve as the main force in leading and guiding the development of children. Taken to its extreme by Thorndike, this perspective led to the assertion that cognitive development is the shadow of teaching.

As Van der Veer and Valsiner (1991) note, these two views were both based on the assumption that cognitive development is a natural process based on biological maturational processes. These two views were mainly distinguished by differences in temporality. The first view emphasized the temporal order of things, maintaining that school instruction should follow cognitive development. The second view emphasized the temporal parallelism, virtually equating education and development.

Finally, there was a third approach taken by Koffka that aimed at reconciling the two former seemingly contradictory viewpoints by taking a middle ground. Koffka acknowledged the existence of two types of development. One corresponded to biological maturational processes and the other was in accordance with school instruction. In this perspective, school instruction formed one type of development but could not be equated with the whole of development as development also presumed biological maturation.

Vygotsky, as his writings in Thinking and Speech evidence, was not satisfied with any of the three aforementioned perspectives describing the relationship between formal education and cognitive development. In Vygotsky’s view, obuchenie and cognitive development were two distinct processes based on complex interdependent relationships with their own developmental curves and dynamics. Based both on his theoretical analyses and empirical investigations, Vygotsky concluded that cognitive development need not precede the formal teaching and learning processes nor should be equated with them. Furthermore, he argued that the dynamic interrelationship between education and development should be more clearly defined than to just assume that the former is one type of the latter.

With this line of reasoning, Vygotsky maintained that there are certain psychological functions that are in place before school instruction begins and that support formal teaching and learning processes. These might be termed the natural or lower psychological functions that follow the path of biological maturation. Formal school instruction, in contrast, brings to life novel psychological functions that could only come into being and develop as a result of formal instruction. These Vygotsky referred to as the cultural or higher psychological (mental) functions (Vygotsky, 1997a). In this context, Vygotsky argued that the only type of effective instruction is one that can point to the road leading to cultural development (i.e., the development of higher psychological functions). In other words, teaching and learning builds on the maturation of certain psychological functions but further leads to the development of new intellectual skills.

In concluding the overview of teaching and learning within a Vygotskian theoretical framework, it should be noted that Vygotsky also thought that the skills and knowledge gained from school instruction (i.e., those that could be referred to as higher psychological functions) could be generalized to other domains to have a general impact on cognitive development. As a simple example, school instruction in mathematics not only allows children to be able to manipulate numbers arithmetically as encountered in school lessons, but also allows them to manage general computations that are required in everyday situations. As will be mentioned later, for Vygotsky, generalizability and transfer were key characteristics of cognitive development (Lantolf and Thorne 2006). Furthermore, as Shilvock (2018) discovered in her research, an important goal of education in the 21st century is to teach students to adapt to any situation they find themselves in. Consequently, one criterion of determining effective school instruction would be to determine whether it leads to generalization and transference.

Assessment within a sociocultural framework

It is important to distinguish between lower (natural) and higher (cultural) psychological functions in discussing assessment within sociocultural theoretical framework not only because Vygotsky argued the necessity for a clear distinction among the two throughout his writings (Vygotsky, 1978, 1994, 1997a, 1997b, 1998) but also because he maintained the latter to be the goal of formal schooling (Vygotsky, 1987).An essential feature of the development of higher psychological functions is that it depends on mediation (Kozulin, 1990, 2003). Mediation is a process by which individuals process external stimuli via means of creating additional artificial stimuli (Van der Veer and Valsiner, 1991). For example, in remembering a list of words read out aloud, one could use picture cards that could represent the words to be remembered. In this way, the words read out aloud (i.e., the object stimuli) are remembered via an additionally created set of picture cards (i.e., the means stimuli). The presence of the means stimuli is what distinguishes higher mental processes from lower ones. Thus, things are not merely remembered (as in direct unmediated memory), but people actively remember things using cultural tools (as in indirect mediated memory).

The artificial means created in this manner are passed down in history from one generation to another. In the process of being passed on from one generation to another, the means themselves are modified and new ones are created to better adapt to the changing circumstances. People also gradually come to internalize the external mediators (Vygotsky, 1997a), such as the picture cards mentioned in the previous example, so that they no longer need to actually have the picture cards present to mediate their remembering. The process by which individuals acquire, modify, and create cultural tools (i.e., mediators) as well as internalize these tools to master their own psychological functions is what constitutes cultural development in a Vygotskian sense. Furthermore, according to Stetsenko (2017), formal schooling constitutes the pathway of cultural development. It is then, through formal schooling, that people become able to function as competent and independent individuals in his or her given society by acquiring the tools made by people of that society and handed down in history (Yaroshevsky, 1989). In this process, Vygotsky also emphasized the importance of human support in his well-known formulation of the zone of proximal development (Vygotsky, 1978). People who are more capable of utilizing the cultural tools help less capable ones in acquiring and internalizing these tools in the context of addressing sociocultural demands (Karpov, 2014). Moll (2014) notes that this human mediation and its relevance for education was of central concern for Vygotsky.

In the context of assessment, the importance of human mediators is pointed out in joint problem-solving and social interaction by Vygotsky inspired scholars such as Rogoff and her colleagues (1995) and Wells (2002). These researchers working within sociocultural theoretical frameworks emphasize the need to consider the sociocultural context of assessment that includes human mediators in addition to what an individual can accomplish in isolation. Van der Veer and Valsiner (1991) also note that Vygotsky thought traditional testing that relied on individual performance alone to be problematic as it tells little or nothing about the stage of children’s cultural development.

It is beyond the scope of this paper to provide an exhaustive overview of Vygotsky’s line of reasoning with regard to the broad topic of the development of higher psychological functions and the meaning of cultural development. For the purpose of present arguments focusing on the topic of assessment, suffice it to say that Vygotsky was generally opposed to the traditional practices of testing that aimed at measuring the development of lower psychological functions (Van der Veer and Valsiner, 1991). These testing practices measuring individual and isolated performances do not provide relevant information regarding the stage of cultural development (i.e., the extent of development of higher psychological functions). Given that the only type of instruction is one that promotes cultural development and that the aim of assessment is to measure the outcomes of instruction, it becomes evident that the only effective type of assessment is one that aims at measuring cultural development.

Two interrelated themes must be addressed to capture cultural development in assessment practices. The first relates to the mediated nature of cultural development, and the second emphasizes the importance of social interaction. These two themes are interrelated because social interaction among people is a specific form of mediation. Kozulin (1990, 2003) provides a general classification of mediators, which include symbolic (e.g., literacy, numeracy, graphs, maps, and other human created symbol systems) and human mediators.

The nature of mediation, especially in the form of human interaction, will become clearer when viewed against the background of current testing practices that do not incorporate mediation and social interaction. Therefore, the next section will briefly survey testing practices that are not aligned with the tenets of the sociocultural theory, in order to return to a more vivid picture of an assessment paradigm that includes cultural development as its core principle.

Unmediated assessment practices: The case of North Carolina End-of-Grade tests

The phenomenon of high-stakes standardized testing is having an increasing impact on the process of education throughout the nation as accountability is rising to the topmost concern (Tuck, 2013) in federal regulations and policies (e.g., The No Child Left Behind Act of 2001). As accountability, assessment, and testing seem to be linked to such an extent that they are at times used synonymously, it would be helpful to clarify these concepts at this point.

There are many ways to hold people accountable and assessment seems to be one approach among many. Similarly, there are diverse ways to assess individuals and testing is one among the various methods. Tests need not always be standardized or high-stakes. As Hargrove (1999) notes, tests become high-stakes due to high-stakes accountability that hold the threat of implementing various sanctions depending on the results of the tests. The same author also mentions that tests are standardized due to the factors of cost and scoring. It should also be added that standardization allows for comparison among individual students as well as individual schools. Comparisons based on normative data serve to rank order students and schools, which may also have strong consequences (e.g., promotion and demotion at the student-level and differential financial support at the school-level).

To give a more concrete and detailed description of high-stakes standardized testing currently in use, an example from an end-of-grade achievement test used in North Carolina will be used. The choice of the example is based on the authors’ extensive experience in working with these tests. In addition, the author had abundant opportunities to review the typical test contents and format while assisting students who had to take these tests at the end of the school year.

Beginning in third grade (usually 8-9 years old), students in North Carolina are required to take the North Carolina End-of-Grade Mathematics Tests and North Carolina End-of-Grade–Reading Comprehension Tests. The tests are high-stakes for both the school and individual students as the results of the tests have a powerful impact on subsequent decisions that may lead to recognition and promotion or retention and demotion. Consequently, there is much emphasis within schools and among individual teachers to maximize their students’ performance on these tests.

As with many other standardized tests, the main response format is presented in multiple choices with four answer options. Test results given in percentages allow an ease of comparison among various groups of students (e.g., by ethnicity/race and gender) including their status of promotion of demotion (refer to http://www.ncpublicschools.org for more detailed information). Percentile ranks also allow for statewide comparisons among various schools. Test results strongly contribute to how individual schools are designated in various public reports (e.g., schools of excellence or schools of distinction). In summary, the characteristics of End-of-Grade (EOG) tests present many attractive features such as ease of scoring and administration as well as ease of comparison and ranking across individual students and groups.

There is nothing in high-stakes standardized tests that is inherently incompatible with socioculturally espoused views on assessment. In determining whether this type of tests is socioculturally oriented or not, the content and format of the tests should be examined. Therefore, the central issue should be about what (i.e., content) is tested and how (i.e., format) it is tested regardless of whether the consequences are high or whether the tests are considered to be standardized or not.

Over the multiple years of working with third grade students in getting them ready for the EOG, the key characteristic derived from observing the interactions between the math practice items and the students was that most, if not all, problems could be solved by identifying key words in the question. Even English as a Second Language (ESL) students, who were not able to comprehend the complete passage contained in the problems, could find the correct answers by identifying the words that cued certain mathematical operations. For example, words such as in all or altogether were associated with additions and words like how many more, how much longer, and change immediately triggered the use of subtraction. Other educational researchers (Bruer, 1993; Wood et al., 1995) have similarly noted that students taking these types of tests look for key words and superficial syntactic cues that reveal which operations (e.g., addition or subtraction) to use.

Furthermore, even if these students could not actually carry out the mathematical operations, the answer choices significantly reduced the burden of identifying the correct answer by the process of elimination as all the answers were presented in multiple choice formats. This held true for problems in which calculator use was allowed as well as for problems that prohibited the use of calculators. Some answers were so recurrent that the students only had to look at the four answer choices to pick out the correct one. This was typical of the questions related to probability, namely, whenever the answer choices contained a number and the word ways (i.e., 3 ways, 4 ways, 5 ways, 6 ways), the answer was always 6 ways. All of the probability questions related to the possible ways of arranging three different things. As a consequence, once the students found out the answer and its recurrent appearance in different guises (i.e., different objects, names, etc), they stopped bothering to even read the question.

A similar phenomenon was observed within the field of reading comprehension. The general format was a reading passage followed by a set of comprehension questions. In order to deal with this type of problem format, the students were guided to read the questions first and then read the passage to identify relevant pieces of information. This direction came from helpful strategies contained within test prep books as well as from their classroom teachers and other tutors involved in helping them prepare for the test.

With the exception of a few questions that required the identification of the main idea of the reading passage, most test items could be answered by locating a single sentence or phrase within the passage. Even with the “main idea” questions, if one were to pick the answer choice with the broadest idea, it was fairly safe to conclude that the correct option was chosen.

Given these contents and formats of the End-of-Grade tests of mathematics and reading comprehension, it is not surprising that even students who were receiving passing scores on these tests were not able to generalize the skills needed to solve the problems presented in the tests to other situations or real-life tasks that required similar intellectual functioning. Within the domain of math, these students were still struggling when there were slight word changes or modifications and extensions to the operations and concepts they had already used to answer questions on the EOG pretest or prep tests. Knowing that 32°F is the freezing point failed to generalize when asked to choose the correct temperature most appropriate for a coat and a sweater (among the four options that included 32°F and other choices that were much warmer for winter clothing). Furthermore, being overly acquainted with the multiple choice format, these students were at a loss what to do when a comparable problem was presented to them in an open-ended manner.

Similarly in reading, asking students to choose their own story and then retell it to the teacher or peers using their own words posed a great challenge. First, they could not choose a story, and when they finally did and read it thoroughly, they failed to retain the content. In the absence of comprehension questions, the students were at a loss in deciding which pieces of information should be retained. This was a consequence of reading materials that were chosen by others instead of basing the selections according to students’ purposes of reading and answering questions other people thought were worthwhile (Gifford, 1992).

When preparing for the EOG tests, the main guidance in the reading process was the comprehension questions that appeared after the texts but were instructed to be read in advance. Consequently, when comprehension questions were not provided in advance, students could not figure out what to do with the reading passages. For example, when students were prompted to write a short essay based on what they had read, their main focus was on counting how many words they had written down, rather than on the overall idea presented in their work. The format of the test had made the children captives of superficial comprehension instead of allowing for deep reflection (Shepard, 1992).

As Wells (2002) noted in the context of describing classroom problem-solving situations, most problems in real life do not come neatly formulated and at times the bigger challenge is to discover and define the problem that needs to be solved. In addition, real problems rarely have a single correct answer, nor do answers come in options of four to choose from. Furthermore, as Rogoff and her colleagues also pointed out (1995), most of the problems are seldom solved by individuals in isolation. Rather, social interaction forms an important mechanism in the problem-solving process (Brown et al., 1992).

In contrast to the demands of real life, the problems in the EOG tests are defined and neatly formulated. The problems have a one-to-one correspondence with a single right option that also comes preformulated in a multiple choice format. Most importantly, there is no room for social interaction as individual performance is what is being tested. As will be discussed in more detail in the following section, these characteristics of currently used achievement tests do not aim at the higher psychological functions that result from cultural development via mediation, especially processes of social interaction. Rather, these tests mostly focus on measuring lower psychological functions that do not generalize to other domains of intellectual functioning and are based mostly on processes of association.

Tests based on association and mediation

In an answer to the question of what should be the goals of education for the 21st century, Wells and Claxton (2002) note that in societies that are undergoing rapid changes, young people need to be equipped with skills and knowledge to thrive in and accommodate to such changing circumstances. In this context, they also define education as a process of simultaneous enculturation and transformation. People not only need to acquire the existing cultural tools but also need to have the ability to modify them for various situations. These ideas are based on the sociocultural theories of Vygotsky who emphasized throughout his pedagogy related writings that the aim of education is to drive cultural development via the mechanism of mediation, which result in generalizable knowledge and skills.

Given the general aim of education is acquiring generalizable knowledge and skills (Shilvock, 2018)(, the objectives of instruction and assessment should be aligned with this fundamental principle. Based on this general objective, two assessment practices based on testing could be outlined depending on whether the tests target lower or higher level cognitive functioning (i.e., cognitive functioning that is generalizable). For ease of reference, the former type will be labeled as tests based on association and the latter will be termed as tests based on mediation. However, it should be noted that this categorization may simplify the existing phenomena as there are many forms of tests that defy strict binary categorization, reflecting the characteristics of both types of testing.

Although strict categorization may be impossible, it is instructive to explore the general characteristics of these two types of tests. Tests of association, by definition, test associative skills, which may be represented in the form of associating test items (i.e., stimulus) with the correct option among given choices (i.e., response). Vygotsky was critical of this type of assessment as it led to measuring lower psychological processes (Van der Veer and Valsiner, 1991). The lower psychological processes do not make use of cultural instruments (i.e., are not mediated) and do not generalize to domains requiring similar cognitive functions. Thus, these types of tests do not assess development, in the strictest sense of the word, as development would require mediation and generalization. It was pointed out that Vygotsky had argued that the only effective type of instruction is one that leads development. If the aim of assessment is to measure the outcomes of effective instruction, the tests based on association do not fulfill this purpose.

Contrary to the practices of tests based on association, tests based on mediation aim at measuring higher psychological processes that have been formed via the mechanism of mediation. Mediation or the internalization of cultural tools is basically a social process that is based on interaction with other human beings. Cultural tools are first used between individuals on the interpsychological plane and only later move on to the intrapsychological plane to serve individual cognitive functioning. This is why Vygotsky (1981) consistently emphasized that even the most private spheres of human consciousness retain the social nature found in concrete interactions. The internalized higher psychological functions, as has been previously emphasized, are characterized by their ability to generalize to other domains that require similar intellectual skills.

In an effort to concretely depict tests based on association, the End-of-Grade Tests used in North Carolina was presented as an exemplar of currently used achievement tests that rely on assessing the development of lower psychological functions. This specific test, however, is but one example of the various high-stakes standardized achievement tests used throughout the nation. As federal policies dictate stronger accountability and pressure for educational reforms increases, a resort to various types of standardized achievement tests are becoming a more frequent phenomenon with the stakes of their results rising ever more rapidly.

As noted before, there is nothing inherently wrong with the progression from stronger accountability to educational reforms to emphasis on high-stakes testing. Everyone involved in the process of education indeed need to be held accountable for the results of instruction, and instruction needs to be improved on a daily basis in order to meet the challenges of accountability. Furthermore, assessment serves as one of the essential ingredients of accountability. What is problematic is the nature of the tests used to assess the results of learning and teaching and holding people accountable for results of these tests when they are not aligned with the goals of instruction. This is precisely why there needs to be a critical evaluation of the content and format of existing tests.

In the following section, a brief outline of tests based on mediation will be presented. In order to move from tests based on association to tests based on mediation, there has to be a shift from overly-positivistic and associationistic views to sociocultural and constructivist perspectives on the nature of human functioning. Only by achieving this change in frameworks, will there be an alignment of goals of instruction with that of assessment and individuals as well as schools will be held accountable for the purpose of formal education, which is none other than the development of higher psychological functions (i.e., cultural competence) for all students.

Toward a socioculturally inspired assessment paradigm

If the objective of assessment is the results of formal instruction, it follows logically that assessment should be aligned with the principles underlying instruction. According to Vygotsky’s theory, the only effective type of instruction is one that leads to the development of higher psychological functions. This type of development occurs via mediation, which involves the use of various cultural tools that are initially acquired in social interactions. From these postulates, it was concluded that the only effective type of assessment is one that is based on assessing higher psychological functions, or cultural competence. Accordingly, it was argued that tests based on mediation have the theoretical potential to fulfill the roles of effective assessment.

The remaining question to be answered relates to the nature of tests based on mediation and to explore why these forms of tests have not been used as a major source of assessment to date. As the first part of the question seems more complex and in need of an elaborated description, the second part will be addressed first.

Many researchers working within the sociocultural tradition (e.g., Rogoff et al., 1995; Van der Veer and Valsiner, 1991) have been aware of Vygotksky’s skepticism toward the long Western tradition of relying on positivistic thinking. This intellectual tradition leads to viewing knowledge as a static product, as opposed to a dynamic, interactive, and constructive process. Accordingly, tests grounded in this tradition aim at measuring discrete pieces of knowledge that the individual must prove he or she possesses apart from any social interaction. What the individual knows on his/her own, and what he or she can do by him/herself is what is valued and thus become the objective of testing. Furthermore, the results of testing must be replicable in standard conditions - thus the term “standardized testing”. All these characteristics of positivism have led to favoring tests that measure discrete skills and knowledge, individual performance, and replicable and quantifiable results in the form of test scores.

As Lidz and Gindis (2003) point out, Vygotsky, despite his criticism towards positivistically oriented standardized testing, never argued for a total abandonment of these types of testing. As these authors note, Vygotsky’s main objection to the aforementioned type of testing was that they were incapable of distinguishing lower psychological processes from the higher ones. Accordingly, if the aim of testing is to gain evidence of the development of lower psychological functions, these types of tests would have appropriate validity. However, as has been argued repeatedly throughout the present discussion, the aim of education is to drive development of higher psychological functions, and educational assessment, therefore, must be based on similar grounds.

The reliance on positivistic tradition may have left indelible traces on assessment practices even as processes of instruction are moving toward socioculturally inspired practices. In this sense, assessment may be said to be lagging behind instruction. Therefore, this might create discrepant situations where pedagogy is informed by sociocultural theories (Daniels, 2001) and assessment is still heavily dependent on positivistic views.

Returning to the first question posed at the beginning of this section, the focus is now on how the higher psychological functions might be tested. Kozulin (1998) provides a beginning to this answer by proposing the importance of mediation of meaning in the development of higher psychological processes. Mathematical principles and literacy skills that are assessed should be viewed as symbolic cultural tools that assist in the mediation of meanings contained in problem-solving situations. Furthermore, as Wells (2002) noted, the challenge in real-life often comes in defining the problem situation itself. The mathematical tools provided by the culture via formal instruction allow individuals to formulate the problem itself.

One implication of these considerations is that in assessment contexts, students should be presented with various scenarios that reflect real-life situations that are in accordance with their sociocultural environments and to have them identify potential problems and their solutions. This type of assessment acknowledges the real challenge of identifying what problems exist and that there could be multiple approaches to solving the problem at hand. Sociocultural approaches to human development acknowledge that people could use various cultural tools to attain the same level and functioning of cultural development. Therefore, multiple perspectives on the problem situation as well as their solutions should be encouraged. As sociocultural theories view knowledge as a constructive process that is formed via means of cultural tools (e.g., mathematical operations learned during formal instruction), process of knowledge construction should be assessed in addition to the product (e.g., a single solution that could be randomly guessed from multiple choice options).

Instead of mechanically associating addition skills to certain words contained within the preformulated questions (e.g., how many in all), students should be encouraged to explain to each other or in writing what type of a problem is contained within the scenario that requires the mathematical operation of addition. Students need to be able to understand the real meaning of mathematical operations and this would be evidenced by their ability to generalize mathematical reasoning to domains requiring similar intellectual functioning. Various real-life based scenarios that demand similar intellectual functioning could tap into students’ potential to generalize and transfer. To give but a couple of examples, these scenarios could include math involved in: shopping at a grocery store (Lave et al., 1984); and selling girl scout cookies (Rogoff, 1990). These are activities children are likely to engage in their everyday lives with a more competent person (e.g., parent or older sibling).

As Karpov (2005) and Rogoff (2003) argue, rote skills that are nontransferable and inert verbal knowledge should be replaced with meaningful conceptualizations and understanding. These test formats allow such transformation to occur. In contrast to standardized testing practices that aim for unmediated processes, mediated assessment practices would not allow slight word changes or modifications in mathematical problems to have any impact on how students approach the problem situation. This is because meaningful conceptualizations attained via cultural tools (e.g., representations of real life scenarios based on social interaction) lead to generalization and transference.

Similarly, reading for meaning must be communicated as the objective of testing in the field of reading comprehension. In real life, children rarely, if ever, read for the sake of answering a set of predetermined comprehension questions. Furthermore, their ability to comprehend is never tested in this manner. Therefore, first and foremost, students should be asked to choose reading materials in accordance with their purposes, interests, desires, and needs. The comprehension questions should also be based on the aforementioned factors. The students could be asked whether their reading fulfilled their purposes, and if not, what might have caused the discrepancy between their intent and the content of the reading. They could also be asked what sources would best complement the reading material they have completed and where they could be found. From a Vygotskian perspective, these are the types of activities that lead to the acquisition of cultural competencies, and consequently to the development of higher psychological functions.

Students should also be engaged in assessment activities that allow them to compare and contrast various types of genres. Children could pair up with a teacher or other students to engage in these types of discussions. As sensitivity to rhetorical devices and stylistic differences is essential in functioning as a culturally competent individual in a literate society, attention to these literary elements should be highlighted by teachers.

Assessment activities could also demand that students write in similar styles to the reading materials offered. This type of practice goes far beyond the concept of reading as “perusing short passages to answer other people’s questions” (Resnick and Resnick, 1992: 46). The results of this kind of writing would also be a telling indication of writers’ comprehension of the structure and organization of the materials read. This is because, as Vygotsky (1987) noted, one cannot imitate what one has not comprehended (Chaiklin, 2003). Therefore, the ability to write in a similar style to a material that one has read indicates that one has comprehended the stylistic and organizational features.

The problems with mediated assessments could be solved individually or in a group setting. Knowledge is ultimately co-constructed and it is informative to assess its acquisition within the context of acquisition. Furthermore, assessing the ability to gain support from others is also a valuable piece of information (Bodrova and Leong, 2007). However, the aims of mediated assessment could be the internalized state of conceptual understanding and in this case individual functioning could become the focus. Although assessed individually, it is important to keep in mind that human learning and development are essentially social (i.e., based on social interaction) and only later become internalized.

Conclusions

As Daniels (2001) notes, pedagogical implications deriving from sociocultural theories have been increasingly implemented into classroom instruction. Accordingly, there is a similar move to change assessment practices to better reflect the outcomes of instruction. The present discussion attempted to build on this foundation to argue that if the goal of education is effective instruction, and if effective instruction depends on the development of higher psychological functions (i.e., cultural competence), then to assess the outcomes of effective instruction, it is only logical that assessment practices must aim for those higher psychological processes that may only be developed and captured through mediational means.

Attempts at assessment reforms and introduction and utilization of Vygotskian ideas within the assessment field is by no means a new phenomenon. Dynamic assessment (Campione and Brown, 1985; Poehner, 2008), and other forms of assisted assessment have been around for decades. What is unique about the present paper is that it emphasizes the need for assessment based on mediation based on the critical distinction Vygotsky (1978, 1994, 1997a, 1997b, 1998) made between lower and higher psychological (mental) functions in his theoretical framework and the need for formal schooling to focus on the latter as it is the latter functions that would allow an individual to become a competent member of his/her society.

Viewed through the lens of the aforementioned theoretical framework, tests that are mainly based on association fail to capture those higher psychological functions developed by instruction. This is because these functions are developed through mediation, which actively use and internalize various cultural tools through social interaction, and have the capacity to be generalized to other domains requiring similar intellectual skills. Tests based on mediation, on the other hand, reflect the mediational and social interactional character of human functioning by incorporating various real-life scenarios and emphasizing the co-constructive nature of meaning. Furthermore, the knowledge and skills assessed are generalizable because the nature of the tests reflects meaningful conceptualizations rather than rote skills and discrete pieces of information.

A move toward socioculturally inspired assessment requires a drastic change in thinking, which is not easily achieved. In addition, the attractive features of standardized multiple-choice tests (e.g., low cost and ease of administration and scoring) are hard to turn away. However, as Kuhn (1996) would argue, and Vygotsky would agree, moving toward a new paradigm does not necessarily imply a complete abandonment of the old. There are times when recognition and recall of discrete pieces of knowledge may need to be tested. These unmediated processes are well suited for tests of association.

In a rapidly changing world, competent individuals must not only be equipped with the necessary knowledge and skills, but they must also be able to apply them conceptually and meaningfully in diverse situations. Formal education should be held accountable for moving students in this direction as well. Moving from tests of association to tests of mediation is a contribution the assessment field can make in this context.