Teaching recurve at university: Zhang 2023 study

Two classes of twelve university students in Beijing, six men and six women each, none of whom had ever shot and none enrolled on a sports degree, took an eight-week elective recurve course in the autumn of 2022. The content, the teacher, the timetable and the equipment were the same; what differed was that one class also had an online component, with materials, homework and discussions on a teaching platform before and after the session on the field.

In the final test, ten arrows at 10 metres on a 60 centimetre target face, the class with the online component scored 46.0 points out of 100 against 36.5 (p = 0.031), and received a technical mark from the teacher of 13.25 out of 15 against 11.25 (p = 0.046). It also grouped its arrows earlier over the weeks, on an indicator that records when at least four arrows out of six first land within a 25 centimetre radius (8.75 against 11.00, p = 0.027). In a mock competition, where the students took turns as judges, the online class made twelve rule errors against twenty-three.

One result goes the other way: assembling and dismantling the bow took the online class 96.8 seconds on average against 82.9 for the other (p = 0.041).

The two classes were formed through the university's enrolment system, not by drawing lots: the first had its lesson on Wednesday afternoon, the second on Thursday at the same hour. The course ran ninety minutes a week for eight weeks, sixteen hours in all, on a standard school range; the men used 18 pound bows and the women 16, and the teacher tried to keep the number of arrows shot equal in the two classes.

The experimental class followed a structure built on Gagné's nine-stage model, with three moments each week: materials and questions online before the lesson, practice on the field, then homework and online discussion afterwards, up to the next lesson.

There are four measures, all collected by the teacher. The score: ten arrows at 10 metres, three hundred seconds allowed, arrows shot late scored zero, scoring read by hand on the target. The technical mark: five elements of the movement, stance, draw, alignment, aim and release, three points each, with fifteen points for anyone without appreciable errors. The grouping indicator, measured in units of half a week of course. The time to assemble and dismantle the bow, timed twice with a ten second penalty for each correction by the teacher, keeping the better attempt. Comparisons are independent samples t tests; the article also reports platform data and a satisfaction questionnaire, collected however in the experimental class only.

Twelve people per class, no allocation by lot and no baseline measure: the two classes are compared only at the end, and nothing says they were equivalent at the start beyond sex and the absence of experience. The teacher is also the author of the study, knows the hypothesis, teaches both classes and personally assigns both the score and the technical mark, which is a judgement on five elements of the movement. For an outcome of that kind, having no independent assessment weighs heavily.

There are four measures, plus the count of rule errors, and they are compared one by one with no correction for the number of comparisons; with twelve people per group and p values between 0.027 and 0.046, it takes little for the picture to change. The count of errors, twelve against twenty-three, was not tested at all.

The grouping indicator is confusingly defined: the course lasts eight weeks, yet the values reported are 8.75 and 11.00, because the scale counts half weeks, and the worked example in the text does not clarify the conversion. Finally the test is at 10 metres over ten arrows, the whole experience concerns absolute beginners on a university elective, the platform data and the satisfaction survey come from the online class alone and have nothing to be compared with, and the work is a conference contribution whose review process is not clear.

This concerns teaching, not performance: it is about what happens when a well-made online component, with materials to go through beforehand and discussion afterwards, is added to a beginners' course. In those two classes in Beijing, the students who had it shot better, received a higher technical mark and grouped sooner; the numbers are small and the comparison is fragile, but the direction is the one an instructor would expect, because the session on the field starts with students who already know what they are about to do.

The exception is the most interesting part. Assembling and dismantling the bow, a fixed procedure learnt by watching and doing, went worse in the online class, and the authors note that those students were slower and had to be corrected more often. For anyone teaching, that is a useful reminder: what is gained in understanding is not automatically gained in handling, which needs hands and presence.

This study says nothing about archers who are not absolute beginners, nor about what remains of those advantages once the course is over.