I posted recently about my experience with ChatGPT, and how it had helped me to clarify my thinking about the use of the coaching tool “Instruction”.
So when I saw that Google had released their own Large Language Model generative AI chatbot, Bard, I went in search of my next guru.
Once again, the “conversation” helped me to crystallise my thinking around around coaching pedagogies, I found out more about (one version) of Direct Instruction, and I learnt a valuable lesson for coaching and the use of questioning — don’t automatically accept the first answer you are given!
Bard is heavily branded as an “experiment” — generative AI can sometimes produce seriously inaccurate responses, even citing fictitious publications — but I can see how Bard might be considered an advance on previous chatbots, most immediately in that it offers alternative drafts of a response and (for most prompts) also includes a “Google this” link.
It can change its mind, though, as I found out when I persisted with a line of questioning.
Bard gave me more insight into Direct Instruction (upper case D & I) — to be fair, I didn’t ask this question of ChatGPT).
Direct Instruction is a specific teaching method developed by Siegfried Engelmann and Wesley C. Becker. It is a highly structured and explicit teaching method that is designed to teach students specific skills and knowledge…[and]…uses a variety of techniques, including explicit instruction, modeling, guided practice, and independent practice.
Bard, Google
All good. A Google search suggested a reference to a discussion of Games-Based and Direct Instruction Coaching by Brendan SueSee and Shane Pill in The Spectrum of Sports Coaching Styles.
— Is DI appropriate in sports coaching?
No, Direct Instruction (DI) as defined by Siegfried Engelmann and Wesley C. Becker is not appropriate in sports coaching. DI is a teaching method that is designed to teach students specific skills and knowledge in a highly structured and explicit way. It is not designed to teach athletes how to play sports.
DI is based on the idea that all students can learn, regardless of their background or ability, and that a structured and explicit teaching method can help them succeed. However, this is not the case with sports. Sports are not academic subjects that can be learned through explicit instruction. They are physical activities that require practice and repetition to master.
In addition, DI is a very teacher-centered method. The teacher provides explicit instruction and demonstrations, and the students are expected to follow the teacher’s instructions and imitate the teacher’s demonstrations. This is not the way that sports are typically taught. Sports coaches typically provide instruction and demonstrations, but they also allow the athletes to experiment and learn through trial and error.
Finally, DI is a very time-consuming method. It requires the teacher to provide explicit instruction and demonstrations for each skill that the students are learning. This is not possible in sports, where coaches have limited time to work with their athletes.
For these reasons, DI is not an appropriate teaching method for sports coaching.
Bard, Google
I certainly hadn’t expected such an unequivocal response.
And, curiously, Bard did not offer the option to “Google this” after the reply to my question on DI in coaching.
But when challenged to expand on this statement, Bard offered a contradictory reply.
The statement that sports are not academic subjects that can be learned through explicit instruction is false.
While sports are not typically considered academic subjects, they can be learned through explicit instruction. For example, a coach can teach a player how to throw a football or how to run a play. This type of instruction can be very effective in helping players improve their skills.
Bard, Google
And this feels closer to a truth.
Explicit instruction does have a part to play in coaching! Sometimes the only way to get something across to a player is to tell them!
I should state here that I did not go to ChatGPT and Bard with the intention of rubbishing direct instruction (or Direct Instruction) as a coaching methodology.
I do have a strong bias towards an ecological dynamics approach to learning movement skills, but I hope I couched my prompts in suitably neutral terms. I did not ask “why is Instruction a poor coaching tool?”.
It is conceivable that Google’s Bard includes some traces of my own Google search activity in the sources it trawls. Google clearly “remembers” a lot of what I do online, quite possible more than I do myself, and that these could indeed indicate a bias. Although that could be counter-productive in a search engine…unless every prompt has to be prefaced with “tell me something I don’t know about…”!
The Teesra 