Formative Assessment: Macbeth

In Daisy Christodoulou’s Making Good Progress?, (click for SchoolsWeek review) she outlines different ways of using formative assessment in a meaningful way. Inspired by her book, I had a think about how I might apply a couple of the strategies she recommends: multiple choice tests, hyper-correction, and Question Level Analysis (QLA) to develop a formative assessment cycle for my GCSE students studying Macbeth.

Below, I will outline the initial steps I took to test out ways of using her recommendations, describe the experience of running it with students, and then share the resulting adjusted test that I developed as a result of the initial trial, which I intend to  continue developing next academic year in a different setting.

Context: This was an activity trialled at the end of the academic year, after the final summative assessment had been sat by students. The class in question was a low-middle ability Year 10 class of 15 students in a ‘challenging’ school setting. Students are predicted grades 4-6 in English, with current attainment varying from Grades 2-5 within the class. We had been studying Macbeth for four weeks before they sat their final examination, an essay, on the topic, and I didn’t consider that I was able to form a valid judgment of students’ current knowledge of the play based on the assessment. For more on what makes a valid assessment read this.

Initial Steps:

In the preceding lessons, we finished reading the play, and I would quiz students at the beginning of each lesson on the content and analysis we had covered the day before. If the majority of the class were unable to satisfactorily recall the content from the previous day, we would spend time going over that learning, this did result in better engagement in the lessons, because the students quickly cottoned on that I was not going to let them get away with not revising what we had learned. In each set of questions, I would include at least one question focussing on a much earlier scene of the play, or a context question, to model spaced/interleaved practice.

I found a 70-Question multiple choice test that had been published online by a teacher in the US called Maria de Bari (who I tried to contact before writing this post): teach.albion.edu/mcd12/files/2015/04/Macbeth-Final-Test.docx (download). I liked the structure of the test. It’s in three phases: multiple choice knowledge of the play, matching quotes to the character, and selecting interpretations of key quotes.

I issued the test as a ‘pop-quiz’ to students (cruel, in the last week of Summer 2, I know). They sat it, and then we went through the answers: I read out the correct answer for each question, and they self-corrected. I then created a QLA spreadsheet on Excel, and coded correct answers as 1, and incorrect as 0. This then automatically filled the cell with green or red, respectively.

Time: setting up the spreadsheet took about 15 minutes (mainly because I’m not very good at using Excel, and inputting the data was relatively quick, once I’d done the first one. It probably took me about an hour to input 15 students x 70 questions, on a keyboard with a number pad; it would take longer on a keyboard with numbers only at the top.

Analysis: with the automatic red-green coding it was immediately clear which questions had gotten incorrect. I identified the ones where students made the most mistakes, and looked back at the test to see the pattern. For example, it became immediately clear that students didn’t have a clear understanding of Act 3 scenes one and two, and that we needed to re-cap the meaning of dramatic irony. I also realised that the test could be structured much more effectively for QLA. If the questions were chronological, it would be easy to see which scenes/acts need to be revised, and if it were split into context, plot and character that would be useful as well.

Feedback: Once I had done the coding, I shared the feedback with the students, with key points about which aspects of the play students should revise. If I had done this earlier in the term, I would have been able to spend more time individually with students, or prepare a revision pack or teach further lessons based on the gaps.

Student response: The students responded well to doing the multiple-choice test. It’s not something I’ve done much of in my English classroom, but they reported they liked how quickly they could get feedback, that there was a right and a wrong answer (although trickier when it came to interpretations: that was a really useful section, because we had a class discussion where some students strongly disagreed or wanted to argue for other interpretations, and this could lead to extended writing). I was encouraged by the fact that they didn’t falsify their corrections, and were happy to share their scores with the class. I think we all would have liked more time to go through the answers in more detail, and set out a clearer revision plan. I will do this next time.

Future Implementation: This is a formative assessment that I would definitely use more to inform my planning and teaching, as well as to encourage students to take more ownership of their revision. I have already added some additional questions, and divided the quiz into sections to make it easier for QLA. I still need to re-design the Excel spreadsheet, because I’d like to find a way to classify question types to drill down further into the gaps in student knowledge. I’m not yet sure how to do that or if it’s possible (if you’re an Excel whizz, please get in touch in the comments or via my contact page).

When I teach Macbeth at my new school next year, I plan to:

• Continue adding to the bank of questions.
• Select questions from the quiz as we go through the scheme of work, for regular formative assessment. As Christodoulou suggests, I will record the raw scores in my own mark-book. I would use this also a way of modelling gap-analysis, spaced practice and revision.
• Administer the ‘final’ multiple choice test at the end of the unit, before the written assessment, and remind students to use it to inform their revision, by modelling the gap analysis for them.
• Use the quote-interpretation questions as a springboard for further analysis and justification.

Below, you can download the test in its current iteration. I still need to add context and language device questions, and will continue adding plot questions. It’s very much a working document. If you need the answers, get in touch and I will email them to you, but I’ve not included them at present because it’s not yet a finished document and I’m still formatting it. Most questions come from Maria de Bari’s original test, and I take no credit as author of those questions.

Macbeth Test