Translate

sábado, 2 de febrero de 2013

Reporte de Doctorantes de Harvard de su visita a la Región Modelo de Orizaba-Zongolica


Veracruz

I.             Description of visit

In January 2013, Jefferson Pestronk and Peter Fishman traveled to the State of Veracruz, region of Zongolica, to observe the work of the Estrategia Integral para la Mejora del Logro Educativo (EIMLE).  We spent a full day at each of three schools:
1) Telesecundaria “La Libertad” in the community of Atlanca (Los Reyes)
2) Telesecundaria “Jose Vasconcelos” in the community of Xaltititla (Rafael Delgado)
3) Primaria Indigena “Adolfo Lopez Mateos” in the community of Tehuipango (Tehuipango)
At each school, we were fortunate to meet students, faculty, staff, and parents, speaking with them about their experience with las relaciones tutoras.  We were also fortunate to be tutored by several exceptional students, engaging in nearly a dozen temas.    
      At Telesecundaria “La Libertad,” Jefferson engaged in the tema, “The Will,” with Victor, learning several key rules of Spanish punctuation. Peter worked with Lucero on the story of “Oetzi,” pulling out main ideas and supporting evidence from a Spanish text.
      At Telesecundaria “Jose Vasconcelos,” Jefferson engaged in two mathematical temas -- plotting coordinates on the Cartesian plane and finding the area of a circular segment -- with Miguel.  Peter also engaged in two mathematical temas -- resolving quadratic equations and finding the area of a circular segment -- with Luz del Carmen.  Later, Peter read “Aplastamiento de las gotas,” a piece by Argentine writer Julio Cortazar, with student Joel. 
      At Primaria Indigena “Adolfo Lopez Mateos,” Jefferson worked on a science tema -- stages of human development -- with student Ana Gabriela.  Later, he engaged in the tema “Ying Yang” with Maria Jose, a student from Secundaria Tecnica Agropecuria #126.  Peter was also tutored by a student from the secundaria, Vladimir, on the coordinates of the Cartesian plane, and worked with a primaria student, Kevin, as he learned to calculate the area of a circular segment from a secundaria student, Juan.
The following sections include our observations and reflections about the tutoria model.  It is important to note that our takeaways are necessarily limited by the period of time that we had with students and the number of schools that we visited.  Both within and between schools, we witnessed a range of implementation of the model -- from some student engagement that very closely approximated the conceptual underpinnings of the model to other engagement that seemed to capture the process, but departed from the spirit, and other engagement that captured the spirit, but departed from the process.    

II.             Observations (when the model is done well)

Focus on the instructional core: In their seminal book, Instructional rounds in education, Elmore, City, Teitel, and Fiarman define the instructional core as “the relationship between the teacher and the student in the presence of content” (Elmore, et al. 2009).  As those authors describe, any large-scale initiative intended to improve student learning must focus attention on the instructional core.  Initiatives that do not in some way impact the relationship between teacher, student, and content, are unlikely to change student learning outcomes.  More than most other educational reforms that we have had the opportunity to study or experience, the tutoria model focuses on the instructional core.  Indeed, the entire model depends upon the teacher (tutor), student (tutored), and content (tema).  No extraneous resources are required or envisioned by the model.  Instead, the model functions by focusing the attention of all stakeholders (teachers, students, parents, administrators) on the student’s engagement with content. 
      Is the content rich and rigorous? 
      Does the tema sit within the student’s zone of proximal development? 
      Does the student carry the cognitive load of engaging with the tema? 
The tutoria model demands that stakeholders ask, and answer, each of these questions.    

Personalization: One of the central promises of the tutoria model is increased personalization of instruction for each student. As the EIMLE leadership described in our kickoff meeting in Mexico City, good learning requires both interest from the learning and knowledge/competence from the teacher. By allowing individual students to select the areas of content in which they want to work, the design of tutoria can use individual interest and engagement to its advantage. The one-to-one or small group structure of tutoring also provides individualized attention, not just simply choice of content. Extensive research has documented tutoring as one of the most effective pedagogical structures when done effectively (see, e.g., Benjamin Bloom’s 1984 paper, The Two Sigma Problem), in large part because an effective tutor can diagnose and address students’ idiosyncratic needs at the earliest possible moment and with customized strategies.

One of the persistent challenges in providing one-to-one tutoring, however, is its resource-intensive nature. Several of the school leaders and staff we spoke with identified the tutoria model as a supplement to the overall structure of the school, which is one way to address resource constraints (e.g., time, people, funding). Students who desire enrichment, or those in need of reinforcement or remediation, can receive tutoring to help move them along, at a lower cost in resources than organizing the entire school day as tutoring. In other cases, one tutor was able to work with multiple students at the same time because there was often significant downtime for the tutor while each student worked on components of the tema. Each of these strategies suggests different ways of allocating limited resources to maximize the impact of the model and may warrant additional investigation to see which is effective in what contexts.

Rigorous discourse: In her work on English Language Arts classrooms in the United States, Pam Grossman of Stanford University has found that deep student learning occurs when students have opportunities to engage in extended discussions in which they develop their ideas and interpretations and in which they are asked rigorous and challenging questions.  Encouraging students to show their thinking increases intellectual acuity and precision.  Too often, however, students are asked rote questions with only right or wrong answers and, in many classrooms, are rarely pushed to show deep conceptual understanding of content.  By asking students to carry the cognitive load and discover answers for themselves, the tutoria model pushes students to develop their ideas and to engage in intellectually precise thinking.  When tutors do not give answers, but rather ask probing questions and when students do not stop at simple solutions, but rather must demonstrate multiple layers of understanding, they are deepening their engagement with content and increasing the likelihood of genuine, sustained understanding.

An example helps to illustrate this point.  Kevin, a primaria student at Primaria Indigena “Adolfo Lopez Mateos” was completing a tema wherein he had to find the area of a circular segment.  Kevin found the area by determining the area of the whole circle and then using angles to determine the percentage of the circle represented by the segment.  However, his tutor, Juan, did not let him merely show that he knew the area of the segment.  Instead, since he had used pi to calculate the area of the whole circle, Kevin’s tutor pushed him to demonstrate the derivation of pi.  Through a series of probing questions, Juan guided Kevin through an exploration of the relationship between diameter, circumference, and pi, showing how its value was derived from the ratio between circumference and diameter.  By the end of the relacion tutora, Kevin could do more than calculate the area of a segment and parrot back the equation representing pi.  He could explain to other students the derivation of pi.        

Knowledge exchange and non-cognitive skills: One of the most heartening experiences of viewing tutoria in action was invisible during the actual process of tutoring but emerged in conversations after working with students. Multiple students indicated that one of their favorite elements of the model was the ability to help their peers understand content better and learn things that they otherwise would not have the opportunity to learn. Put briefly, the tutoria model expanded the leadership role of students in their own learning, and the students grasped that role eagerly. Two brief vignettes help illustrate the point:
      Miguel, a secundaria student at Telesecundaria “Jose Vasconcelos,” wants to be a math teacher when he grows up because the tutoria approach has shown him how much he enjoys helping other students who are struggling
      Ana Gabriela, a primaria student at Primaria Indigena “Adolfo Lopez Mateos,” learned the tema about Stages of Human Development because the only student in her school who knew that tema would graduate to secundaria shortly and Ana Gabriela did not want that knowledge to be lost
Both of these examples show students taking what is, in our experience, unusual agency both for their own learning and for the learning of their peers.

This ownership of learning fits in with other “skills” that we observed in some of the students with whom we worked, all of which might fall under the umbrella category of the so-called “non-cognitive skills.” These include skills like motivation (as discussed earlier under personalization), and grit and persistence (overcoming challenges in the learning process, for example). Similarly, redefining who has knowledge worth knowing and teaching by making the role of teacher and student more fluid, and emphasizing the primacy of wanting to learn something over already knowing something, the tutoria model may have an impact on the development of growth mindsets as described in Carol Dweck’s research. Ultimately, because the tutoria model aims to help students learn how to learn, these non-cognitive skills may be just as important as the knowledge exchanged.


III.          Challenges identified

The takeaways described above were what we saw when tutoring seemed to go well. However, it must be said that not all of the interactions we observed demonstrated these positive characteristics. We believe that at the heart of this may lie two interrelated challenges: the ambitious expansion goals established for the tutoria program and the difficulty ensuring the right balance of consistency and local adaptation inherent in such an ambitious expansion.

Ambition of scaling: The expansion goals laid out for the program are truly astounding. As a central element of the Ministry of Education’s efforts to improve Mexico’s lowest-achieving schools, EIMLE aimed to bring the tutoria model to approximately 9,000 schools. As a point of reference, this is nearly twice as many targeted schools as the United States Department of Education’s School Improvement Grants (SIG), which target 5,000 low-achieving schools in the United States. The SIG program was initially funded with an investment of $3.5 billion, with additional funding in subsequent years. There were few indications that a similar commitment was made to the tutoria program. To be sure, funding is not sufficient on its own to improve schools, but resources to support additional staff, capacity-building at different levels of the system, development of new instructional materials, and so on are enablers of success.

Consistency of the model: In an expansion of the scale described above, it is neither surprising nor (perhaps) bed that implementation was not consistent across sites. What was concerning, however, was that the variation did not seem to be the result of a planned process or in response to particular contextual differences across sites. Across the different schools, we observed different descriptions of the tutoria model, different structures for carrying out tutoring, different beliefs about whether the school should be creating its own temas, different structures (and adherence to the structures) for demonstrating mastery of content, and so on. Based on conversation with the Veracruz team, it also seems that the level of support available to schools varies greatly (perhaps related to the scaling ambitions): some ATPs support a handful of schools and others tens of schools; some ATPs are former educators or other professionals while others received the job through political connections; some schools receive significant training from their regional offices while others do not.

In our experience, reform efforts and programs undergo a process of deciding what must be “tight” about the model and what can be “loose.” Put another way, any school or state implementing tutoria MUST do some things while being flexible on other things. However, what these particular components are was not always clear. Federal leaders made clear that a core tenet of the model is that the first tutoria experience must be lived, but other than that laid out few non-negotiables. Hence, the implementation of the model was inconsistent. Whether this should be the case is a separate question, but we observed it to be the reality.




IV.            Questions

Our observations about the model and our acknowledgment of the challenges inherent in expanding this pedagogical practice to rural schools across Mexico necessarily raised several questions about the model going forward.  We present these questions within the framework of the four broad areas of observation noted above.

Instructional core: The tutoria model focuses attention on the relationship between teacher (tutor) and student (tutored) in the process of content.  Elmore and his colleagues note that coherence among the elements of the instructional core is key in increasing student learning.  With this in mind, we wondered: How does EIMLE create coherence between elements of the instructional core, while allowing for flexibility, customization, and student-driven choice of temas?  Coherence occurs when teachers and students shared a common vocabulary and a common understanding of the tutoria process and when content is aligned with the process.  The model’s vocabulary of temas, guiones, anticipacion, and demonstraciones, when understood commonly by teachers and students, contributes to the coherence of the model.  Moreover, there appears to be an appropriate “grain-size” of content -- a discrete skill embedded within a broader context with real-life application -- that is well-suited to the tutoria model. 

As EIMLE seeks to scale the model, we wondered what elements of the model must remain consistent and what elements might be adapted for a local context.  For instance, when tutoria is applied to the topic of carpentry, as the community of Atlanca intends to do, does the “grain-size” of content shift or remain consistent?  Could a tutor offer a tema on the building of a chair or must he select a more discrete skill, such as the smoothing of a piece of wood?  Implicit in this question is whether the “grain-size” of the content is key to ensuring coherence of the model.  The same question may be asked of various elements of the model.  For instance, in at least two schools that we visited, use of the culminating demonstracion was infrequent.  Students demonstrated their knowledge to the tutor, but were rarely asked to present to the broader class community.  When we did witness demonstraciones, the format and depth of each presentation varied widely.  This element of the model lacked coherence.  We wondered whether expectations for a demonstracion should be consistently applied across students and across schools or whether this element of the model was one that might be highly differentiated based upon the individual school and the individual student. 

Our research and experience in scaling programs in the United States suggests that there is not a simple formula for determining when and how to be consistent and coherent and when and how to allow for flexibility and differentiation.  The critical practice is to be in the habit of consistently evaluating when to be loose and when to be tight.        

Personalization: We had the good fortune to experience tutoring from students at both the primaria and secundaria levels. The experience was relatively different; the younger students we worked with were less likely to treat the tutor’s role as the possessor of knowledge who should provide it when the tutee struggled, rather than to help the tutee discover answers himself by providing gentle guidance. This seems unsurprising that some of the specifics of the methodology would be more difficult for younger children to grasp, but it does raise the question of the extent of applicability for a single methodology. Specifically, is the same tutoria methodology equally appropriate for students of different ages?

A second question is whether the methodology is equally appropriate for all types of knowledge. The intellectual roots of the tutoria model share much with the “deschooling” movement articulated by theorists such as Ivan Illich. Both are premised on the idea that individuals who are personally motivated to learn and are given the opportunity to connect will glean valuable knowledge from interacting. However, in Illich’s explication of his idea in Deschooling Society, he gives examples of interactions that are more unstructured, such as individuals meeting through the use of technology (in a pre-internet era) to discuss a book of common interest. In this example, there are not right or wrong answers; what two people take from a piece of literature is almost certainly going to be different. However, the same is less true in subjects such as mathematics (see below for an example), science, or in teaching principles of language such as grammar. In these cases, more explicit structure or guidance may be valuable to ensure that students retain accurate knowledge. Given these differences in types of knowledge, need the methodology change?

In both of the cases described above, the experience of the EIMLE team may suggest that the methodology is flexible enough to accommodate learners of different ages and of different subjects. If that is the case, then our experience in Veracruz suggests that schools and students implementing the methodology may need more help understanding how to apply it.

Rigorous discourse: As noted earlier, a key feature of the tutoria model is the depth of exploration embedded within each tema.  This depth is partly achieved through scripting of the tutoria process.  Each tema should include an anticipacion section, wherein students access prior knowledge and predict what they will learn.  At the conclusion of each relacion tutora, the student analyzes his or her learning process by answering four key questions (What did I learn?  How did I learn it?  What difficulties did I encounter?  How did I resolve these difficulties?).  Scripted procedures such as these ensure that the student engages in meta-cognitive reflection.  The intention is for learning to be dynamic and deep, not rote and shallow. 

In several instances, however, we witnessed student activity that was more procedural than substantive.  This occurred both when students acted as tutors and when they were tutored.  In one example, a secundaria student answered an anticipacion question by saying that he expected the tema to be “interesting.”  The tutor, who was also a student, accepted this response and moved on to the next step in the process, despite the fact that the answer lacked both substance and evidence of deep thinking.  The tutor might have asked, “What do you think will be interesting?  Why do you think it will be interesting.”  However, the tutor seemed happy to have “checked off the anticipacion box” and to move on. 

We saw this type of procedural thinking play out again and again.  It made us wonder: How does the model ensure that the focus is on the learning itself rather than the procedures underlying the learning?  We wondered if more might be done with students to develop understanding of the purpose behind each procedure.  Indeed, at times, it seemed that students had learned a procedure without knowing why, for instance, they might ask their tutees to predict the content of a given tema.  If a tutor knew, for instance, that he was looking for evidence of prior knowledge applied to the a given topic, he might be less likely to accept “interesting” as an adequate response to an anticipacion.     

Knowledge exchange: As noted earlier, the interaction between students and their desire to take collective responsibility for their own learning is a strength of the tutoria model when it is done well. However, there are risks to students, rather than more experienced or knowledgeable teachers, taking the lead. (Note that we do not assume here that all teachers are more experienced or knowledgeable, but it seems a safe assumption that on balance teachers have more instructional experience and content knowledge than students.) Most notably is the risk of incorrect or incomplete exchange of knowledge, something that both of us experienced or witnessed during our school visits. It is worth noting that our colleagues who visited other states also indicated they experienced the same phenomenon.

Specifically, during temas that we either received or observed, students taught content that was explicitly incorrect, not simply a matter of questionable interpretation or pedagogy. An example was a mathematical formula that had been taught incorrectly to the student, which one of us discovered when our answer did not match the answer that student, acting as the tutor, had recorded. Fortunately, in this particular instance the error could be remediated because the student had not tutored any other students in that particular tema yet, but that seems unlikely to be the case in other circumstances.

This raises questions about how to catch this type of misunderstanding quickly and systematically. At least in theory, the presentation that the student performs at the end of the tutoring session should be the check on mastery, but in our experience those presentations were often much more focused on process than on content mastery; alternatively, some other form of assessment could be conducted in an effort to identify potentially incorrect knowledge.