An examination of teacher questioning in a Year 8 Classics class

Context

This assignment investigates the effects of various questioning strategies on the engagement and attainment of a Year 8 Classics class¹. I became increasingly interested in questioning whilst reading more widely about Assessment for Learning (AfL) for my School Experience Report (SER), and decided that the omnipresence of questioning in every classroom would make it a valuable focus for this report. Building on my understanding of the importance of questioning as an AfL strategy, my aim was to devise a series of lessons making explicit use of different types of questioning and to analyse the relative successes and failures of each approach. This was measured by assessing overall class participation and by specifically studying the responses and involvement of five ‘focus’ students. Profiles of these students are below and each student has been allocated a pseudonym.

The school in which this research was conducted is a London girls’ comprehensive school. Students take Classics from Year 8 and this is taught through a myth-based scheme of work devised by the Head of Department, with the sequence of lessons taught for this report centring around the myths of ‘Perseus and Medusa’ and ‘Orpheus and Eurydice’. The focus class consisted of 30 girls of mixed attainment and engagement and a number of the students in the class are on the ‘Supported Students’ register, something I was careful to consider when planning.

Literature Review

The term ‘questioning’ refers to ‘interrogative utterances’ generally followed by an answer (Dillon, 1981, p. 51). The function of these utterances is variable, but in a classroom context, the control and delivery of them lies almost exclusively in the hands of the teacher (Dillon, 1981, p.51). This means that what should be a ‘two player game’, actually can manifest itself as a ‘deluge of teacher questions’ and a comparatively small number of student questions (Rowe, 1974, p.81; Susskind, 1969, p.146). Stevens (1912, p.15) found that teachers asked an average of 395 questions per day, of which 65% related to the recall of facts. Although this study is outdated, the correlation of these findings with those of Gall (1970) who found that only 20% of questions required students to think beyond recalling facts, and with the more recent research of Brown and Wragg (2001) indicates how little the use of classroom questioning has changed. The effect of this is that students are rarely pushed beyond the introductory ‘knowledge’ level of Bloom's Taxonomy, and fail to engage in more challenging, higher-order thinking (Black et al., 2003). It is therefore unsurprising that students have become accustomed to their role as passive respondents, and teachers are used to ‘controlling’ and initiating discourse (Dillon, 1981, p.51). This ingrained passivity inhibits active participation from the student, and consequently, restricts the potential for teaching and learning which effective questioning could foster (Morgan, 1991, p.7).

This literature review will evaluate the body of research surrounding strategies for effective questioning. It will focus on three chosen areas of questioning research: (i) the use of scaffolding to promote higher-order thinking, (ii) peer questioning, and (iii) wait time.

The use of scaffolding

The importance of stimulating higher-order thinking through effective questioning was first addressed by Bloom's pioneering The Taxonomy of Educational Objectives (1956). His hierarchy of learning categorised the levels at which student thinking occurred with objectives, moving from the simplest level (recall) to the most complex level (creation) (Ramirez, 2017, p.146). The taxonomy was cumulative, building on the successful completion of previous levels before progressing to the next (Haag Granello, 2001), with the additional benefit of offering a clear structure for more difficult learning processes (Ramirez, 2017, p.151).

Despite the clear scaffolding offered by Bloom's hierarchy, subsequent research has found that teachers have struggled to develop the questioning techniques required to move beyond asking recall questions and stimulate higher-order thinking. Although recall questions are widely acknowledged to be an effective method of testing initial student knowledge (Brown and Wragg, 1993, p.14), they fail to build upon or stretch student thinking and can therefore foster passivity (Dillon, 1981, p.53). Brown and Wragg attribute teachers’ failure to move beyond recall questions to an over-reliance on questions arising spontaneously. They therefore suggest planning sequences in advance, to ensure that students reach the highest cognitive levels (1993, p.14). Morgan and Saxton also emphasise the importance of using a scaffold of questions to deepen student involvement with study material, with the ultimate aim of students taking ownership of their learning (1991, p.21). This process of student ownership can only occur at the highest cognitive level (creation), where students understand material well enough to be able to apply knowledge to new contexts and scenarios (Morgan and Saxton, 1991, p.25). The use of pre-formulators (French and MacLure, 1981, p.35) or ‘advance organisers’ (Ausubel, 1978, p.255) provide clues to the answers expected through the question being asked. This scaffolding technique operates on the principle of shared knowledge, which helps direct students towards the correct area of expertise (French and Maclure, 1981, p.35).

This technique is a valuable method for allowing both students and teachers to assess their learning, to ensure that the lower levels of Bloom's taxonomy are consolidated before moving on. This also gives teachers the opportunity to reformulate questions if they initially fail, helping to guide students more systematically towards higher levels of cognition (French and MacLure, 1981, p.38).

However, the effective use of Bloom's taxonomy as a questioning framework relies on a teacher's ability to exhibit flexibility, and to use the answers provided by students to direct their line of inquiry (Morgan and Saxton, 1991, p.9). The process of question-planning should therefore always involve teachers asking themselves ‘What kind of thinking is this question generating?’, as well as a consideration of the value of each question for engaging students in higher-order thinking (Morgan and Saxton, 1991, p.11). This comes with the additional recognition of the need for adaptability, and a willingness on the part of the teacher to alter planned question scaffolds, so as not to inhibit the natural direction of discussion (Morgan and Saxton, 1991, p.9). This is essential for promoting student ownership, which Schaffner notes as increasing student satisfaction, and consequently improving learning outcomes (1983, p.40). An additional difficulty with closely following Bloom's taxonomical structure, particularly when reaching the highest evaluative level, is differentiation and inclusion. Although these higher-order questions are useful for promoting discussion, challenging an individual to take their thinking further can lead to some students ‘taking over’, while other students lose interest (Morgan and Saxton, 1991, p.16; Wragg and Brown, 2001, p.34). Peer questioning offers a potential solution to this issue, and will be considered in the next section of this discussion.

Peer Questioning

Gall and Artero-Boneme (1994) proposed the ‘heads together’ approach as a means for engaging all students, and for tackling the issue mentioned above. This involved students working collaboratively in mixed-ability groups to brainstorm answers and the teacher then addressing questions to these groups, rather than to individuals. King suggests taking this further by encouraging these groups to question each other, forcing students to clarify their understanding through student-to-student interaction (1990a, p.664). The value of peer questioning is widely acknowledged: it forces students to think about material in new ways and externalise their thoughts, particularly when provoked by conflicting peer opinions (King, 1990b, p.134; 1990a, p.666). Dillon supports this view, suggesting that student responses to each other tend to be more complex and in the language of the student, prompting better student connection with thinking and information (1988b, p.154). Peer questioning can be particularly beneficial for lower-ability students, who can model their questions and answers on their more able colleagues (King, 1990b, p.134). However, these learning gains depend upon the mixed-ability composition of working groups.

King and Dillon agree that the value of peer questioning is determined by the guidance offered by the teacher (1990a, p.1991). Although students are generally able to construct generic questions, King found that students provided with explicit questioning guidance gave more elaborate explanations than those involved in unstructured discussion (1990a, p.665). She suggests using question stems as a strategy for promoting critical thinking, and to ensure that students are asking questions at a range of cognitive levels (p.680ff). Graesser and Goodman also adhere to this view, emphasising the importance of probing students about the ‘why’ and ‘how’ of questions to encourage them to form links between past, current and future learning (1985, p.682). Modelling a range of effective questions for students and advocating brainstorming questions before asking them can also help to maximise learning gains from questioning (Dillon, 1981, p.15; Hunkins 1974, p.99), by making students think critically about the content they are questioning. Although this research was carried out in an American context, it seems that this methodology could be valuably applied within a UK context.

Wait Time

Research concerned with questioning is universal in its agreement that increasing the time allowed for student response improves both student and teacher outcomes. Rowe's pioneering study on wait time (1974), found that teachers waited an average of 0.9 seconds before rephrasing the question or asking a different one. This resulted in students having insufficient time in which to formulate an answer, and discourse was restricted to being closed and superficial (Black et al., 2003, p.33; Rowe, 1974, p.81ff.). When wait time was increased to three to five seconds, the length of student responses increased, as did the incidences of speculative thinking and the proportion of students offering responses (Rowe, 1974, p.89ff). Teacher outcomes also improved: the total number of questions decreased, and response flexibility improved, exemplified by the way in which teachers built upon existing conversation (Rowe, 1974, p.91). The results of this study are corroborated by research undertaken by Black et al. (2002; 2003) and Tobin (1987), who emphasises the importance of pausing to encourage the engagement of more students.

Tobin's study on wait time is informed by the work of Winne and Marx, who encourage teachers to allow uninterrupted time for information-processing to occur after a question is asked (1987; 1983). Siegman and Pope suggest that the time allocated should be directly proportional to the difficulty of the question asked (1965, p.524ff). As a result of allowing students enough time to construct a response, Tobin found that the likelihood of students setting up a discussion independent from the teacher increased (1987, p.71). Wragg and Brown also subscribe to this view but distinguish between drill questions, which can be asked quickly, and complex questions, demanding a longer wait time (2001, p.33).

Allied to this is the importance of the response given after the wait time has elapsed. There is a general consensus amongst educational researchers that an effective response should reinforce understanding and offer feedback, whilst also generating continued interest (Brown and Wragg, 1993, p.22; 2001; Purkey, 1978, p.73). Morgan and Saxton are particularly emphatic on this point, and the importance of acknowledging every student contribution to foster future engagement (1991, p.87). This comes with the clear caveat that praise should not be used dishonestly, as this can cause disengagement (Morgan and Saxton, 1991, p.90). Probing and prompting are useful strategies for doing this, helping students to build upon their thinking and simultaneously correct any misunderstandings (Brown and Wragg, 1993, p.33). This not only encourages classroom dialogue, but also promotes higher-order thinking through allowing students to develop understanding, thus making their thinking more explicit.

The body of research surrounding strategies for effective questioning is wide-ranging, and sometimes contradictory. There is however a consensus regarding the general overuse of factual questions and a failure on the part of the teacher to allow adequate time for students to clarify their thinking (Black et al., 2002; Rowe, 1974). This can consequently inhibit potential learning gains, and prevent students from reaching the higher cognitive levels established by Bloom's taxonomy (Ramirez, 2017, p.148). Having a better understanding of these prospective questioning pitfalls should make them easier to avoid in my own classroom practice. The educational value of effective questioning as a strategy for AfL is clear, allowing students to clarify and reflect upon their learning and therefore take ownership of their progress. In order for these gains to occur, it is essential to move away from the traditional teacher-to-student nature of dialogue and acknowledge the significance of student-to-student dialogue in the learning process (Morgan and Saxton, 1991, p.5). The potential worth of student talk is something I will work particularly hard to capitalise on through peer questioning, both for the purposes of this sequence but also to inform my future teaching practice.

Lesson Sequence

I used the strategies and findings discussed in my literature review to inform the planning of my sequence, concentrating on the questioning techniques themselves more than the material being covered. The sequence was taught over three weeks during six lessons of 40 minutes, with each week having a different questioning strategy as its central focus and with wait time as an underlying priority throughout. The first week was based on peer questioning and promoting student-to-student dialogue (King, 1990; Morgan and Saxton, 1991), the second week centred on Bloom's taxonomy and higher-order thinking (Ramirez, 2017), and the final week on peer-brainstorming (Gall and Artero-Boneme, 1994). My sequence was continuous. However, one of my ‘focus’ students was absent for three of the six lessons, potentially reducing the validity of my findings.

Lesson One

This was my first lesson teaching the class, concentrating on the study of Perseus which they had begun the previous week. I started the lesson with a scaffolded question sequence based on Benvenuto Cellini's bronze statue of Perseus.

The sequence began with recall questions, which Brown and Wragg note to be useful for assessing knowledge and instigating thinking (1993, p.14), and built to more open questions regarding what the students felt for Perseus and Medusa.

Who do you think the statue shows?

How can we tell this?

What do you think it is made of?

What is he holding?

What is he wearing on his head/body?

What do you think he is standing on?

How is he made to appear heroic?

What do we feel for Medusa here?

I was careful to allow a wait time of at least three seconds, particularly for these more open questions, to promote higher-order thinking (Rowe, 1974). The preplanned sequence was effective and further responses and thinking were elicited through additional probing questions (Graesser and Goodman, 1985, p.682), but the rigid structure made it difficult to adjust questions reactively to fit with the class conversation. Students Creusa and Juno were particularly engaged with the starter and offered thoughtful responses. However, I feel that if I had offered more positive feedback and paraphrased or repeated student replies, it would have allowed more students to hear and therefore participate in discussion.

The main activity of the lesson involved the students working in small groups to devise a tableau summarising their allocated section of the Perseus and Medusa story. Time was allocated for students to read their section of the story and prepare their freeze-frames, and they were also instructed to brainstorm what their characters were thinking and why (Black et al., 2002, p.6). Black et al. suggest that this can increase participation and foster higher-order questioning (2002, p.6) and was designed to equip students for the peer hot seat questioning activity which happened when they presented their tableaux to the class. Morgan and Saxton cite hot seating as a useful strategy for peer questioning, but warn against allowing it to disintegrate into a series of quick-fire, short answer questions which only reach the lowest cognitive domains of Bloom's Taxonomy (1991, p.120). This was something I hoped to prevent through question modelling. The first group performed their freeze frame at the end of the lesson and the rest of the class were told to think of questions for each character while it was being presented. Juno was particularly forthcoming with hot seating questions, some of which were higher-order e.g. ‘How do you think that Danae felt about being abandoned by Zeus?’ In general, however, it would have been useful to provide more explicit structure for the kinds of questions students should be asking, as advocated by King (1990a, p.666). A ‘hands up’ policy with performers choosing questioners would have reduced the amount of shouting out and chaotic questioning which occurred towards the end of the lesson. In my evaluation of the lesson with the class teacher, we decided that students needed more context for the story than the freeze frame provided and therefore chose to readjust the plan for the following lesson. Overall, the class engagement with questioning and with the lesson content was good: using the ‘focus’ students as a measure of success, students made valuable suggestions regarding the starter image and participated in the peer-questioning activity. The remainder of the groups were scheduled to present next lesson, now to include a short summary of their section of the Perseus story, to tackle the lack of plotline understanding.

Lesson Two

Lesson two was a continuation from the freeze frame activity based on the ‘Perseus and Medusa’ myth, which the students had begun in the previous lesson. The lesson started with a brief recap of the story so far, provided by the group who had performed their tableaux in lesson one. Student Penelope was particularly restless coming into the classroom, so this acted as a good method by which to engage her and prevent possible disruption. Her recall of the story was excellent and I was satisfied with the success of the questioning from last lesson, exemplified by this retention of information. The remainder of the groups were then given a short time in which to run through their freeze frames and remind themselves of their extract of the story.

Following this, the remaining groups were set off to perform their tableaux for the rest of the class. As decided in the evaluation of lesson one, each group was asked to provide a short summary of their section before presenting their tableaux, to reduce possible confusion over the storyline. This worked well, demonstrated by the further higher-order questioning prompted from the students. Students were also provided with model ‘why’ and ‘how’ questions which improved the quality of student-to-student questions, directly correlating with the findings of Hunkins (1974, p.99). The responses offered were more detailed than those provoked in the previous lesson, with student Helen giving a notably thorough response when asked why Andromeda had agreed to go with Perseus. Given the perceptibly minor role of Andromeda in this myth, I was impressed with the well-considered backstory that student Helen had constructed for her. This response was probed by another member of the class, making student Helen elaborate upon and thus clarify her thinking. This reflected King's research, which found that students were forced to reorganise their thinking when confronted with conflicting peer opinions (1990b, p.134). I had been concerned that allowing students to direct their own questions, would lead to some members of the class not fully engaging with their character and the activity. However, this was unfounded, and the class demonstrated effective distribution with little prompting. As a student-led lesson, I was pleased with how it ran and the peer-questioning approach did give the students control over their own thinking and learning, as King suggests (1990b, p.132). The lack of student-teacher interaction made behaviour management challenging, emphasising the worth of questions as a means by which to control the classroom (Morgan and Saxton, 1991, p.42). The peer-questioning structure also made it difficult to monitor the second variable - wait time - meaning that thinking was possibly not probed as far as it could have been for each question and for all students.

Lesson Three

The ‘Perseus and Medusa’ myth remained the focus for lesson three but with an emphasis on the heroic characteristics of Perseus, rather than the myth itself. From a questioning perspective, wait time remained the underlying consideration and Bloom's Taxonomy the primary priority for the week's lessons. In the starter activity, students were challenged to work up a question scaffold closely affiliated to the cognitive levels of Bloom's Taxonomy (and based on the initial stimulus image depicted below).

As a result of last week's evaluation, I worked hard to provide students with more positive feedback and to encourage students to speak loudly to engage as many individuals as possible in discussion. To ensure that students had sufficient time to formulate their answers, I instructed them to brainstorm each question with the person next to them before responding. This was useful for guaranteeing that wait time was a minimum of three-five seconds, as suggested by Rowe (1974, p.81).

The question scaffold was completed successfully, with especially elaborate responses garnered at the synthesis stage of questioning: student Creusa's likening of the image to Kanye West (Perseus/Trump) and Taylor Swift (Medusa/Clinton) was well justified. Student Creusa's ability to synthesise this information was indicative of a strong understanding of both the world affairs event referred to (US election) and the significance of the Perseus and Medusa statue studied in the previous lesson. This was perhaps partly attributable to the clear structure provided by the Bloom's question sequence and the help that this provided in allowing students to subconsciously sequence progressively difficult thinking processes (Ramirez 2017, p.151). When walking around the classroom to monitor student discussion, students Helen and Minerva were absorbed in a heated conversation regarding which way round their chosen celebrities should be. As with lesson two, this aligned with King's research regarding peer-questioning forcing students to reconcile conflicting views and thus further clarify their thinking (1990b, p.134). In hindsight, it would have been beneficial to address a question directly towards these students when the class returned to feedback, rather than calling upon pairs with their hands up.

The next task required students to recall their knowledge of the ‘Perseus and Medusa’ myth by drawing a logo to represent each short sentence. This activity did not offer much opportunity for questioning, but students were set a higher-order extension question asking them to determine the point in the story where they thought Perseus was the most heroic. The plenary activity was designed as a precursor to lesson four, which would consider the characteristics needed to be a modern hero versus those required to be an ancient hero: a series of images was put on the board and students were asked to work out which heroic trait each picture represented, before determining whether the attribute would have been more advantageous to an ancient or modern hero. Broadly speaking, students were able to identify correctly the trait from each image with little additional probing. In some instances, students gave alternative suggestions regarding what the images might represent. These were generally well justified and when questioned further, students were able to rationalise why the attribute would be nonetheless important for an ancient or modern hero. This demonstrated individuals tapping into the ‘synthesis’ and ‘creation’ levels of Bloom's hierarchy, validating the importance of effective questioning as a means for generating higher-order thinking (Black et al., 2003, p.61). Based on the engagement of the ‘focus’ students, this lesson allowed the vast majority of the class to access the higher cognitive levels of Bloom's Taxonomy, exemplified in their responses to the starter and plenary activities.

Lesson Four

Having set up the idea of modern heroes and ancient heroes and their different skillsets in the previous lesson, students began lesson four by completing a quick ‘modern versus ancient hero’ table activity (below).

While circulating to clarify understanding, I realised that there was some confusion concerning what constituted a modern hero: fictional heroes or real-life heroes. This is something I should have clarified from the outset, but allowing students to produce their own definitions was perhaps valuable in encouraging them to access the ‘creation’ level of Bloom's Taxonomy. Students were then asked to feedback their responses to the rest of the class, before being asked whether there were characteristics which applied to both modern and ancient heroes. Student Helen was quick to respond to this, suggesting that some attributes were shared but had more value to a modern hero than an ancient hero or vice versa. When probed further, she substantiated her argument by proposing that the ability to fight monsters would be more useful to an ancient hero than a modern hero, unless the modern hero was a film character in a specific scenario, citing Percy Jackson as an example of this. I also used this as an opportunity to ask the rest of the class whether or not they agreed, ensuring that they did not lose interest: a side effect of probing which Brown and Wragg warn of (1993, p. 20).

Following this discussion, the class watched a video on the making of the Cellini statue they looked at last week and were asked to consider the significance of the final line of the video:

So Cellini has one bit of mischief to play out at the expense of Michelangelo – positioning the sculpture where it would seem it had caught David's attention. The petrifying gaze of Medusa turns David into cold, lifeless stone.

Student Penelope immediately raised her hand to point out the irony of Medusa turning people to stone, positioned looking at Michelangelo's stone ‘David’. This demonstrated the value of a video mini plenary with a clear focus question (Lawson et al., p. 2006): student Penelope has a tendency to become quickly disengaged and the video acted as a useful ‘resetting’ device for her and other, similar characters within the group. Having resettled the class, I worked through another Bloom's Taxonomy-based question sequence considering the relative worth of the heroes displayed on the board and their shared characteristics (see below).

Student Minerva correctly recognised all of the heroes on the board in the ‘knowledge’ stage of questioning. In the ‘application’ question stage, students were quick to identify that all of the men on the board were white and that this would not be the case for modern heroes. However, students became restless and therefore struggled to reach the ‘synthesis’ and ‘evaluation’ cognitive levels. I should have realised this earlier and terminated the activity as the class began to get noisy towards the end of the hierarchy. This demonstrated a possible conflict between stimulating higher-order thinking through questioning and behaviour management, emphasising the value of balancing questioning with independent work.

As a final activity, students wrote a short passage from the perspective of Heracles, advising Perseus on how he could be more heroic. Student Penelope struggled with this due to a lack of prior knowledge about Heracles, but was able to recall some information about his labours when prompted individually. If there had been more time available, I would have encouraged students to brainstorm their ideas about Heracles before beginning the task, to help clarify and organise their thinking.

Lesson Five

Lessons five and six focused on encouraging peer-brainstorming as a method for clarifying thinking before responding: a technique which Black et al. also suggest as being useful for promoting class participation (2003, p.6). Peer-brainstorming also had the additional benefit of automatically enforcing a longer wait time, thus helping to fulfil this overall sequence aim (Rowe, 1974). As students arrived in the classroom, they were immediately set off to brainstorm their knowledge of the Greek underworld with the person next to them. Having the activity on the board as students were coming in settled the class and also ensured that the aims of the activity were shared with students, encouraging them to take control of their own learning: King cites this as a fundamental objective of peer-to-peer learning and this is also an important part of assessment for learning, for which questioning is a central strategy (King, 1990b, p.134; Black et al., 2002, p.6ff). After discussing their prior knowledge in pairs, students were called upon to contribute their ideas to a whole-class brainstorm on the board. This elicited a significant number of responses, with students Juno and Minerva demonstrating an understanding of the role of Hades in governing the underworld and knowledge of the existence of Charon, although neither was able to name the boatman. I spent too long unsuccessfully probing these individuals on this issue, and came close to potentially losing interest from the rest of the class (Brown & Wragg, 1993, p.20). When I realised this was occurring, I swiftly moved on to the next activity, which involved reading through an abridged version of the story of Orpheus and Eurydice, with short recall questions asked at various points to clarify understanding. This activity resettled the group and the majority of students were able to answer the knowledge-based questions when called upon, which indicated both an understanding of the story but also of the class staying on task. The lesson plenary was a task designed to elucidate higher-order thinking from the group, by asking them to write a short response to the question ‘Who do you feel more sorry for – Orpheus or Eurydice?’ Before setting students off to work individually, students were asked to briefly discuss the question in pairs and then to contribute to a class table in the board. This acted as both as scaffolding to help students write their answers later on, but also as a means of encouraging students to test and challenge their thinking on their peers (King 1990b, p.134). Student Penelope responded particularly well to this approach, concluding that she did not feel sorry for either characters: Eurydice seemed weak with little self-respect and Orpheus showed a lack of resolve by turning back as he left the underworld. I was impressed with the detail of her response and her consideration of the characters in the context of the story. More generally, the table produced by the class was balanced with well-justified responses elicited from the majority of respondents.

Lesson Six

The conditions for lesson six were not ideal due to the absence of three of the ‘focus’ students (students Juno, Minerva and Helen). This lesson continued with the peer-brainstorming emphasis from lesson five, but students were given less time in which to discuss each question with their partner. A short summary video was shown as the starter and students were asked what similarities they noticed between what they had read in the previous lesson and what they saw. This drew upon the knowledge level of Bloom's taxonomy and is mentioned by Brown and Wragg (1993, p.14) as a useful way of assessing existing knowledge at the beginning of a lesson. The class quickly picked up on the lack of snakes in the video, which had caused the death of Eurydice in the version of the myth read previously.

The main activity continued to draw on this idea of myths being represented in alternative ways, asking students to look at images (Appendix G) depicting different sections of the story and to consider two key questions in their pairs:

• – Which section does the image represent?

• – Do you think the image is effective?

Images 1 and 2 were similar, but students were able to differentiate between the two images by acknowledging the dark colours used in the second image to denote the setting of the underworld and the light colours in the first image to reflect the living world above. Students required some probing to consider aspects of the images such as colours, the expressions and body language of the characters, and objects in the scene (Wragg & Brown, 2001, p.33ff). When directed in this way, their analysis of the pictures became more detailed and Penelope made a well-considered point regarding the polarised lighting in Image 3: the light on Orpheus’ side of the picture showed him entering the living world, whilst the darkness on Eurydice's side reflected her remaining in the underworld. Creusa also made a valuable point regarding colours in Image 3, acknowledging that Eurydice goes from white to translucent, perhaps as a physical manifestation of her fading away and becoming a ghost. Although this lesson required some additional direction from me, I was generally satisfied that the lesson content was widely accessible. The use of peer-brainstorming and the wait time this naturally allowed, ensured that the majority of the class were able to engage with the higher levels of Bloom's taxonomy, exemplified in the complex responses given.

Conclusion

Overall, I was pleased with how the lessons ran as a sequence and I felt that my questioning did improve as the sequence progressed. This is particularly true with regard to wait time and I consistently ensured that I allowed at least the three-five seconds which Rowe suggests as optimal (1974). However, there were certainly areas of each lesson which I would rethink if teaching them again and providing additional positive feedback was a recurring target for improvement throughout the sequence. This is something I will look to concentrate on more in my future teaching.

With regard to the chief weekly focuses I devised my lesson content around (peer-questioning, Bloom's Taxonomy and peer-brainstorming), the second lesson each week was broadly speaking the more successful at subscribing to these predetermined strategies. This was due to the valuable time spent after the first lesson evaluating my questioning practice and establishing specific areas for improvement. The contrast was particularly noticeable in the tableaux activity which bridged lessons one and two: I provided model questions for the second lesson as a result of feedback from the first, which gave the students a clear guide for generating their own questions. This is something which King strongly suggested doing (1990b, p.131) and the use of this strategy produced a greater number of higher-order responses. I also realised that the students required more direction regarding how best to distribute their hot seat questions and therefore established a ‘hands up’ policy, which worked well in this student-to-student questioning environment (Brown & Wragg, 1993, p.19).

Lessons three and four failed to follow this trend. Lesson three was palpably the most successful lesson of the whole sequence, allowing students to access the highest level of Bloom's Taxonomy ‘creation’ (1976), with the use of brainstorming naturally ensuring that wait time was extended, thus helping students to reach these higher cognitive levels. Lesson four posed the most significant problems: the students were already restless upon entering the class and rather than trying to persevere with the Bloom's question scaffold as planned, I should have abandoned this in favour of independent work or distributed the questions more effectively (Wragg & Brown, 2001, p.31). For me, this reinforced that the value of questioning is determined entirely by the extent to which students engage with it. In this case, questioning was acting in opposition to behaviour management and, in future, I will try to adapt my lessons more swiftly and reactively to prevent any potential loss of control.

From a data collection perspective, this lesson sequence had multiple issues. It was virtually impossible to gauge the engagement and attainment of the whole class with questioning as the primary focus due to the limited amount of written work involved, but also on account of the sheer size of the class. This was countered to an extent by the use and careful selection of ‘focus’ students, but these students could still not represent the views of a class of 30. The validity of using ‘focus’ students was also decreased due to absence, which affected the pool of data gained from each lesson. With a different group, I would perhaps have devised a questionnaire or exit ticket to assess student learning gains at the end of each lesson. However, this was unrealistic both due to the number of students and their age.

The teaching of this sequence has confirmed the incontrovertible value of questioning as a strategy for AfL, as the literature had already suggested. The potential it offers for allowing students to take control of their own learning, a fundamental facet of AfL, and the variety of questioning techniques for doing this, appears unrivalled by the other strategies for AfL previously studied as part of my SER. Although this study only evaluated a small number of questioning techniques, whose execution will require significant refinement, I believe that, overall, learning gains were made and students were engaged. I will seek to build upon these findings and my own questioning in my future teaching practice.