This term was developed in the context of informal learning within communities of practice, equivalent to apprenticeship to a trade, or induction into a profession - but I haven't seen it used in relation to students' learning.

I've used situated learning in some of my research. I believe the answer to the question of how to assess is really about what the goals are for learning. So, if your goal for students who are on the periphery is that they can apply X concept to a new scenario, you would assess them on their ability to do just that. The learning outcomes for a course (or other learning experience) should drive the assessment.

" Legitimate peripheral participation (LPP) describes how newcomers become experienced members and eventually old timers of a community of practice or collaborative project (Lave & Wenger 1991). According to LPP, newcomers become members of a community initially by participating in simple and low-risk tasks that are nonetheless productive and necessary and further the goals of the community. Through peripheral activities, novices become acquainted with the tasks, vocabulary, and organizing principles of the community. LPP was then  in later work changed by Wenger (1998) who introduced duality instead, however the term LPP is still widely used in relation to situated learning.en.wikipedia.org/wiki/Legitimate_peripheral_participation

Please look some of my articles in this profile. I have extensively described the use of cognitive apprenticeship. 

Kind regards

NP

I would have to agree with Amy.  Making explicit from the outset the goals of LPP engagement is the trick.  This type of measurement or assessment process doesn't fit neatly into a standardized or systematized reviewing model.  "Benchmarking" for lack of a better term, at least acknowledges that the application skills and the abilities to be analyzed, are evolving under this type of learning condition.  Avoid ignoring the learning "thresholds" that are surpassed while endeavouring to arrive at the learning goal.  When measuring "did you reach the goal?" you might discover that your originally explicit goals either assisted, distracted from, or became irrelevant to, what was actually achieved.  

In my experience, people who are interested in measuring, get a little uncomfortable with the "fuzziness" of "potential outcomes" in these complex (LPP) learning contexts.  The problem is, applying simpler measurement processes doesn't make the learning process undertaken any simpler, so the validity of the measurement is directly connected to the "sample size" taken.

Classrooms are indeed social situations where members are socialised to be "pupils" "members of groups"  or as Reynaldo suggests "test  takers" etc. This often is not the objective of teaching or of formal assessment but much is learnt which may be more permanent.  Research into the aspect of apprenticeship into communities of practice in "online environments"  that Ye Chen mentions may provide interesting answers in a research project. It may be worth comparing the quality of learning of the stated curriculum as opposed to the hidden curriculum. 

My friend Arindam Bose has done some studies, you can read about his work here- https://www.researchgate.net/profile/Arindam_Bose4

https://www.researchgate.net/profile/Arindam_Bose4

I also believe the answer to the question of how to assess is related to expected goals for learning. In Chaklin and Lave (1993) you can found some empirical studies discussing this kind of research context.

no

No

In my limited experiences reading about Lave and Wenger's ideas of legitimate peripheral participation via conversation analyst's perspective (namely, John Hellermann, 2008), I have always thought of legitimate peripheral participation as the degree of active participation and input in discussion. Obviously, there is also the theoretical idea that it is about the gradual shift into becoming a 'legitimate member of the community of practice'. I think both Dean and Reynoldo are right about being explicit and defining exactly what and how to define 'legitimate peripheral participation'. 

I would imagine if you wanted to 'measure' this you will need to first find a way to quantify participation in the dimensions, which you will first have to define. Then you will have to determine how you want to measure learning and what aspects of learning you are going to measure (e.g. memory recall, ability to think and reason etc.). Then you can find a statistical model to examine their correlations. At least, that's probably how I will go about trying to assess students' learning in their legitimate peripheral participation. Pre- and post-teaching session design with recorded observations of classroom interaction to allow you to develop and apply an instrument to 'measure' degree of peripheral participation? 

The late Professor Nate Gage (2009) might be a good start to help you identify ways to quantitatively conceptualise variables of teaching/learning. He has been credited to be the father of modern classroom research, and his 2009 book is a review of classroom teaching that recommends the development of a theory of teaching. 

I think how you are going to measure and assess learning will depend on how you define learning and your philosophical paradigm about what learning is. For example, from a positivistic and cognitivist model of learning you might measure learning based on memory recall and retention. If you consider learning more from a socio-constructivist view than learning becomes a socially situated-phenomenon. In this case, learning is like what Reynaldo have written, learning becomes demonstrated in the way participants display, manage, renew and negotiate their identities and their formations. 

So, it depends on your decisions about how you want to define these. You can also observe your implicit assumptions by reflecting on how you would go about measuring learning. (See Gage 2009, Chapter 2 about different philosophical position and developments in relation to what is scientific)

In my own research using 'Conversation Analysis" to study anatomy laboratory teaching-learning interactions, I have noticed that students tend to default into dyadic (pairs) or triadic interaction despite being arranged in a small group configuration of (7-10 members). Depending on the physical setting and physical positioning of students, learners may 'peripherally listen-in' on neighbouring conversations going on around them. Students in the anatomy laboratory often pay special attention to teacher-student instructional interaction, which they are not active participants in, but they can learn from that discussion as legitimate peripheral participant. I believe in a classroom/formal education institutional setting 'legitimate peripheral participation' may be observed in larger group configurations where it is not possible for everyone to active participate in the dialogue of the interaction. Rod Gardner's work, "When Listeners Talk" is evidence of how peripheral participation may be observed in multi-party instructional interaction. 

Kevin's last paragraph speaks to some of the traditional ways to approximate "position" of the learner within the context of the larger community of identification.  Measuring relativity means exactly what that implies: multi-dimensional, scaled approximation requiring the measurer to determine confidence levels on at least two planes: determination of confidence of the participant and determination of confidence of the assessor. - so rather complicated  As I mentioned, there must be an acceptance of margins of error if you want to gain/see validity in this methodology.

I think what Kevin and Reynaldo accurately point out is that most educators believe that L&W were identifying some of the "procedurals" present in experiential domains.  Today there is a better sense of the impact of relational causality in social learning, so measurements need not be constrained to post hoc assessment tools.  The projective aspects of what learning unfolds (for example: how the participant evolves their definition of valuable capability within context) is dependent on multiple judgements and is never a consensus of one.

Legitimacy is a wonderful thing in its messiness.  Placing limits on that messiness to get a clean picture or measure, kind of defeats the reason one would places value on why learning that inherently involves thinking migration, should now be contained - for the sake of the observer. 

I could go on and on about this, but suffice it to say, the tools to measure matter, but if you limit the measuring to the procedurals, you won't capture the richness of the real take-aways of domain sensitive learning episodes.