Should I Provide Assignment Examples and if so, What is the Best Way to do so?
This article explores whether providing past students’ examples of assignments and assessments is effective as a means of improving student marks by conveying work quality expectations.
Students frequently ask for assignment and exam examples to help them determine what they need to do. Instructors often fear that providing examples of past student work (with permission, and with personal information removed, of course) will unduly constrain students because many will simply do their own version of what they see rather than what they could imagine if they had to figure it out on their own. This is similar to a criticism of grading rubrics as making assignments too formulistic and process-focused, thus stifling creativity.
Turns out that research shows that under certain circumstances, providing exemplars does improve student performance, so the article also provides guidance on what works best should you decide to use exemplars.
Optimal Way of Providing Exemplars
Research suggests that the approach resulting in the most positive improvement in student marks is to provide three examples of varying quality (low, medium, and high), in conjunction with marking sheets or answer keys or some equivalent type of indication of where things went astray, such as comments.
Variations on this theme are providing two exemplars, borderline and excellent, with a marking sheet (Rust et al. 2003); and providing one exemplar for each grade level from borderline to high distinction (Newlyn and Spencer, 2009; Wimshurst and Manning, 2013).
With all of the approaches outlined above, there was a significant improvement in marks.
A critical part of most of the studies outlined below is that there was an associated instructor-organized activity that got students involved in applying or critiquing the application of the marking scheme, rubric or comments and discussing the results. Helping students and instructors come to a common understanding of what is expected is the key. The exemplars and marking schemes and discussion are effective tools for doing so.
Only those study results with control groups that compared student marks are included. All had student perceptions surveys, which are fine, but I wanted indication of actual student marks improvement. One study, Hendry, Bromberger and Armstrong (2011), had wonderful methodology and great insights from student perceptions survey results, but there was no statistically significant improvement in student performance. Hence the “research suggests” and not “research indicates.”
Kean (2012) gives a good example of the optimal three exemplars approach. She used this method at Leeds Metropolitan University in her first year Journalism, Research and Writing assignment worth 100%, done in stages. There were 41 students. Summative assessment for this module was an exam in which students had to write a press release from a brief, plus one other piece of specialist writing. A mock exam was given in preparation. It was returned in class along with answers by the previous year’s students – a good (68%), a middling (52%) and a fail (35%) – but without the marks on the exemplars. Students graded the exemplar papers and used the same feedback/marking sheet to explain their marks as had been used for their own papers. The average mark on the module final exam was 56% compared to 50.8% the previous year, and there were fewer failures: 2 out of 41 (or 0.05%) compared to 9 out of 59 the previous year (or 0.15%).
Scoles, Huxham and McArthur (2012) used qualitative and quantitative analysis to report on a project they had undertaken to measure the effectiveness of using three annotated exemplars (average 50-60%, good 60-70% and excellent 70%+) to improve the exam performance of their students at a Scottish University. Undertaking a statistical analysis between those students who accessed the available exemplars (available online) and those students who did not access them showed a highly significant difference in the mean marks students achieved in the final exam (54.8 vs 48.7). (Of 520 students, 397 accessed exemplars with feedback, and 123 did not.)
Newlyn and Spencer (2009) conducted research on using exemplars with over 2,500 students at an Australian university over a two-year period in an attempt to improve the students’ final exam performance in introductory law. (For some time, the introductory law unit had seen relatively good interim assessment marks but very poor summative exam ones). A series of exam exemplars from previous years’ work representing the full range of grades (Fail, Pass, Credit, Distinction, and High Distinction) was provided to students during the teaching semester. They were annotated to reflect the published criteria and standards. The researchers concluded that the provision of exemplars led to an increase in students marks (“before” medians of 24.2 and 26.4 vs “after” medians of 37.0 and 37.9), particularly at the highest grade levels.
Wimshurst and Manning (2013) conducted exemplar research on 97 students in a criminology course at an Australian university. Eighty-seven students from the previous year‘s course were the control group. Researchers provided six exemplars from previous students of grade levels from borderline pass to high distinction. The researchers took into account various factors such as gender, age, mode of study (on campus or distance based) and grade point average (GPA). They concluded that not only did exemplars improve students overall grades for the unit (by 7%), but they also enhanced student interest in and involvement with the subject.
Rust et al. (2003) provided first-year business students at Oxford Brookes University with a marking sheet, a borderline exemplar and an A grade exemplar, and asked students to mark the assignments individually. In an optional 90-minute ‘workshop’ held one week later, students discussed and shared the justification for their marks, and staff explained how the criteria and standards were applied to the assignments. Students who participated in the optional workshop (approximately half of just over 300 students, for two years in a row) subsequently achieved significantly better results in their own summative assessments compared with those who did not attend.
Hendry, Graham D.; Bromberger, Nikki; and Armstrong, Susan. (2011). Constructive guidance and feedback for learning: the usefulness of exemplars, marking sheets and different types of feedback in a first year law subject, Assessment & Evaluation in Higher Education, 36(1), 1 — 11. URL: http://dx.doi.org/10.1080/02602930903128904
Kean, J. (2012). Show and Tell: Using Peer Assessment and Exemplars to Help Students Understand Quality in Assessment. Practitioner Research in Higher Education, 6(2), 83-94.
Newlyn, D. (2013). Providing Exemplars in the Learning Environment: the Case for and against. Universal Journal of Educational Research 1(1), 26-32. DOI: 10.13189/ujer.2013.010104
Newlyn D, Spencer L. (2009). Using Exemplars in an Interdisciplinary Law Unit: Listening to the student's voices, Journal of the Australasian Law Teachers Association, 2(1 & 2), 121-133.
Rust, C.; Price M.; and O’Donovan, B. (2003). Improving students’ learning by developing their understanding of assessment criteria and processes. Assessment & Evaluation in Higher Education 28(2), 147–64. Doi: 10.1080/0260293032000045509
Scoles, J.; Huxham M.; & McArthur, J. (2012). No longer exempt from good practice: using exemplars to close the feedback gap for exams. Assessment and Evaluation in Higher Education, 38(6) DOI:10.1080/02602938.2012.674485
Wimshurst, K. & Manning, M. (2013). Feed-forward assessment, exemplars and peer marking: evidence of efficacy. Assessment and Evaluation in Higher Education, 38(4), 451-465.