Open Access

Dynamic assessment of writing ability in transcendence tasks based on Vygotskian perspective

Asian-Pacific Journal of Second and Foreign Language Education20172:10

https://doi.org/10.1186/s40862-017-0033-z

Received: 4 April 2017

Accepted: 29 August 2017

Published: 5 September 2017

Abstract

This study aimed to investigate the dynamic assessment of writing proficiency in Iranian English learnersˊ writing ability in transcendence tasks based on Vygotskian perspective. The questions which this research intended to answer were: 1) Does dynamic assessment affect Iranian English learners’ writing ability? 2) Do the results of dynamic Assessment differ from the results of a non – dynamic assessment? 3) If learners progress, are they able to maintain their improved performance in the transcendence tasks when the assessment context changes? To answer the questions, 60 learners were selected following an Oxford Placement Test, and then randomly divided into two groups, experimental and control. The experimental group received dynamic assessment based instruction while control group received a non-dynamic writing instruction. After eight sessions, both groups were post tested and underwent a static and a dynamic assessment. The results indicated the outperformance of the experimental group over the control one. After two weeks, the experimental group undertook two transcendence tasks in order to trace their growing proficiency in more difficult and to increase innovative tasks. Moreover, upon analyzing the protocols, an inventory of meditational moves was detected as a posterior. As the result, the study suggests the use of dynamic assessment as a development-oriented procedure to diagnose and develop the learnersˊ emerging abilities.

Keywords

Dynamic assessmentMediationWriting proficiencyTranscendence

Introduction

This present research concentrates on the positive impact of dynamic assessment of L2 writing proficiency with reference to Vygotsky. L.S. Vygotsky was an early twentieth – century Russian psychologist who had a significant influence on the development of social theory. According to Vygotsky (1978), in order to understand the human mind, we need to understand the processes from which it emerges or develops. He tried to develop a theory of social, cultural, and historical formation of the human mind and emphasized the social nature of human cognition. Socio – cultural theory is linked to the work of L.S Vygotsky who attempted to provide an account of learning and development as mediated processes. Recently testing researchers have begun to express concern over the power of tests in our lives. For example, Messick (1988), states that more attention should be paid to the social consequences of introducing a test into an existing instructional setting and accepting the resulting scores as the only indicator of learners’ abilities. The impact and influence that tests have on instruction and learning cannot be denied. This is generally referred to as wash back which manifests itself in testing situations where getting high test scores come to be goal of education (Bailey, 1996). Dynamic assessment is a subset of interactive assessment that includes deliberate guided or meditational teaching. The roots of dynamic assessment are traced back to Vygotsky and Feuerstein. Dynamic assessment is based on the point that there are many obstacles and factors that can mask one’s ability. It implies that every individual performs or functions at less than 100% of capacity (Tzuriel, 2000). In DA the focus is not on the success or failure of examinees at completing a given task. Instead, the focus is on the analysis of the amount and kinds of assistance they needed. From Vygotsky’s perspective, analysis of examiner – examinee collaborations reveals the future performances of examinees if they are given appropriate instructions (Vygotsky, 1988). According to Vygotsky (1988) learning does not exist in isolation i.e. he believed in a non – personal view of knowledge and learning. Writing has always been present in applied linguistics and it is an important skill because it provides a way of monitoring EFL learners’ language production; it is a source of stable data analysis which shows how language was learned. According to Stanley (1993), although the importance of writing has been recognized in applied linguistics, generally, it remains one of the least understood, if not misunderstood, subjects in applied linguistics. One reason is the ambiguity of the term “writing which has been used in referring to orthography, written discourse, and the act of writing in linguistic sciences. Writing continues to be marginalized in SLA research. Writing is currently considered as a dynamic, creative and contextualized process of communicating meaning. Writing is not decontextualized process; it is situated in the social and cultural context in which it is produced. Writing involves a dynamic interaction among the text, the writer and the reader. So writers need to consider these three elements and write accordingly. The need to teach this ability is crucial because writing is not only a way of discourse manifestation but a way of manifesting the linguistic, pragmatic, intercultural and strategic competence also. In doing so, teachers will raise learners’ awareness of all these elements in the communicative act of writing and as a result, will encourage them to communicate through writing.

This research investigates to discover how dynamic assessment techniques influence L2 learners’ writing at paragraph level, since the participants were intermediate learners, and according to ACTFL Proficiency Guidelines (2012), writers at the intermediate level are characterized by the ability to meet practical writing needs, such as simple paragraphs, messages and letters, requests for information, and notes.

The DA approach that was developed for this study follows Feuerstein’s preference for flexible interaction between the mediator and the learner as the two cooperatively perform the assessment tasks. The researcher decided to start this study when she was teaching English to some Iranian university students. The researcher found that these students had always the tendency to write what they wanted to express in Farsi before turning it to English (L2). So they usually resorted to the direct translation method. The researcher observed that whenever the students did this, their answers were heavily influenced by their mother tongue grammatically. From the researcher’s point of view, they were not good at converting the sentences into L2. In this relation, research questions of this study are as follow:
  1. Q1.

    Does Dynamic Assessment affect Iranian EFL learners’ writing ability?

     
  2. Q2.

    Do the results of DA differ from the results of a non – dynamic assessment?

     
  3. Q3.

    If learners progress, are they able to maintain their improved performance in the transcendence tasks when the assessment context changes?

     
To answer these questions the following assumptions were formulated:
  1. H1.

    Dynamic Assessment does not affect Iranian EFL learners’ writing ability.

     
  2. H2.

    The results of DA do not differ from the results of a non – dynamic assessment.

     
  3. H3.

    If learners progress, they will not be able to maintain their improved performance in the transcendence tasks when the assessment context changes.

     

Methods

This study design was based on an interventionist DA approach (the Sandwich format). A pretest-enrichment-posttest-transfer session format was followed in this study. A pretest was conducted to diagnose the students’ independent performance abilities and their main sources of difficulties (syntactic and lexical). To address the learners’ recurring problems an enrichment program lasting for eight weeks was offered. Then a posttest was administered followed by two transfer/transcendence (TR) sessions aimed at understanding the extent to which the students could extrapolate their newly acquired knowledge to novel contexts. The experimental group received the same pre-posttests at two levels: Zone of Actual Development assessment (ZAD) and Zone of Proximal Development (ZPD) assessment. Only the experimental group received TR tasks (not the control group). An experimental research design was performed. The research design is summarized as follows (See Table 1 and Fig. 1).
Table 1

Research design for each group

Experimental group

Control group

Period

Pretest (ZPD, ZAD assessments)

Pretest

1 week

DA based instruction

Non-dynamic instruction

8 weeks

Posttest (ZPD, ZAD assessments)

Posttest

1 week

Transcendence 1

 

1 week

Transcendence 2

 

1 week

Fig. 1

Experimental research design

Participants

The participants of this study were 60 homogenous EFL learners (i.e. they had the same L2 proficiency level) who were studying at University of Applied Science and Technology, Tehran, Iran. The participants of the study were male and females, native speakers of Persian and in their early twenties and thirties. Also, they had not been to English speaking countries. Their average exposure to English was about 6 years during which they had received a traditional version of language learning syllabus and curriculum. Based on a placement test and an interview, they were considered intermediate learners (Their OPT scores were one standard deviation above the mean). Since they were studying English in Iran only, they were described as foreign language learners. Most of them had received their second language (L2) knowledge (in English language) at school; so, they were exposed to non-dynamic instruction of L2 learning. As a result, they had received a grammar-based, teacher-oriented method based on the course in the classroom (See Table 2).
Table 2

Demographic information about the subjects (N = numbers)

Learners

N of classes

Ex group

Cntr group

Total

18–35 years old

1

30

 

30

18–35 years old

1

 

30

30

Total

2

30

30

60

An OPT (Oxford Placement Test) was administered to all the participants (N = 87). With their OPT results, the learners were divided into an experimental and a control group.

Research materials

In this study four types of tests (pretest i.e. DA1 and first static assessment/SA1, posttest i.e. DA2 and second static assessment/SA2, TR1, TR2) were used. The tests consist of a general subject about which the participants had to write about at paragraph level. Also, IELTS standard tests were used to assure the validity and reliability of them. One kind of pretest was used (from IELTS writing samples) for each group. So all participants underwent a static and a dynamic assessment at the beginning and at the end of the enrichment program. In addition, the enrichment learners also completed two transfer assessments. After conducting the 8th instruction session, the posttest was given to 60 students to measure the learners’ achievements. The students’ writings helped the students to showcase a variety of abilities which required remediation. As already stated, grammar and pragmatic use of vocabularies proved problematic for all the participants and therefore the enrichment focused mainly on grammar and usage of different words in different situations. The transfer tasks were developed to assess the extent to which learners had internalization. They assessed how far the learners had extended the mediation provided. Together with the results of the posttests, this information tracks the gains made. In the present study, two transfer tasks TR1 and TR2 were used to round out the development diagnostic of each participant at the end of the enrichment program. The SAs and DAs were centered on learners’ narrative which they wrote. TR1 paralleled the SAs and the DAs in that it too involved a general subject to write about. However the genre was different, so the emotional response and the attention the subject demanded were different i.e. the students had to write about a topic was not as concrete as the initial tests. TR2 differed from the other assessments in an even more important way: the medium of prompt itself. The prompts were given only in English language and the students were allowed to use English to English language (not English to Persian or Persian to English dictionaries). So the two TRs differed from the initial tests regarding the topic factor (topics became more abstract) and the prompt factor (they were given implicit feedback mostly). A final point regarding the transfer assessment has to do with the enrichment and non-enrichment learners. The transfer tasks were designed to determine how well the enrichment learners could extend or transfer the abilities they had developed through their interactions with the mediator to novel problems. It implies that non-enrichment learners would not show substantial change over time so it did not seem likely that they had new abilities to transfer.

Procedures

The procedure of the study is based on the principles of interactional DA. Regarding the mediator’s (teacher) way of offering the mediation, the strategic mediation was strongly linked to the learners’ needs i.e. the mediator adjusted her assistance based on the specific response of the learners. The DA procedure took the following steps:
  1. 1.

    At the onset of each assessment session, the students were given a general concrete topic to write about (at paragraph level). The students were told to use any kind of dictionaries if necessary. Moreover, they could ask for help or support whenever they face difficulties in their writing process.

     
  2. 2.

    Upon the students’ failure to express their ideas, the mediator intervened and offered her leading questions, prompts, hints and explanations in order to uncover the students’ writing potential.

     
  3. 3.

    The meditational strategies were detected after the analysis of the teachers’ interactions with the students. The teacher used them during the mediated assessment sessions. The frequency and types of meditational moves (implicit/explicit) offered by the teacher during the assessment sessions indicated the students’ improved abilities and ZPD/ZAD functioning. The assessment procedure adopted in this study coincides with those of Poehner (2005). The teacher analyzed the learners’ performances qualitatively on both process and product. Through dynamic assessment the teacher acquired the ZPD as well as the qualitative, quantitative diagnostic information for each learner. So, in order to interpret the data, the first step was to transcribe the audiotapes. The second step was to read the protocols to obtain an overall impression of the meaning the students tried to express.

     

The mediator responded to problems as they occurred and in this way the mediator helped the learners to perform the task at a higher level than were capable of on their own. So the mediator, in this study, offered different form of help and assistance throughout the assessment. Also, the learners were free to ask for help as needed. The mediation itself was based on the principles of the mediation emerged out of the cooperative dialoguing between the mediator and the learners. Therefore there were no priorities or hierarchies of hints or prompts.

During the sessions in which mediation was provided (i.e. the DAs and the transfer assessments), the mediator would interrupt at different points to ask questions, offer suggestions and provide help when necessary. Sometimes the mediator attempted to provide a correction to question something that was said or to make general comments. The writing topics that were used for the compositions came from IELTS sample writings. In the following examples, the dynamic procedures revealed that the learners did have different levels of control over different structures. Some students were able to improve their performance after a simple one-time reminder but some of them needed a very explicit from of help or assistance (i.e. the choice between two alternatives). While checking the students’ writings, the mediator first mentioned that something was wrong with a given sentence e.g. "something is wrong in this sentence. Guess what." If the student could not spot the problem or if the examiner could not elicit an appropriate response from him/her, she would provide a more explicit form of mediation and so on until the learner was able to spot the problem and make corrections. Eventually, if necessary, the examiner would explicitly correct the error with the needed explanation to make sure that the learner comprehended.

The learners’ performances during SA1 (pretest) and SA2 (posttest) were examined for changes in their independent functioning or Zone of Actual Development (ZAD). DA1 (pretest) and DA2 (posttest) were compared to determine if there were changes in the amounts and kinds of mediation required at these two points in time as well as how learners responded to the mediator’s moves. This indicated their ZPD at times 1 and 2. Finally, their performance in DA2 (posttest) was compared with TR1 and TR2 in order to assess how well the learners were able to maintain their level of functioning as changes were introduced to the assessment context. Participantsˊ performances in the control group were analyzed in the same way but through a two-way comparison since they did not take part in the transfer assessments. The learner-mediator interactions/explanations were transcribed for analysis. The transcripts of the enrichment lessons were examined and they demonstrated signs of struggle and development (See Appendix 1).

Data collection and analysis

Students’ writings were collected by three different raters (to calculate inter-rater reliability) who were told not to count errors of capitalization, errors of lexical choice (e.g. kids vs. children) unless they impeded meaning, spelling errors and punctuation error.

In order to answer the first question of this study, a paired-sample t-test was applied between the pretests and posttests of the two groups.

To answer the second question of the study, an independent samples t-test was applied between the posttests only. In an attempt to answer the third and the fourth questions of the study, the recorded transcripts were analyzed and the protocols (i.e. examples) were used as illustrations. Micro genetic methodological procedure was conducted to analyze the developmental changes and descriptive analysis was conducted to identify the frequency of the emerging interaction pattern. Also, inter-rater reliability and agreement was calculated and checked for each test (pretests, posttests, TR1, TR2).

Results

The results reported here take into account the DA interactions collected during the pretest, posttest, and TR sessions. The protocols have been mainly drawn from the mediated portions of the assessment sessions that involved interaction and assistance. As it was noted earlier in the previous sections, the present study aimed at investigating the impact of DA-based versus non DA-based instruction and assessment on Iranian EFL learners’ L2 writing skill in transcendence tasks. So this section is concerned with data analysis in which the following terms such as group statistic, Paired Samples t-test and Independent-Samples t-test have been used. The main objective of this study is to examine the effectiveness of Dynamic Assessment on L2 writing proficiency in transcendence. The obtained data of this study were analyzed by utilizing SPSS (version 21) software. To reject or accept the aforementioned research hypothesis, the following procedures were taken into account and the obtained data were analyzed by T-Test (paired Samples t-test and Independent-Samples).

Inferential/quantitative statistics

In this section, the researcher presents the results and findings of t-test which confirms the positive effect of DA instruction on the experimental group. As mentioned before, OPT was administered to make sure that the participants were homogenous in terms of their language proficiency. The subjects’ scores were between 33 to 41. Thus the entire participants (N = 60) of this study were at the intermediate level (their scores on OPT were one standard deviation above the mean). The researcher assigned the homogenized subjects to two groups: the experimental group and the control group. The subjects participated in 8 sessions of instruction during research period. Then, in order to measure the participants’ writing proficiency at paragraph level, writing topics of IELTS were used as a pre-posttest for the two groups. In order to measure the reliability of the students’ scores, inter-rater reliability was calculated. In this study, the researcher intended to answer the following questions:
  1. Q1.

    Does Dynamic Assessment affect Iranian EFL learners’ writing ability?

     
  2. Q2.

    Do the results of DA differ from the results of a non-dynamic assessment?

     
  3. Q3.

    If learners progress, are they able to maintain their improved performance in the transcendence tasks when the assessment context changes?

     
Thus to compare the performance of the experimental group and the control group on pre-posttests and the two TRs, their mean scores were subjected to Matched t-test (paired Samples Statistics). The analyzed data for the two groups were shown in the form of tables Table 3.
Table 3

Checklist of assessing writing

To be counted

Not to be counted

Grammaticality

Errors of capitalization, spelling and punctuation

Lexical Appropriateness

 

Content

 

Inter-rater reliability

The students’ writing compositions were analyzed for fluency, accuracy, grammaticality lexical items and content. This was done by using inter-rater reliability i.e. students’ writing papers were collected by three different raters who were asked not to count errors of capitalization and word choice (e.g. buy vs. purchase) unless they impeded meaning. So, inter-rater agreement was checked by three scorers as explained by Wiggle (2008). All errors in spelling, punctuation and capitalization were ignored. The required checklist was adopted from Wigglesworth and Storch (2009).

As Table 4 indicates, in order to test reliability of the students’ scores across the four tests in this study, inter-rater reliability was calculated. Three different raters collected the papers and scored the students’ writing compositions. Their main focus was the grammar and content of the writing compositions. The results are summarized in Table 4. As Table 4 shows, the control group was given two tests (pretest = SA1 and posttest = SA2) while the experimental group was given six tests (pretest = ZAD and ZPD assessments, posttest = ZAD and ZPD assessments, TR1 and TR2).
Table 4

Inter-rater reliability

Test

Control group

Experimental group

Pretest (ZAD, ZPD)

ZAD = 0.74

ZAD = 0.82, ZPD = 0.77

Posttest (ZAD, ZPD)

ZAD = 0.81

ZAD = 0.79, ZPD = 0.76

TR1

………

0.83

TR2

………

0.85

Analysis of the first research question of the study

A paired-samples t-test was conducted to evaluate the impact of Dynamic instruction and assessment on students’ ZAD scores on L2 writing proficiency. As Table 5 illustrates, the mean score of the posttest of the experimental group (M = 10.2667) is higher than the mean scores of the pretest (M = 6.1000) i.e. the ZAD mean scores of the experimental group were raised from 6.1000 to 10.2667. Therefore, the results reveal a significant difference from time 1 (pretest) to time 2 (posttest).
Table 5

Descriptive Statistics of the Experimental group’s ZAD performance

  

Paired Samples Statistics

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

Post Exp ZAD

10.2667

30

1.33735

.24417

 

Pre Exp ZAD

6.1000

30

.92289

.16850

According to Table 6, df (i.e. the degree of freedom) is 29 and the 2-tailed sig is .000 which is less than 5, standard deviation is 1.28877 and standard error mean is .23530. Also, tobs = 17.708 is more than the critical t. Thus, the first null hypothesis is rejected. The experimental group improved significantly after the DA instruction.
Table 6

Paired samples T-test of the experimental group’s ZAD performance

Paired samples test

Paired differences

t

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

Post Ex ZAD-

Pre Exp ZAD

4.16667

1.28877

.23530

3.68543

4.64790

17.708

29

.000

A paired-samples t-test was conducted to evaluate the impact of the Dynamic instruction and assessment on students’ ZPD scores on L2 writing proficiency. As Table 7 indicates, the mean score of the posttest of the experimental group (M = 12.5000) is higher than the mean scores of the pretest (M = 8.2333) i.e. the mean scores of the DA group were raised from 8.2333 to 12.5000. The results of the two paired samples T-test show that experimental group’s ZPD scores are higher than their ZAD scores. That is, by providing help and mediation during the ZPD assessment of posttest, the students’ performances were changed positively.
Table 7

Descriptive statistics of the experimental Group’s ZPD assessment

Paired samples statistics

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

Post

Ex

ZPD

12.5000

30

1.10641

.20200

 

Pre Ex

ZPD

8.2333

30

.77385

.14129

According to Table 8, df (i.e. the degree of freedom) is 26 and the 2-tailed sig is .000 which is less than 5. Also, tobs = 28.235 is more than the critical t. Thus, the first null hypothesis is rejected. The experimental group improved significantly after the DA instruction.
Table 8

Paired-Samples T-Test of the Experimental Group’s ZPD performance

Paired Samples Test

Paired Differences

t

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

Post Ex/ZPD-

Pre Exp/ZPD

4.26667

.82768

.15111

3.95761

4.57573

28.235

29

.000

Table 9, indicates that a paired-samples t-test was conducted to evaluate the impact of a non DA-based instruction on students’ scores on L2 writing proficiency. In other words, a Matched t-test was conducted between the pretest and posttest of the control group. The number of students in the non-DA class (i.e. the control group) was 30. The mean scores of the non-dynamic assessment (NDA) group were raised from 6.1667 to 6.9000. So, the mean scores of the control group were not raised significantly. Table 8, includes descriptive information of the control group i.e. number (30), per-post mean scores (6.1667 and 6.9000), standard deviation (.74664 and 1.49366) and standard error mean (.27270 and .13632) respectively.
Table 9

Descriptive statistics of the Control Group: Non-DA-based

Paired Samples Statistics

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

Post Control

6.9000

30

1.49366

.27270

 

Pre control

6.1667

30

.74664

.13623

According to Table 10, the observed t value is calculated to be 3.832 and the degree of freedom is 29 (df = 29). Also 2-tailed sig is .001 which is less than 5. According to the results of the paired-samples t-test, the control group’s performance on the posttest (time 2) was not significantly different from the pretest (time 1) which shows that the non-dynamic method of teaching and assessing writing skill are not useful.
Table 10

Paired Samples Test of the Control Group (non-DA-based)

Paired Samples Test

Paired Differences

t

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

Post control Pre control

.73333

1.04826

.19139

.34191

1.12476

3.832

29

.001

Analysis of the second research question of the study

According to Table 11, two types of assessment were employed in the current study i.e. DA (Dynamic Assessment) and NDA (Non-Dynamic Assessment). The number of participants in each group was 30. Independent-Samples T-test was conducted between the posttests of the two groups only (i.e. DA and NDA) Also, the mean scores of the DA group (10.2667) are higher than the mean scores of the NDA one (6.9000). Therefore it reveals that the DA group outperformed the NDA group because their mean scores are significantly higher.
Table 11

Descriptive statistics of the two groups’ ZAD assessments on the posttests

Group Statistics

 

Instruction Type

N

Mean

Std. Deviation

Std. Error Mean

Writing

DA-bases

30

10.2667

1.33735

.24417

NDA-based

30

6.9000

1.49366

.27270

According to Table 12, there were two groups each containing thirty students. The two groups received two different types of L2 writing instruction (i.e. dynamic assessment and non-dynamic instruction). The experimental group was exposed to Dynamic Assessment of writing proficiency while the control group was exposed to a placebo. Sig (2-tailed) is .000 which is less than 5 so the second null hypothesis is rejected. Also, the mean scores of the experimental group who received a DA (mean = 10.2667) is higher than the mean scores of the control group who received SA (mean = 6.9000).
Table 12

Independent Samples Test

Independent Samples Test

  

Levene’s Test for Equality of Variances

t-test for Equality of Means

  

F

Sig

T

df

Sig. (2-tailed)

Mean Difference

Std. Error Mean

95% Confidence Interval of the Difference

         

Lower

Upper

Writing

Equal variances assumed

.182

.671

9.198

58

.000

3.36667

.36604

2.63396

4.09937

 

Equal variances not assumed

  

9.198

57.305

.000

3.36667

.36604

2.63377

4.09956

Table 13, illustrates a matched t-test which was conducted between the ZPD and ZAD of the experimental group on the posttest. The number of students in both tests was 30. Also, the table provides the descriptive analysis of the two tests such as mean (ZAD = 10.2667, ZPD = 12.5000), standard deviation (ZPD = 1.10641, ZAD = 1.33735) and standard error mean (ZPD. 20,200, ZAD = .24417).
Table 13

Descriptive Statistics of the ZPD and ZAD of the Experimental Group on the Posttest

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

ZPD-Exp

12.5000

30

1.10641

.20200

ZAD-Exp

10.2667

30

1.33735

.24417

Table 14, shows that the posttest of the experimental group had two levels: ZAD assessment and ZPD assessment. That it, first, the students were given a ZAD assessment and the then a ZPD assessment based on the same test. Table 13 reveals the calculated standard deviation = .62606, standard error mean = .11430, df (degree of freedom) = 29 and 2-tailed sig = .000 which is less than 5. Thus the second null hypothesis was rejected. Also the calculated tobs = 19.539. The current table confirms that the ZPD scores of the experimental group are higher than their ZAD scores. In other words, there was significant difference between the mean scores of the solo performance of the experimental group and their assisted performance.
Table 14

Paired Samples Test of the ZPD and ZAD of the Experimental Group on the Posttest

Paired Differences

t

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

ZPD- Exp

ZAD- Exp

2.23333

.62606

.11430

1.99956

2.46711

19.539

29

.000

Analysis of the third and fourth research question of the study

Table 15, shows a matched t-test which was conducted between the posttest and TR1 (of the experimental group). The number of students in both tests was 30. Also, the table provides the descriptive analysis of the two tests such as mean (TR1 = 13.7000, pest Ex = 12.5000), standard deviation (TR1 = 1.87819, post Ex = 1.10641) and standard error mean (TR1 = .34291, post Ex = 20,200).
Table 15

Descriptive statistics of the posttest of Ex. and TR1

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

TR1

13.7000

30

1.87819

.34291

Post Ex

12.5000

30

1.10641

.20200

Table 16, reveals the calculated standard deviation = 1.20000, standard error mean = .18815, df (degree of freedom) = 29 and 2-tailed sig = .000 which is less than 5. Thus the third null hypothesis was rejected. Also the calculated tobs = 6.378. One of the question that the current study tried to was whether the students could apply their newly gained knowledge in new contexts i.e. transcendence. The writing topic of TR1 was more difficult than the posttest. However, the learners’ mean scores reveal that they could successfully apply their newly learned knowledge to TR1.
Table 16

Paired-Samples t-test (between the posttest and TR1 of the experimental group)

Paired Samples Test

Paired Differences

T

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

TR1-Post Ex

1.20000

1.03057

.18815

.81518

1.58482

6.378

29

.000

A Paired Samples t-test was run between the posttest of the experimental group and TR2 in order to see whether the students could apply their newly learned skills to TR2. This test was used to show whether progress could be transferred to TR2 (which was even more challenging than TR1). Table 17, provides descriptive information such as number, mean standard deviation and standard error mean of TR2 which are 30, 9.8667, .93710, .17109 respectively.
Table 17

Descriptive statistics of the Post-Ex and TR2

  

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

TR2

9.8667

30

.93710

.17109

Post Ex

12.5000

30

1.10641

.20200

As Table 18, indicates the mean difference between posttest of the experimental group and TR2 is 2.63333 and the degree of freedom is 29. Also, 2-tailed sig is .000 (which is less than 5). The calculated tobs is 25.939. The calculated standard deviation is .55605 and the standard error mean is .10152.
Table 18

Matched t-test between Post-Ex and TR2

Paired Samples Test

Paired Differences

T

df

Sig (2- tailed)

  

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

   
     

Lower

Upper

   

Pair1

TR2-

-

.55605

.10152

−2.84097

−2.42570

-

29

.000

 

Post Ex

2.63333

    

25.939

  

Performance of the DA group across assessment sessions.

Table 19, is an indication of the mean scores of the experimental group (which was exposed to DA – instruction) across assessment sessions. The experimental group was given four different tests (pretest, posttest, TR1 and TR2). The raised mean scores from the pre-to posttest reveals the significant effect of dynamic assessment. Also the scores of the two TRs are close to the mean scores of the posttest (however not higher); this indicates that the students could successfully transfer their knowledge in more difficult tests. TR1 was near far i.e. a little more difficult than the posttest and TR2 was even more difficult. Both of them were taken from the book ‘for and Against’ which is for advanced level L2 learners.
Table 19

Mean Scores of the Experimental Group across Assessment Sessions

Test

Mean scores

 

6.1000

Pretest (ZAD assessment)

8.2333

Pretest (ZPD assessment)

10.2667

Pretest (ZAD assessment)

12.5000

TR 1

13.7000

TR 2

9.8667

Figure 2 illustrates the development of the experimental group across assessment sessions. DA1 is the pretest which was given to the experimental group and DA2 is the posttest which was administered to the experimental group. As the figure shows, there has been a significant growth on the posttest i.e. DA2 (after the treatment of the study). TR1 was to some extent more difficult than the posttest so there is a bit of decline in the performance of the experimental group. However, the participants could still apply their newly gained knowledge to this task. TR2 was even more difficult than TR1 so the subjects’ performance dropped even more but it is still better than their functioning on the pretest.
Fig. 2

The progress of the experimental group

Qualitative analysis of the data

Following the analysis of interactions, an inventory of meditational strategies emerged which is as follows: Confirming/rejecting response, Repeating the erroneous guess with a questioning tone, Asking leading questions, Using the Internet,Using dictionary, Translation Providing correct response and explanation. Following Aljaafreh and Lantolf (1994), the menu of meditational strategies offered here was arranged from the most abstract (implicit) to the most concrete (explicit). The strategies outlined here were not prescribed in advance but developed out of interactions between the mediator and learners. Like Aljaaferh and Lantolf (1994), the mediational strategies developed in this study followed the abstract-concrete (implicit-explicit) principle. A description of detected meditational strategies along with on-the-spot examples mentioned in Appendix 2.

Descriptive analysis of the data: Frequency of mediator’s strategies in the ZPD

From an SCT perspective, one can track the learners’ progress in the ZPD by referring to the number of mediations offered (Poehner, 2005). This section presents the frequency of the mediator’s strategies identified during the DA interactions between the mediator and learners. Table 20 represents a summary of frequency of meditational strategies over time at three stages of DA sessions (the pretest, posttest, and TR sessions).
Table 20

Descriptive Analysis: Frequency of Meditational Strategies

 

Pretest

Posttest

TR1

TR2

1. Confirming/rejecting response

65

60

54

57

2. Repeating the erroneous guess with questioning tone

60

55

46

45

3. Asking leading questions

80

58

55

56

4. Using the Internet

100

61

70

83

5. Using dictionary

94

69

62

62

6. Translation

98

47

15

18

7. Providing correct response and explanation

74

37

0

10

Total

570

387

302

331

The first meditational strategy (MS1) is the most implicit strategy because the teacher just rejects or confirms the learners’ response; it is left in the hands of the learner to find the correct answer. MS2 just informs the learner that something is wrong whit his/her response. MS3 acts as a prompt or clue because it leads the learner towards the right path so it is less implicit. MS4 provides a model for the learner because according to Bandura (1971), human being needs a model to follow; thus, it shows that the meditational strategies are becoming less and less implicit and more explicit. MS5 is even more explicit because learners can look up unknown items in a dictionary. MS6 is the most explicit meditational strategy because the teacher gives the correct answer and explains the reason. A comparison of the frequency of meditational strategies indicates the microgenetic growth of the students’ ZPD. An important idea underlying DA perspective is that the students’ reduced demands for external and explicit mediation is an indication of self-regulation, more control over their knowledge and last but not least cognitive development (Poehner, 2008). According to Table 20, this claim is supported because it is illustrated that there is a decline in the use of meditational strategies in the posttest.

Discussion

This research presented the results of a qualitative and quantitative study of dynamic assessment across the pretest, posttest and TR assessment sessions. This study provided insights into the learners’ independent/solo as well as joint/dependent writing performance qualitatively and quantitatively through close examination of different types of meditational strategies offered by the mediator during her DA interactions with the learners. The analysis of the dialogic interactions between the mediator and learners was presented and illustrated with protocols from the assessment sessions. The typology of meditational strategies paved the way to know how to offer DA – based mediation and thus gives and thus gives new insights into how to incorporate DA procedures in a classroom to assess and teach L2 writing. A close analysis of the DA interactions showed the employment of assess and teach L2 writing. A close analysis of the DA interactions showed the employment of 7 types of meditational strategies which were placed on a regulatory scale (Table 20). The strategies were arranged based on the abstract/concrete (i.e. implicit/explicit) criterion (Aljaafreh and Lantolf, 1994) which are as follows: confirming/rejecting response, asking the erroneous response with a questioning tone, asking leading questions, using a dictionary, translation, providing correct response asking the erroneous response whit a and explanation. The detected strategies were classified into five categories on their functions:
  1. 1)

    Managing the interactions.

     
  2. 2)

    Helping the learners to reconsider their L2 production.

     
  3. 3)

    Helping the learners to notice the clues.

     
  4. 4)

    Helping the learners to use the artifacts.

     
  5. 5)

    Enhancing writing abilities.

     

In this study, the mediator interacted with the students and the mediations were adjusted to an individual student’s reactions. The regulatory scale was developed after the learners’ developmental path. Also, the frequency table helped track the learners’ developmental processes. The comparison of the learners’ L2 production in the pretest with those in the posttest and TR sessions clearly showed the learners’ reduced demands for explicit meditational and their tendency towards self – regulation. This observation was documented with reference to the learners’ reliance on more implicit kinds of meditational moves in the TR tasks. The observation revealed how DA interactions could create a sense of belonging in the social atmosphere of the classroom and how the learners could benefit from the help and scaffolding provided by their classmates and teacher to improve their L2 writing skill problems. The teacher assigned a writing topic to the students and asked them to write about it and put their ideas into written words. The evidence reported in the present research showed the significant role of interaction and mediation in providing the learners with an opportunity to resolve their writing problems. The DA procedure helped the mediator identify certain problems learners faced during L2 writing which were not visible during unmediated assessment i.e. NDA. For example the data clearly showed that one major problem of the learners in this study was their undeveloped grammar and Vocabulary. The results of diagnostic analysis of mediator – learner interactions supported this claim. Another problem was their unfamiliarity with English expressions and sayings. The findings of the current study go in line with the ex-researchers’ findings such as Baek and Kim (2003), Poehner (2005), Wigglesworth and Storch (2009), and Shabani (2014) confirm them. This study followed Poehner’s (2005) study on French speaking proficiency; however in this study, the researcher focused on L2 writing abilities. Also, Poehner’s study was exclusively qualitative and based on interactional model of DA but in this study the data were analyzed both quantitatively and qualitatively. As a result, the design of the study was based on the Sandwich format of DA (i.e. interventional DA) but the procedure was based on the interactional model of DA.

Conclusion

A humanistic approach sheds lights on the idea that learners are different individuals and it aims at helping learners become more like themselves and less like each other. A DA-based instruction which has it’s roots in Humanism minimizes anxiety and maximizes the sense of security. DA extends and increases the aspect of learning. The analysis of this study reveals that the participants gained more proficiency than did the ones who were in the control group being exposed to a static and standardized practice of L2 writing instruction. As the results indicate, a DA approach of writing proficiency to Iranian EFL learners proves to be useful in uncovering the underlying traits. The present study examined each participant’s performance during the assessments and presented frequencies of the different kinds of meditational moves produced. Comparisons of interactions during the dynamic session at time 1 (prior to the enrichment program) and time 2 (following the enrichment program) provide evidence regarding the extent to which learners have control over their knowledge and performance. In order to confirm and ascertain the quality of the observed changes, the interactions during DA2 (i.e. the second dynamic assessment) were compared to the two TRs (i.e. the transfer assessments). From a Vygotskian perspective, development means going beyond the here-and- now performance on a given challenging tasks. Therefore, the reduced amount of support the learners needed as they performed more challenging tasks, showed their improvement and development. The learners’ verbalization can reveal learners’ problems, insufficient knowledge, etc. As it was mentioned earlier, the analysis of students’ performance during the mediated sessions is based on Aljaafreh and Lantof’s (1994) argument that development can reflect itself not only in improved solo performance but also through changes in the amount and kinds of mediation learners need. With this in mind, the meditational sessions i.e. DA1, DA2, TR1, TR2 were developed for the presence of the different meditational moves that were included in the typology. During the TRs, the students were asked to write about topics of varying complexity. All the participants found TRI to be more challenging and to some extent, it was because of the task’s specific lexical demands. The participants performed mostly independently during TR2. They needed little interaction with the mediator. One of the major contributions of dynamic assessment over static assessment is that the inclusion of mediated interaction gives us more information apropos of a learner’s abilities After all, two heads are better than one.

Abbreviations

ACTFL: 

American Council on the Teaching of Foreign Languages

Cntr: 

Control

DA: 

Dynamic Assessment

Ex: 

Experimental

N: 

Number

NDA: 

Non- Dynamic Assessment

OPT: 

Oxford Placement Test

S: 

Student

SA: 

Static Assessment

T: 

Teacher

TR: 

Transcendence

ZAD: 

Zone of Actual Development

ZPD: 

Zone of Proximal Development

Declarations

Acknowledgments

Hereby, I would like to express my heartfelt and gratitude to Ayda Rahmani, cooperating in this study. Besides, I wish to express my appreciation to the students for their contribution in this study.

Authors’ contributions

This work was carried out in collaboration between the two authors. Author PF supervised the work, decided upon and provided the sources, and revised and edited the drafts. Author AR collected the data and wrote the drafts of the manuscript, performed the statistical analyses, and implemented the revisions. Both authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

(1)
Department of English Translation, Lahijan Branch, Islamic Azad University
(2)
Department of English Language, Tonekabon Branch, Islamic Azad University

References

  1. Aljaafreh, A., & Lantolf, J. P. (1994). Negative feedback as regulation and second language learning in the zone of proximal development. The Modern Language Journal, 18, 465–483.Google Scholar
  2. Baek, S. G., & Kim, J. K. (2003). The effect of dynamic assessment based instruction on childrenʼs learning. Asian Pacific Review, 3(1), 189–198.Google Scholar
  3. Bailey, K. (1996). Working for washback: A review of the Washback concept in. Language Testing, 13(3), 257–279.View ArticleGoogle Scholar
  4. Bandura, A. (1971). Social learning theory. New York: General Learning Press.Google Scholar
  5. Messick, S. (1988). The once and future issues of validity: Assessing the meaning and consequences of measurement. In H. Wainer & H. I. Barun (Eds.), test validity. NJ. Erlbaum: Hillsdale.Google Scholar
  6. Poehner, M. E. (2005). Dynamic assessment of oral proficiency among advanced L 2 learners of French. Pennsylvania State University, University Park: Unpublished Doctoral Dissertation.Google Scholar
  7. Poehner, M. E. (2008). Dynamic assessment: A Vygotskian approach to understanding and promoting second language development. Berlin: Springer Publishing.Google Scholar
  8. Shabani, K. (2014). Dynamic assessment of L2 listening comprehension in transcendence tasks. Social and Behavioural. Science, 98, 1729–1737.Google Scholar
  9. Stanley, N. V. (1993). Gifted and the zone of proximal development. Gifted Educational International, 9, 78–81.View ArticleGoogle Scholar
  10. Tzuriel, D. (2000). Dynamic assessment of young children. New York: Plenum Publishers.Google Scholar
  11. Vygotsky, L.S. (1988). The problem of age. New York: Plenum.Google Scholar
  12. Vygotsky, L. S. (1978). Mind in society. Cambridge: Mass: MIT Press. Google Scholar
  13. Wiggle, S. C. (2008). Assessing writing. Cambridge: Cambridge University Press.Google Scholar
  14. Wigglesworth, G., & Storch, N. (2009). Paired versus individual writing: Effects on fluency, complexity and accuracy. Learning testing Journal, 26(1), 445–452.Google Scholar

Copyright

© The Author(s) 2017