Introduction

Second language teacher education (SLTE) can be described as an evolving field. ^{Wright (2010)} contends that in the 1980’s SLTE mainly focused on teaching methods and techniques whereas towards the end of the century, it became more concerned with learning to teach. ^{Freeman and Johnson (1998)} argue in favour of a reconceptualised knowledge base of SLTE, which encompasses three domains: “(a) the nature of the teacher-learner, (b) the nature of schools and schooling, and (c) the nature of language teaching” (p. 406). In a similar vein, socio-cultural theory (SCT) (^{Vygotsky, 1978}) has influenced SLTE as teacher education programmes have come to conceive of student-teachers as “a community of learners engaged in social practices and the collaborative construction of meanings” and have advocated teaching modes which involve “dialogue and collaborative inquiry” (^{Richards, 2008}, pp. 164-165). Activities that foster this form of joint participation help the student-teachers engage in conversation with teacher educators, school-based tutors, peers, and school authorities and learn from these formative meetings.

Teaching practice offers a multiplicity of opportunities to work with others; however, most of the research carried out has tended to focus on post-observation conferences and the feedback that the student-teachers are given after teaching a lesson (^{Brandt, 2008}; ^{Copland, 2010}; ^{Tang & Chow, 2007}). Much less is known about other instructional practices such as tutoring sessions which involve dialogue and collaborative inquiry. From the perspective of SCT, teacher educators play a crucial part since they need to effectively address each student-teacher’s individual zones of proximal development in order to enhance their potential for learning and focus not only on what they can already do on their own but also on what they can attain with the help of others. Therefore, one of the key skills that they need to develop is to scaffold the student-teachers’ learning-to-teach process. How and to what extent scaffolding unfolds in these tutoring sessions is the main concern of the present investigation.

Literature Review

The last decades have witnessed a steady growth in research on teacher education and development from the perspective of SCT. Adequate guidance or scaffolding is an essential process to assist and guide the prospective teachers to complete a task successfully or achieve a goal (^{Wood, Bruner, & Ross, 1976}). In a case study involving a university supervisor (US), a cooperating teacher (CT), and a pre-service teacher (PT) in physical education during the field placement, ^{Cartaut and Bertone (2009)} explored the specific and complementary scaffolding modalities used by the two teacher educators as well as their effects on the PT’s professional activity development. The researchers reported that the US supported the PT by suggesting directions for finding solutions and by raising new concerns about the requirements of the teaching profession in general and of the training institute in particular. It was further found that the CT’s scaffolding activity comprised the provision of alternative actions both in terms of goals and concrete operations. In addition, the PT revealed that his views and actions in the classroom changed as a result of the US’s and CT’s joint and complementary scaffolding processes in the advisory visit and the CT’s follow-up in the field training interactions. These results provided evidence of the scaffolds the two teacher educators deployed and the role that the support or scaffolding modalities played to help the PT develop professionally. Scaffolding can also be used as a strategy to prompt student-teachers to self-analyse and question their past experiences and current beliefs. ^{Van Zoest and Stockero (2008)} conducted research to examine the role of synergistic scaffolds (^{Tabak, 2004}) in supporting knowledge of self-as-teacher. The researchers designed and implemented six scaffolds in a secondary school mathematics teacher preparation program. They concluded that the student-teachers had developed a sense of self-as-teacher and that the scaffolds had served the purpose of prompting changes. To a lesser extent, some changes not prompted by the instructor were introduced; an action which points at influences other than the specific six scaffolds. They also explored the pre-service teachers’ perceptions of the usefulness of the six scaffolds in supporting their thinking. All participants reported that the scaffolds had encouraged them to think more thoroughly than they would have otherwise, and alluded to their synergistic use. The researchers claim to have addressed the student-teachers’ learning needs when designing and implementing the scaffolds; nevertheless, when stating the limitations of the study, Van Zoest and Stockero call for the systematic diagnosis of pre-service teachers’ learning needs, which represent another dimension of scaffolding that was not analysed in this study.

In order to research trainer talk from a linguistic point of view in the context of an MA class of Turkish pre-service English teachers, ^{Engin (2013)} analysed the trainer’s intervention strategies deployed in the post observation conferences as a means to scaffold the trainees’ reflections and classify them according to different levels of intervention. Afterward, the author found five different levels of trainer scaffolding in the interactions studied, ranging from least to direct intervention.

^{Cartaut and Bertone’s (2009)}, ^{Engin’s (2013)}, and ^{Van Zoest and Stockero’s (2008)} research studies examine scaffolding provided to either pre-service or in-service teachers as an aid to support different activities. Nevertheless, these studies only analyse the scaffolding process from the perspective of support, thus focusing solely on scaffolded help in terms of intervention strategies. Van de Pol (2012) points out that not all forms of support can be equated with scaffolding since the assistance provided needs to be contingent upon the learners’ current level of understanding and faded over time.

Some studies have analysed scaffolding from the perspective of contingent support. For example, ^{Chin (2007)} carried out research to explore how teachers use questions to scaffold student thinking and knowledge construction. Lessons taught by six secondary school science teachers were audiotaped and videotaped. The different questioning techniques used by the teachers were identified. It was also found that the teachers’ questions built on a preceding student contribution and served as “rungs of a ‘cognitive ladder’ enabling students to gradually ascend to higher levels of knowledge and understanding” (p. 837). Therefore, the teachers’ questions were contingent in that they adjusted to the knowledge base of the students.

Researchers ^{van de Pol, Volman, and Beishuizen (2011)} conducted research to investigate the one-to-one and small-group teacher-student interactions in order to describe the process of scaffolding. The study involved three social studies teachers working at secondary education innovative schools of lower prevocational education in the Netherlands. These innovative schools, contrary to more traditional schools, were expected to show more instances of contingent teaching and autonomous student learning, which are key features of scaffolding. The researchers observed that in two thirds of the non-contingent interactions, the teachers did not use diagnostic strategies. Lack of diagnostic strategies was found to occur along with miscommunication in many cases.

That teachers resort to different strategies and skills to support and scaffold student learning is generally accepted notion. The studies that examine scaffolding and equate it with teacher help or support have identified numerous ways in which this scaffolded help can be realized. In addition, since these studies are mainly descriptive, the number of skills and strategies identified is quite comprehensive depending on their context of study. The review of the literature also emphasizes the fact that teachers need to diagnose and identify their learners’ needs before actually giving support. Therefore, they scaffold their students’ learning by making use of diagnostic strategies and providing contingent support.

Theoretical Framework

This research study is theoretically anchored on SCT (^{Vygotsky, 1978}), which provides a detailed account of the interrelationship between learning and development. Vygotsky (1978) contended that psychometric tests only reflected the learners’ current developmental level and disregarded their potential abilities. SCT distinguishes two developmental levels. The actual developmental level refers to the individual’s mental functions that are already completed or matured, which enable the individuals to perform activities on their own without any help. However, an individual’s mental development is also indicated by those abilities that are under the process of maturation and that enable individuals to achieve different learning goals with the assistance of others. This stage is called the potential developmental level. The child is conceived of as an integrated whole of relationships that comprise developed and developing higher mental functions acquired through collaboration (^{Chaiklin, 2003}). Vygotsky introduced the core concept of the zone of proximal development (ZPD) and defined it as “the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers” (p. 86). This key construct helps explain how learning can foster qualitative developmental changes (^{Lantolf & Thorne, 2006}). Therefore, from a Vygotskian perspective learning is in advance of development as the former triggers different developmental processes when the individuals interact with experts or more capable peers in their environment. Co-constructed knowledge and abilities become internalized and new ZPDs emerge. Chaiklin (2003) states that the ZPD is not a fixed feature as it evolves across age periods and constitutes a potential for learning (^{Wells, 1999}).

The ZPD has significant implications for teaching and learning. ^{Vygotsky (1978)} calls for diagnostic procedures to ground the teachers’ pedagogical interventions and assess the learners’ ZPD by determining maturing functions. In classroom settings, teachers can play a key role in guiding the learners’ course of learning and, thus, development by drawing on their existing capabilities and creating the appropriate conditions for learning (^{Lantolf & Thorne, 2006}). Therefore, scaffolding (^{Wood et al., 1976}) can be a useful strategy to help learners move across different ZPDs since it involves the support a teacher gives to a learner when carrying out a task which he/she would not be able to accomplish on his/her own (^{van de Pol, Volman, & Beishuizen, 2010}).

Method

Bearing in mind the numerous characteristics of qualitative research identified by different scholars (^{Creswell, 2003}; ^{Dörnyei, 2007}; ^{Hatch, 2002}; ^{Mackey & Gass, 2005}) one can see the present study is mainly framed within a qualitative methodology. To start with, the transcriptions of the on-going interactions between the participants enabled me as a researcher to have access to rich and complex details which, in turn, catered for a rich description of the scaffolding process in the setting investigated. Secondly, the research setting itself involved a naturalistic kind of inquiry (Lincoln & Guba as cited in Hatch, 2002) since scaffolding was studied in its naturally-occurring situation without manipulating any variables. Thirdly, studying the phenomenon in its natural context also contributed to understanding scaffolding from an “insider perspective” (Dörnyei, 2007, p. 38). Finally, although the study comprised a thick description of the scaffolding process, I sought to go further into an interpretative stage by making personal meanings while taking into account my role as researcher and my own personal biases and value stance as a practicum supervisor.

The research also involves a case study design. The interactions in which a student-teacher’s learning is scaffolded by a practicum supervisor during tutoring sessions in the context of an English as a foreign language (EFL) teacher education programme is, therefore, the case or unit of analysis of the present work.

Context

This study took place in the context of the last practicum student-teachers had to undertake in an EFL teacher education programme at a tertiary level teacher education college in Córdoba, Argentina. The practicum comprises both observing lessons taught by a graduate teacher and taking a total of about ten practica. The practicum supervisor and the student-teachers met regularly during one-to-one tutoring sessions and post-observation conferences. This study was particularly concerned with the one-to-one tutoring-sessions that the practicum supervisor held with student-teachers weekly as a course requirement. They met in order to discuss lesson plans, require assistance, comment on previously taught lessons, and exchange views on changes made to the lesson plans after being suggested by the supervisor by email, among other issues.

Data Collection

All the one-to-one tutoring sessions between the supervisor and each student-teacher were audio-recorded. The supervisor was asked to hold them as usual and record the full interactions no matter what subject matter was being discussed. The use of the audio-recordings allowed me to capture the scaffolding process in a naturally occurring situation, in which I was interested as a practicum supervisor and researcher. Furthermore, since the focus of the study was on the on-going interactions between the participants, especially the ways in which they discussed different aspects of teaching practice and responded to each other’s words and/or comments, the audio recordings provided me with accurate and detailed data about the complexities and subtleties of the phenomena studied. They also enabled me to get a verbatim account of everything that was said, listen to the conversations countless times, and carry out macro- and micro-level analyses.

A word must also be said about the limitations of using audio recordings. In this particular case, although it was agreed with the supervisor that the full interactions would be recorded, some parts might have been omitted since it was the supervisor who decided when to turn on the recorder and when to turn it off; therefore, the recordings might not provide a fully accurate picture of the interactions. Moreover, the impact that recording the conversations might have had on the participants’ behaviour should not be overlooked and, therefore, claims should be carefully made.

Data Analysis

Data analysis comprised different steps. Firstly, the data from the audio recordings were transcribed without deleting any parts in order not to make any judgments in advance and to get to know about all of the information collected thoroughly (^{Dörnyei, 2007}). Following Dörnyei (2007), a “pick and mix procedure” (p. 248) was adopted to transcribe the data in order to fit my own research concerns (^{Mackey & Gass, 2005}). The rationale for choosing certain conventions was: a focus on content and function as well as accessibility to readers. Next, the data were read several times and then several criteria were established for selecting fragments for analysis.

Since the results of some of the research carried out earlier in the field of scaffolding and measuring instruments are already available, I opted for a tighter or deductive design (^{Miles & Huberman, 1994}) with a pre-established set of analytic categories to describe and measure scaffolding. The teacher-students’ on-going interactions in the one-to-one tutoring sessions were analysed by resorting to the data-analysis instruments devised by ^{van de Pol (2012)}. In particular, the model of contingent teaching (MCT) was deployed to carry out the analysis. Nevertheless, I followed an open-ended data-led procedure and allowed newly conceived categories to emerge and be included in the analysis.

The MCT (see Table 1) devised by ^{van de Pol (2012)} serves as an instrument to describe the process of scaffolding in qualitative terms. The MCT consists of four steps:

Table 1 Steps of the Model of Contingent Teaching (Adapted From ^{van de Pol, 2012}, p. 85) 

These steps uphold the interactive nature of scaffolding since each teacher’s turn is followed and thus determined by a student’s turn.

The analysis focused on a qualitative examination of the scaffolding process. Drawing on the MCT, for each fragment, all teacher turns were coded as Step 1, 2, 3, or 4, bearing in mind the function they served in relation to the student-teachers’ turns. The following step consisted of identifying the combinations of steps in each interaction fragment, that is to say, cycles of contingent teaching. To round off the analysis, all interaction fragments were coded for contingency. A fragment was considered contingent, and consequently, a scaffolding example if the supervisor first made use of a diagnostic strategy and then provided assistance which was tailored to meet the student-teacher’s needs or level of understanding. A fragment was coded as non-contingent, and consequently, a non-scaffolding example if no diagnostic strategy was employed and immediate support was provided by the supervisor.

Results

The data analysed consisted of 24 tutoring sessions which were recorded by the practicum supervisor. The sessions varied in length, ranging from six to nineteen minutes. Most of the sessions, however, lasted about 11 minutes. The sessions were further divided into 102 interaction fragments. All the student-teachers’ turns were analysed as a unit, so I did not distinguish among the different student-teachers.

The results section examines how the scaffolding process unfolded in the one-to-one interactions between the practicum supervisor and the student-teachers. I provide a descriptive account of the steps taken by the supervisor by drawing on the analytic categories provided by ^{van de Pol (2012)}. It must be noted that the steps identified in the supervisor’s speech reflect the function they served in relation to the student-teachers’ reactions and/or responses, thus depicting the interactive nature of scaffolding.

Discussion and Conclusion

This study has captured the interactive nature of scaffolding as the ways and the extent to which the supervisor’s actions and utterances served to scaffold the student-teachers’ learning-to-teach process in the context of the one-to-one tutoring sessions could only be understood by analysing them in relation to the student-teachers’ actions and utterances. In other words, the role of dialogue in the on-going interactions between the participants is a crucial component of scaffolding as several researchers contend (^{Puntambekar & Kolodner, 2005}; ^{Stone, 1998a}, 1998b; ^{Tharp & Gallimore, 1991}; ^{Wertsch, 1979}).

From a qualitative perspective, the findings of this study suggest that the scaffolding process in the tutoring sessions comprises two main steps or phases: a diagnostic phase and an intervention phase. The MCT distinguishes Step 1 turns (diagnostic strategies) from Step 2 turns (checking the diagnosis). Resorting to only Step 1 turns or both Step 1 and Step 2 turns together reveals the supervisor’s need to gather essential information in which to ground her decisions as to what type of and how much help or assistance to give the student-teachers. In the tutoring sessions, the use of Step 2 turns seems to reinforce the diagnostic phase since it serves to round-off the supervisor’s assumptions and/or get a more focused idea of the student-teachers’ level of understanding. Therefore, Step 2 turns may be subsumed under the diagnostic phase since their purpose resembles and complements that of Step 1 turns. The intervention phase is manifested by the use of multiple and simultaneous ways of offering help, which lends support to the use of synergistic scaffolds proposed by ^Tabak(2004).

In the study described here, teaching cycles, which consisted of 1-3 or 1-2-3 steps of the MCT, were the most recurrent ones. Consequently, the findings indicate that a diagnostic phase made up of either Step 1 or of both Steps 1-2 is common practice in the tutoring sessions. However, other studies have found the use of diagnostic strategies to be scarce (^{van de Pol et al., 2011}). This difference may be motivated by the expected or defined structure of the one-to-one tutoring sessions in the context researched here. One of the teachers in van de Pol et al.’s (2011) study was found to base his help on his beliefs about what is difficult for students, and it was found that another teacher rarely resorted to diagnostic strategies due to time-constraints. In the tutoring sessions, the student-teachers were expected to explain their choices and the decisions behind lesson planning. In other cases, the supervisor read their lesson plans. There seemed to be a negotiated agreement between the participants as to how the tutoring sessions should proceed. Both activities provided the supervisor with clear insights into the student-teachers’ level of understanding as well as their learning needs. This pre-defined structure of the tutoring sessions gives the supervisor plenty of information on which to draw in order to provide the most adequate amount and type of help or assistance required. The supervisor hardly ever provided support without first gathering information about the student-teachers’ level of understanding. This was only found to occur in cases in which the student-teachers initiated the interaction by raising a difficulty or concern they had, which was immediately addressed by the supervisor. To sum-up, the structure and the overall purpose of the tutoring sessions imply an initial stage of diagnosis which pre-determines the function of both the supervisor’s and the student-teachers’ actions and utterances.

The fact that complete teaching cycles consisting of Steps 1-2-3-4 were not identified in the one-to-one tutoring sessions analysed here is noteworthy. The supervisor was found to take great effort to diagnose the student-teachers’ current level of understanding and, thus, provide tailored support but she did not check the student-teachers’ new learning afterwards. It appears as if the supervisor assumed that teaching necessarily amounted to learning. In other words, the supervisor seemed to take for granted that all the support she provided the student-teachers with by different means led to new understandings and learning. Of all the steps of the MCT, the supervisor focused mainly on Steps 1 and 3, but the lack of Step 4 should make us wonder to what extent true scaffolding occurred. This finding requires further research.

The qualitative analysis also sought to reveal the patterns of contingent and non-contingent teaching cycles. The present research has found the practicum supervisor to act contingently upon the student-teachers’ level most of the time because she usually resorted to diagnostic strategies before providing actual support. In keeping with the findings reported here, ^{Chin (2007)} found the teachers to provide their students with contingent support because they showed evidence of offering situated help and thus adjusting to the knowledge base of the students. In the context of the tutoring sessions, contingency can be best understood by resorting to Chin’s metaphor, which describes contingent support in student-teacher interactions as “rungs of a cognitive ladder” (p. 837) since the teacher’s help builds on the students’ prior knowledge and, at the same time, it helps them achieve higher levels of competence. From the perspective of SCT, the use of diagnostic strategies helps teachers determine the students’ maturing functions and, therefore, their ZPDs. The situated support they provide them with helps them to become self-regulated and to internalize knowledge and skills and reach higher levels of cognitive development since their ZPDs gradually evolve (^{Chaiklin, 2003}). In conclusion, diagnostic strategies seem to be a crucial dimension of scaffolding and a stepping stone for fostering learning and development in the context researched here because they appear to be a necessary condition for providing contingent support, enhancing the student-teachers’ potential for learning (^{Wells, 1999}) and gradually handing over the responsibility for teaching to the student-teachers themselves.

Scaffolding is a complex and dynamic phenomenon which is gradually shaped by the participants’ intervention modalities and, at the same time, influences the participants’ on-going interactions. What both the supervisor and the student-teachers do and say are closely intertwined since they are two sides of the same coin. Even though on the surface the analysis of scaffolding may seem to focus solely on the supervisors’ roles and skills, it necessarily incorporates the student-teachers’ perspectives since true scaffolding is characterized by its interactive dialogic nature. Consequently, this study has attempted to narrow down a gap in research by including the recipients of the teacher’s help (Randall & Thornton, 2001) in the analysis.