Abstract
Adapting support contingently to students’ needs by diagnosing students’ existing understanding first – i.e., scaffolding – is considered a crucial aspect of excellent teaching. However, performing diagnostic strategies and using the information in subsequent support is found to be difficult and rare in classroom practice. Therefore, the aim of the current experimental classroom study was to investigate whether and how a model of contingent teaching (consisting of the following steps: using diagnostic strategies, checking the diagnosis, giving contingent support and checking students’
learning) promoted teachers scaffolding behaviour. Seventeen teachers participated in an intervention programme based on the model of contingent teaching whereas thirteen teachers did serve as a control group. All teachers (prevocational education, social studies) taught the same five-lesson project on the European Union and their development of the quantity and quality of the steps of contingent teaching was investigated. Not so much the quantity but the quality of the steps of contingent teaching appeared to increase while working with the model of contingent teaching; this included contingent support. The model of contingent teaching proved to be a powerful tool in promoting teachers’ scaffolding behaviour and can therefore fulfil an important role in future scaffolding research as well as in future teacher education.
12 This chapter is based on:
Van de Pol, J., Volman, M., Oort, F., & Beishuizen, J. (resubmitted). Teacher scaffolding in small-group work: An intervention study.
INTRODUCTION
Scaffolding or adaptive support is recognised as a distinct feature of excellent teaching (Borko, Mayfield, Marion, Flexer, & Cumbro, 1997; Seidel & Shavelson, 2007;
Wittwer & Renkl, 2008). Ausubel recognised this more than four decades ago: “If I had to reduce all of educational psychology to just one principle, I would say this: The most important single factor influencing learning is what the learner already knows.
Ascertain this and teach him accordingly” (Ausubel, 1968, p. vi). Scaffolding is found to be an effective way of helping learners (Mattanah, Pratt, Cowan & Cowan, 2005;
Pino-Pasternak, Whitebread & Tolmie, 2010; Pratt, Green, McVicar, & Bountrogianni, 1992). In addition, the metaphor of scaffolding attracts both teachers and researchers because of the powerful nature of the metaphor: it appeals to the imagination (Mercer
& Littleton, 2007; Saban, Kocbeker, & Saban, 2007).
However, because scaffolding is not just any support but support that is temporary, tailored to students’ needs and aimed at transferring responsibility, performing it is rather challenging. Several studies have found that – when adhering to this definition of the concept – scaffolding in classroom practice is scarce (Elbers, Hajer, Jonkers,
& Koole, 2008; Lockhorst, Wubbels, & van Oers, 2010). Especially first diagnosing students’ understanding and then using that information in order to tailor the support is difficult, particularly when more than one student is involved such as in supporting small groups of students (Elbers, et al., 2008; Nathan & Petrosino, 2001; Van de Pol, Volman, & Beishuizen, 2011; Webb, Nemer, & Ing, 2006; Wittwer, Nückles, & Renkl, 2010).
Therefore, finding ways to promote teachers’ scaffolding skills is vital. However, to our knowledge, few studies have investigated how scaffolding can be stimulated.
We conducted an experimental study involving 30 teachers to investigate the effects of a professional development programme (PDP) aimed at promoting scaffolding. The model of contingent teaching – in which diagnosing students’ understanding before giving support stands central – (Van de Pol, Volman, & Beishuizen, 2012) formed the basis of this programme. The goal of the current study was to investigate how teachers – who worked with the model of contingent teaching – differed in their ways of teaching compared to teachers who did not work with the model. The research question of this study was: What are the effects of a PDP based on the model of contingent teaching on teachers’ scaffolding behaviour?
The Concept of Scaffolding
The concept of scaffolding is related to Vygotsky’s sociocultural theory and
especially to Vygotsky’s Zone of Proximal Development or ZPD (1978). The ZPD describes the difference between what a learner can do alone (actual level of performance) and what a learner can do with support (potential level of performance).
The concept of scaffolding was coined in 1976 by Wood, Bruner, and Ross and was used to describe support that a tutor could give to a tutee to succeed at a task that he or she could otherwise not perform (Wood et al., 1976, p. 90). Scaffolding can thus be seen as support that helps learners reaching a potential level of performance.
Scaffolding can be characterised by three main features: (1) adaptivity or contingency, (2) fading of support over time, and (3) transfer of responsibility for a task or for learning to the student (Van de Pol, Volman, & Beishuizen, 2010). Contingency refers to adapting support to a student’s needs (e.g., Wood, Wood, & Middleton, 1978).
In addition, support needs to be faded over time in order to transfer the responsibility for learning or for a task to the learner. To be able to be adaptive, estimating or diagnosing a student’s current understanding first before giving support is crucial (e.g., Snow &
Swanson, 1992; Wittwer & Renkl, 2008).
Diagnosing Students’ Understanding
Diagnosing students’ understanding is inheritably connected to the concept of scaffolding (Pea, 2004; Puntambekar & Hübscher, 2005). To be able to tailor support to students’ understanding, this students’ understanding should be explored first, for example by asking questions (Wittwer & Renkl, 2008). Asking diagnostic questions appears important as teachers’ more general estimates of learners’ understanding often appear inaccurate (e.g., Begeny, Krouse, Brown, & Mann, 2011; Chi, Siler, & Jeong, 2004). Teachers often do not first diagnose students’ understanding before providing support (Elbers et al., 2008; Lockhorst et al., 2010; Nathan & Petrosino, 2001; Van de Pol et al., 2011; Webb et al., 2006; Wittwer et al., 2010). Reasons for lack of using diagnostic questions are, amongst others, (1) deficient diagnostic skills, (2) using more general information about the students (e.g., “this student is a bad reader” or “this students cannot concentrate well”), (3) helping directly is an automatic reaction, and (4) time constraints (Elbers et al., 2008; Morrison & Lederman, 2003; Wittwer et al., 2010).
And if teachers diagnose students’ understanding, using this information while giving support also appeared to be difficult (Ruiz-Primo & Furtak, 2007). Moreover, the few diagnostic questions that are being asked are mostly of low quality, i.e., questions that elicit claims of understanding (e.g., “I get it”) only (Wittwer et al., 2010) or low-level factual recall questions (Chin, 2007; Morrison & Lederman, 2003; Roth, 1996). More open diagnostic questions that elicit demonstrations (e.g., students’ explanations) and questions that elicit more than just factual recall (i.e., focus on conceptual understanding)
will give teachers more detailed information about the students’ understanding and provoke deeper reasoning and could therefore be considered of higher quality. These types of questions are known to be related to higher students’ achievements (Barnes, 1975; Redfield & Rousseau, 1981; Ryan, 1973). Promoting the use of diagnostic strategies and especially high quality diagnostic strategies is thus crucial. However, little is known about how to promote teachers’ diagnostic skills; the few available studies on fostering diagnostic strategies are conducted in (peer) tutoring contexts (e.g., King, Staffieri, & Adelgais, 1998; Roscoe & Chi, 2007), not in classroom contexts.
Creating Common Understanding
Before giving support, creating common understanding is an important aspect of the scaffolding process. To prevent misconceptions of the students’ current understanding, a teacher can actually check whether his or her understanding of the students’ understanding is correct; i.e., checking the diagnosis. The purpose here is not to support students, but to check one’s own understanding of the students’
understanding; checking the diagnosis is thus different from what many researchers (Yifat, & Zadunaisky-Ehrlich, 2008) call revoicing or recasting for which the main goal is giving feedback.
Creating common understanding is for example stressed in a teaching method that is closely linked to scaffolding; the Socratic Dialogue (Heckman, 1981). In a Socratic Dialogue, students’ learning is guided and directed by a facilitator using certain ground rules of communication. One import ground rule is using a check on whether the student’s meaning is understood correctly (Knežiç, 2011). Knežiç (2011) is one of few who investigated the effects of a Socratic Dialogue intervention on – amongst other things – the use of this checking step. Her experimental study showed that teachers who participated in the PDP increased their use of this step significantly more than the teachers who did not participate in this programme. The overall use, however, in both conditions stayed considerably low. Increasing the use of this check thus seems possible, but still progressions could be made as this is an important step in the scaffolding process. In addition, no attention has been paid to the quality of this step yet.
Contingent Support
When information on the students’ understanding is collected and checked, a teacher can proceed with giving support that is using and/or adapted to the diagnostic information that is gathered; i.e., giving contingent support.
Contingency is recognised as a crucial characteristic of effective support, also in the context of small-group work (Blatchford, Baines, Rubie-Davies, Bassett, & Chowne, 2006; Chiu, 2004; Webb, 2009) and it is pointed out as a crucial predictor of the success of support given (Mattannah et al., 2005; Pino-Pasternak et al., 2010; Pratt et al., 1992).
Contingency is considered as a necessary condition for scaffolding; if fading or transfer of responsibility takes place in a noncontingent way, we cannot speak of scaffolding.
Therefore, we focus in this study on the aspect of contingency. Contingency is defined as the adaptation of a teacher’s support to students’ responses. Note that the definition of contingency does not include any notion of the effects or effectiveness of contingent support on students’ understanding. We distinguish between two interrelated types of contingency: (1) the tailoring of the degree of control that is exercised in support to the students’ understanding; contingency-control, and (2) the degree to which the teacher reacts to and uses (i.e., takes up) what students say; contingency-uptake.
Wood and his colleagues elaborated the concept of scaffolding by emphasising that the support (in terms of degree of control) needed to be tailored to the understanding of the students, in order for the students to reach a deeper level of understanding (e.g., Wood et al., 1978). Help with a high degree of control was for example modelling the answer whereas help with a low degree of control was for example a general verbal encouragement. Scaffolding does not represent just any support, but support that is tailored; the level of control exercised in the support given should be increased when a learner fails and decreased when the learner succeeds (Wood & Middleton, 1975).
This principle was called the contingent shift principle and it is highly related to the other key characteristics of scaffolding, namely fading and transfer of responsibility (cf.
Wood, 1991). That is, following the learners’ successful progression, the tutor fades the degree of control which gives the learner more responsibility.
Another type of contingency is contingency-uptake. Adapting support to students’
responses can also be done by acknowledging and using what the students have said during the conversation; i.e., uptake (e.g., Nystrand, Wu, Gamoran, Zeiser, & Long, 2003; Wells, 2010). If uptake takes place the student influences the course of the conversation and he/she makes a contribution to that conversation. It means that the teacher has to listen carefully to what students say in order to be able to incorporate this into the ongoing conversation; the teacher’s response is depending on and tailored to the students’ responses (Nystrand et al., 2003).
Tutors (in one-to-one situations) seem to be able to give contingent support – if they receive information on students’ understanding (Wittwer & Renkl, 2008). However, Ruiz-Primo and Furtak (2007) found that teachers in classroom situations had difficulties using the diagnostic information; the teachers in their study did gather information on students’ understanding, but found it difficult to use in their support. In a classroom
situation, using the information gathered thus seems more difficult and this calls for finding out ways of promoting this important skill. However, how to stimulate teachers’
degree of contingency is not clear yet.
Checking Students’ Learning
A final step in the scaffolding process is that of checking students’ learning. This is different from the initial phase of diagnosing students’ understanding in that it no longer focuses on students’ initial understanding but on students’ new understanding or learning as it developed in the course of the conversation. Checking students’ learning is also different from creating common understanding in that it focuses on the students’
understanding, not the teacher’s understanding of the students’ understanding.
Checking students’ learning is mainly performed by asking questions, preferably those questions that elicit a demonstration (e.g., a students’ explanation or elaboration), rather than only a claim of understanding (e.g., “I get it!”) (Koole, 2010; Sacks, 1992).
Only if a demonstration is given, the teacher can determine how a student understood the new subject-matter and whether he or she can transfer the responsibility back to the student. In addition, it appeared that students often have difficulty in gauging their own understanding and often overestimate this (Freund & Kasten, 2012). Therefore performing such a check while eliciting demonstrations is a vital part of scaffolding.
However, when using this check in classroom practice, teachers appear to mainly elicit short and simple answers (Kao, Carkin, & Hsu, 2011), as opposed to demonstrations.
Finding ways to promote this skill – as a part of scaffolding behaviour – thus seems important.
The Model of Contingent Teaching
A model that integrates the four aforementioned crucial aspects of the scaffolding process are summarised in the model of contingent teaching (Van de Pol et al., 2011;
based on Ruiz-Primo & Furtak, 2007) (Figure 1).
Figure 1. Model of contingent teaching
This model contributes to the current conceptualisations of scaffolding as it provides a concrete step-by-step operationalisation of the theoretical concept of scaffolding.
Following the modelling of contingent teaching, teachers should first diagnose students’ current or actual understanding by for example asking diagnostic questions or reading the work of the students. Second, teachers can check their diagnosis with the student, to increase common understanding. Third, the teacher can then support the student contingently, using the gathered information. Finally, the teacher can check the students’ new (potential) understanding or the students’ learning. Examples of each step can be found in the results section in Excerpt 1 and Excerpt 2. This model of contingent teaching stood central in the PDP that was used in the current study to promote teachers’ scaffolding behaviour. Emphasis on gathering diagnostic information and checking the diagnosis is assumed to facilitate contingent support (e.g., Chiu, 2004). Applying these steps in a high quality manner can be seen as scaffolding.
The Current Study
In the current study, we investigated the extent to which the model of contingent teaching promoted teachers’ scaffolding or provision of contingent support. The model of contingent teaching can be seen as the curriculum for the participating teachers in order to learn how to scaffold. The PDP itself was informed by knowledge from recent reviews on factors that enhance the effectiveness of PDPs (Darling-Hammond
& Bransford, 2005; Loucks-Horsley & Matsumoto, 1999; Timperley, 2007; Van Veen, Zwart, & Meirink, 2012). The programme was designed using the following features:
(a) learner centred (i.e., focus on and connect to the learners’ (i.e., teachers’) prior knowledge); (b) knowledge centred (i.e., teachers need to have opportunities to develop well-organised bodies of knowledge, e.g., by using a central conceptual framework); (c)
3. Intervention
assessment centred (opportunities for reflection, feedback, & revision); (d) immersion (engaging with the actual learning materials); (e) integration of knowledge and practice;
(f) collaborative work, e.g., with externals such as scientists or researchers; and (g) use of various vehicles (e.g., workshops, coaching etc.). The PDP was developed in a small-scaled, exploratory study (Van de Pol et al., 2012).
The subject area of focus in the current study is that of social studies. This subject area is a largely untouched area of study; scaffolding research has mostly focused on students’ developing literacy (Van de Pol, et al., 2010). However, scaffolding may also be a useful teaching approach in social studies, as students have difficulties in understanding and using substantive concepts13in this domain (e.g., Kneppers, Elshout-Mohr, van Boxtel, & van Hout-Wolters, 2007; Limon, 2002; Van Drie & van Boxtel, 2008). Van Drie and van Boxtel (2003) showed that students find understanding and using substantive concepts difficult because of their abstract and theoretical nature.
Studies focusing on students’ understanding of substantive concepts showed that generally this knowledge is rather meagre and limited (Van Drie & van Boxtel, 2008).
As scaffolding is expected to be an effective way of supporting students, this teaching method is expected to support students in this problematic area of social studies.
In the current study, we investigated to what extent and in what ways the model of contingent teaching promoted teachers’ supporting behaviour in social studies lessons. The study had an experimental design; about half of the teachers participated in the PDP on scaffolding (scaffolding condition) whereas about half of them did not (nonscaffolding condition). All teachers were prevocational social studies teachers who taught the same five-lesson project on the EU in year 8 (age 12-14). For diagnostic strategies (step 1), checking the diagnosis (step 2), and checking students’ learning (step 4), we had relatively similar hypotheses. These steps are crucial parts of the scaffolding process and are hypothesised to stimulate and facilitate contingent support (step 3). Because these steps stood central in the PDP, we expected firstly that the quantity of these steps would increase (hypotheses 1a, 2a, and 4a).
Secondly, we expected that the quality of these steps would increase in terms of the focus of the steps: a focus on conceptual understanding is considered of higher quality than a focus on facts and we therefore expected a shift towards a focus on conceptual understanding in the scaffolding condition (hypotheses 1b, 2b, and 4b).
Thirdly, we expected that the quality of step 1 (diagnostic strategies) and step 4 (checking students’ learning) would increase in terms of the mode that is elicited:
a more extensive mode of expression (i.e., demonstrations) is considered of higher quality as it provides the teacher with more information on the students’ understanding 13 Substantive concepts refer to “historical phenomena, structures, persons, and periods” (Van Drie & Van Boxtel, 2008). Examples are democracy, internal market, and export.
and it forces the student to elaborate his or her understanding compared to for example a claim of understanding (e.g., “I get it”). Therefore, we expected that the teachers’
elicitations would shift more towards the elicitation of demonstrations in the scaffolding condition compared to the nonscaffolding condition (hypothesis 1c and 4c). We did not have this expectation for step 2 (checking students’ learning): it is in the nature of this step that most of the times quite a short answer will be elicited (e.g., “Am I correct that you understand internal market as the marketplace on Saturday?”).
With regard to step 3 (intervention strategies), we had different hypotheses. The quantity of this step is not the problem; it’s the quality that is often too low (Chiu, 2004).
Therefore, we focused on the quality of this step in terms of contingency. As hypothesised by others (e.g., Chiu, 2004; Webb, 2009), diagnosing students’ understanding will facilitate teachers to tailor their support to the students’ understanding because they have more information about the students’ understanding. Therefore, working with the model of contingent teaching – with an emphasis on diagnostic strategies – is hypothesised to enhance teachers’ contingency. Teachers have more information to:
(1) decide what level of control is needed (i.e., contingency-control) and (2) use and refer to when giving support (i.e., contingency-uptake).
With regard to contingency-control we hypothesised that the teachers who worked with the model of contingent teaching (i.e., scaffolding condition) show a greater increase in their control adaptation than teachers who did not work with the model (nonscaffolding condition) (hypothesis 3a; control). With regard to contingency-uptake, we hypothesised that the teachers who worked with the model of contingent teaching (i.e., scaffolding condition) show a greater increase in uptake than teachers who did not work with the model (nonscaffolding condition) (hypothesis 3b; uptake).
METHODS
Participants
Thirty teachers from 20 schools participated in this study; 17 teachers (of 11 schools) in the scaffolding condition and 13 teachers (of 9 schools; always between 1 and 3 teachers per school) in the nonscaffolding condition. Nine schools were for prevocational education only; eleven schools were also for higher levels of education.
The schools were spread over the Netherlands. Of the participating teachers, 20 were men, 10 were women. The teachers taught social studies in the 8th grade of prevocational education. The average teaching experience of the teachers was 10 years. Each teacher participated with one class, so a total of 30 classes participated.
The mean class size was 27 students and the lessons lasted 53 minutes on average. During the project lessons that all teachers taught during the experiment,
The mean class size was 27 students and the lessons lasted 53 minutes on average. During the project lessons that all teachers taught during the experiment,