3/23 Laura: Science is Social

I really enjoyed both readings this week as they highlighted the communicative element of scientific practice, a factor which I think is integral and often under-valued.  I think that both Reiser et al and Sampson and Gleim offered powerful examples of learning communities and the importance of developing and revising explanations with the help of peers, the teacher, and diverse resources.  The ability to create sound explanations, which articulate causal mechanisms and account for observed data, is a powerful tool in all subjects.  By supporting our students in developing strong logic in the sciences, we can help them become better writers and more critical consumers of information, goals which I think we all have as teachers.   To develop this logic, Reiser et al suggest highlighting the distinction between observation and interpretation.  I think this is an astute decision as here you can get to the root of student’s misunderstandings of what is ‘knowable’ from an observation, and gently steer them away from assuming, projecting or anthropomorphizing by encouraging metacognition.  The ExplanationConstructor journal is a valuable tool in this metacognition as it allows both the student and the teacher to trace the student’s thought process through notes and the decision tree.  This journal is reminiscent to me of my lab notebooks working in marine science labs, and I hope to offer my students a similar space, either through BGuILE or through written observation journals, where they can work through inquiry and data and be able to visualize the experience later, from which they could glean more sophisticated and personal correlations and theories.  

            I was also very excited that Reiser et al envisioned their program working seamlessly into existing curriculum and side by side with physical representations and experiences.  I think that traditional and computational modeling and experiences can and should be complementary, to utilize the power and scope of computational models as well as the familiarity and experiential nature of physical models.  Reiser et al make an excellent point about unfamiliarity, in that you have to be aware of the level of transfer your student is going to experience when using a new tool and optimally prepare them or scaffold with the familiar to help make the experience less dissonant.             

             I am concerned that it is very challenging to know what to pull from a large dataset, even for professional scientists, and that could be a limitation for student’s experience.  Reiser et al suggest doing one exercise beforehand with a smaller data set and then ‘strategic conversations with students about what they are trying to achieve and what they will learn from particular queries of the data’.  However, I think these conversations would need to be heavily scaffolded as I can imagine the blank stares when asking students what they would learn from different statistical queries, potentially with a whole class lesson or discussion on what is possible to ask.  

            I am disappointed with teaching practice section of Reiser et al, as did not really indicate good practices for ‘creating and sustaining a climate of inquiry’, but rather just said to ‘augment’ software, which I think minimizes the role the teacher plays in prompting inquiry.

Questions:

Is the timeline presented by Reiser et al realistic?  With 36 total school weeks, and 6-7 week units, you could cover max 6 concepts. 

What is the best way to scaffold students before/while introducing them to large data sets where correlations may not be readily apparent?

Is Sampson and Gleim’s suggestion to allow students to invent their own method too open ended?