Weaver & Cohen report that "no empirical evidence has yet been provided to determine the best overall method for conducting strategy training" (1998: 71), though they identify three main instructional frameworks designed to raise student awareness about strategy use, to give students opportunities to practice the strategies that they are being taught, and to help them understand how to use the strategies in new learning contexts. The first of these, by Pearson & Dole (1987), shows a number of similarities to the "3Ps"[8] approach to language teaching (cf. Skehan 1998:94, White 1988), teaching specific strategy use as lesson content through modelling, explanation, functional practice, and transfer of the strategy to new learning contexts:

The second framework identified by Weaver & Cohen is that of Oxford et al. (1990), who  include strategy awareness, self-evaluation and self-monitoring of language performance, in a more student-centred approach:

Chamot & O'Malley (1994) provide the third framework, which is more process-oriented and is especially useful after students have already had practice applying a broad range of strategies in a variety of contexts. This approach offers a four-stage problem-solving process in which learner-training is part of language learning:

1. Planning: The instructor presents the students with a language task and explains the rationale. Students are asked to plan their approaches to the task, choosing strategies they feel are appropriate.
2. Monitoring: During the task, students are asked to 'self-monitor' their performance.
3. Problem-solving: Students are expected to find their own solutions to problems as they arise.
4. Evaluation: Learners are given time to 'debrief'. (Weaver & Cohen 1998:73)

In addition to these frameworks, a number of guidelines for strategy training have been proposed (e.g. Wenden 1987a; 1991b; Chamot & Kupper 1989; Dickinson 1992; Huang & van Naerssen 1987; Oxford 1989;1990b;1996a). Chamot & Kupper draw attention to the necessity for preparation, stressing strategy-needs analysis in their six steps to learning strategy instruction: i) identify students' current strategies; ii) assess their strategy needs; iii) plan strategy instruction; iv) teaching strategies; v) evaluate strategy use; vi) help students transfer strategies to new tasks (1989:19).

A number of methods of assessing existing learning strategies have been used over the past two decades: informal observation and discussion (cf. Rubin, 1975); think-alouds (e.g. Cohen & Hosenfeld 1981), interviews (e.g. O'Malley, Chamot, Stewner-Manzared, Kupper, Russo 1985), questionnaires (e.g. Politzer 1983; Oxford 1986, 1989), computer assessment (e.g. Jamieson & Chapelle 1987), and product analysis (Vann & Schmidt 1993; Abraham & Vann 1994). Oxford (1996a) compares a number of strategy-assessment methods, as in table XVII, below: 

TABLE XVII: COMPARISON OF STRATEGY-ASSESSMENT TYPES (OXFORD 1996A:35-36) 

From this table, Oxford infers a number of implications for research and instruction:

1. Language researchers must conceptualise language learning strategies in a way that includes the social and affective sides of learning.
2. Through strategy assessment, teachers can help their students recognise the power of using language learning strategies for making learning quicker, easier and more effective.
3. Based on the information from the strategy assessment, teachers can weave strategy instruction into regular classroom events in a natural, comfortable, but explicit way (cf. Chamot & Kupper 1989; Oxford 1990b; O'Malley & Chamot 1990). Teachers must also consider differences in motivation, learning style, gender, and other factors that affect learning strategy use.
4. Teachers need to be judicious in their selection of strategies to use in instruction.
5. Strategy assessments using different measurement modes with the same sample of students could be cross-related. This would contribute to the validity of various assessment techniques.
6. Studies need to be replicated so the more consistent information becomes available within and across populations. (Oxford 1996a:36-37)

Wenden (1987b:159) suggests that explicitness of purpose, content, and evaluation be taken into account in the development and implementation of activities and materials for learner training, and offers "tentative guidelines for teachers and administrators who wish to provide students with more systematic training in learning how to learn":

1. Inform students of the value and significance of the strategies you train them to use.
2. Provide training in both cognitive and metacognitive strategies. Training in metacognition should include both awareness raising or reflection on the nature of learning and training in the skills necessary to plan, monitor and evaluate learning activities.
3. To determine how to integrate learner training with language training, take into account the following factors:

1. Range and specificity
2. Autonomy of application
3. Learners' needs

1. In evaluating learner training it is important to consider the following:

1. Learner attitudes: has learners' appreciation of learner training changed?
2. Skills acquisition: has the learning skill been learned?
3. Task improvement: does the skill facilitate performance of the language task?
4. Durability: does the skill continue to be utilised"
5. Transfer: is the skill utilised in similar contexts? (Adapted from Wenden 1987b:166)

Dickinson (1992:23) derives an "IDEAL" procedure for learners to select and monitor their use of learning strategies (Identify, Define, Explore, Act, Look), which is mirrored in Williams & Burden's (1997:165) fundamental questions for learners:

1. What do I want to achieve? (Identify)
2. Do I know exactly what I am doing? (Define)
3. Why am I doing it? (Explore)
4. How will it be of value to me? (Explore)
5. Which strategies shall I use to achieve my aim best? (Act).
6. How do I tell when I have succeeded? (Look).(Adapted from Williams & Burden 1997:165)