Research Bites: There is No Best Method – Why?


Prabhu, N. S. (1990). There is no best method—why?. Tesol Quarterly, 24(2), 161-176. [link]

Twitter Summary

Prabhu: there is no best method. Besides, understanding learning and teaching, engagement, PD more important.


Prabhu examines the thought terminating cliche that “there is no best method.” You can hear this phrase often when people who follow one or more different methods discuss teaching approaches. The phrase is often used as a dismissive or defensive statement which typically hampers more thoughtful pedagogic discussion – instead of discussing the exact reasons why there is no best method, this statement basically ends the conversation with a lot of sheepish head-nodding. The fact that most teachers would agree with this statement without ever having analyzed it may be a problem. Therefore, Prabhu examines the reasons why this trusim is said, and then offers a concept outside of the realm of methods that may be of more use to teachers. His ideas are quite dense, so I thought it would be best to summarize them as an imagined conversation between teachers.

Note: there is also an alternative, more concise explanation at the bottom of this post.

Conversation 1

B: There is no best method because it all depends on context.
A: OK, but doesn’t that mean there is a best method. You know, a best method for each context?
B: No, because in those contexts, there are still many methods that have to be used to meet the needs of the learners. One method can’t do it all.
A: Isn’t it a bit difficult to identify which variables actually affect learning, and therefore which methods to apply? It seems a bit random.
B: We can do needs analyses, surveys, questionnaires, and things like that to determine what variables are important and how to apply the method.
A: Isn’t it a bit difficult to determine which approach to take with a specific variable? I think more research is needed on just what affects and does not affect learning, and to what degree.
B: I agree, that’s why context matters. We have to consider everything when applying a method.
A: Don’t you think it may be more useful to concentrate on similarities between groups and not focus on the dissimilarities? You know, if a doctor had to consider every little physiological variation, from heart rate to fingerprints, we’d all be dead. The body is the body, and learning a language is learning a language. Are you saying that this fundamental human trait differs so much that something can’t be applicable in many different contexts?

A: Yes, because it all depends on the context.

Conversation 2

C: There is no best method because there is some truth to every method.
A: Yeah, that makes sense, except, who decides what part is the truth?
C: I think the teacher can make that decision.
A: OK, but then this ‘truth’ becomes a method right?
C: Exactly!
A: So, then, there is only some truth to this new method, right? If you claim that the other methods contain some partial truth, then this new method must contain a partial truth, too, right?
C: No, because the teacher takes only what is true from the original method and then ends up with only the truth.
A: But just because you have taken what you consider to be true and blended it up with other truths doesn’t mean your method is fully true. Won’t it have holes and half-truths too? It seems a bit haphazard to me. It’s seems like you mixed what you thought was good in the hope that it would be effective for learning, or in the hope that it would cover all aspects of learning.
C: No. The best part of each method is taken and put together so that it is much more effective than any other single method.
A: pedagogical understanding
A: I see, so you’ve kind of created a new, ecclectic method?
C: Yes.
A: But, there is some truth to every method, right?

Conversation 3

A: There is no best method because we don’t know what ‘best’ really means.
D: Sure we do. The best method is the method that brings about learning the quickest.
A: But, it seems like objectively measuring a method is very difficult. We’re measuring people, not chemicals. There are lots of unseen variables involved. And besides, its very difficult to establish a direct link between teaching and learning because learning isn’t that simple – it’s quite complex.
D: Yes, but there is a lot of research on methods, and the research clearly shows which methods are effective.
A: Most research shows possible correlations, but nothing is ever definitive. But, OK, let’s suppose there is a best method. Then, in order for it to be effective, a teacher would have to follow the method exactly, not deviating even once from what is essentially a highly specific teaching routine. If they do deviate, the method they are using is no longer the empirical method proved by research. So, the teacher must be very specific in applying the method. In fact, it seems like the teacher doesn’t even matter.
D: That’s absurd!
A: Exactly.

What Prabhu suggests is that a teacher’s sense of plausibility may be a more important factor than the methods one chooses. A higher sense of plausibility will be linked to a more engaged and thoughtful teacher, and this is likely to be more effective than a teacher who simply adheres to a method. Plausibility here means “teachers’ subjective understanding of the teaching they do. Teachers need to operate with some personal conceptualisation of how their teaching leads to desired learning” (p. 172). In other words, teachers have to ask themselves ‘How plausible is it that my teaching can influence learning?’ A teacher with a high sense of plausibility asks themselves these questions and is likely confident in their teaching and more engaged in their classrooms.

This conceptualization is most certainly subjective and will vary, but the point is that knowing how one’s teaching is linked to learning, as well as the extent to which one is involved in their own “real” and “active” (as opposed to “mechanical”) teaching, is a crucial factor in teaching effectiveness. The alternative is a teacher who simply follows a method because it is true or made of partial truths, little concerned with why or how it is true. Particular attention should be paid to the terms Prabhu uses: “real” and “active”. Prabhu says that “The enemy of good teaching is not a bad method, but overroutinisation” (p. 174). If you are engaged in your teaching – meaning you are not mechanically stating grammar rules or asking students to complete exercise 1 and then exercise 2 and then exercise 3 – if you are “real” in your approach to teaching, your teaching will be effective.

Ask yourself this: Who would you rather have as a teacher – one who is enthusiastic and fully believes in the audolingual method or one who is both not confident and somewhat robotic about applying tasked-based learning, CLT or even conversation-driven learning (Dogme)?


Prabhu claims that, in order to develop and maintain a high sense of plausibility, interactions and discussions between teachers is necessary. One way to imagine how this works out in the real world is focusing on professional development activities: workshops, conferences, reading, writing, and Tweeting! In an interview with The Teacher Trainer Journal, Prabhu himself suggested an open-house for the sole purpose of discussing teaching. This forces the teacher to constantly think, rethink, and adjust their teaching. In addition, interaction between teachers is likely to lead to a lot of useful advice, tips, and ideas that can be applied in the classroom – helping teachers break whatever routine they may have accidentally stumbled into.

Alternative Explanation

  1. There is no best method because it all depends on context.
    • This means that each context has a best method.
    • It is very difficult to determine the relationship between numerous variables and appropriate methods.
    • Catering to every possible need means creating a methodology that cannot be flexible, otherwise one of the variables of learning may not be met and the whole package falls apart.
    • Focusing on a theory or practice that is more general, rather than specific, may be more beneficial. After all, the language learning process itself cannot be overly dissimilar across contexts. Or, in Prabhu’s metaphor: “If all physiological variation among individuals (including fingerprints) were assumed to call for matching differentiation in medical treatment, no medical practice would be justifiable” (p. 164).
  2. There is no best method because there is some truth to every method.
    • This means that the methods that do exist are not actually complete but only contain partial truths.
    • Believing this means making an eclectic mix of methods based on what is perceived to be true, or just randomly hoping what is conglomerated actually works.
    • In this view, the eclectic method is a method in itself, and therefore likely only contains partial truths.
  3. There is no best method because we need to rethink what “best” really means.
    • Objective evaluation of a method is quite difficult.
    • The classroom is too complex and environment to without a doubt state one method works better than another.
    • It is also difficult to demonstrate a direct link between teaching and learning because learning is a very complex process.
    • Even if we were to completely verify a method, it means we would have to follow that method precisely in order to reap its benefit. The method would be nothing but a routine – one which anyone could complete.

2 thoughts on “Research Bites: There is No Best Method – Why?

  1. Ana says:

    Excellent post! It does a great job explaining the dense ideas presented in Prabhu’s article. Thank you for sharing.

    • Anthony Schmidt says:

      Thanks! Even after parsing through Prabhu carefully, and after writing this, I still find it a bit overly dense!

Comments are closed.