Showing posts with label Data-Driven Teaching. Show all posts

Deriving Insights of Value

#EdChat, Big Data, Data Science, Data-Driven Teaching, The Data Scientist 0 comments

Quartier de la Sorbonne, Paris, France

Photo by Jonathan Acuña (2018)

Deriving Insights of Value

A simple problem, but what’s the answer?

By Prof. Jonathan Acuña-Solano, M. Ed.

School of English

Faculty of Social Sciences

Universidad Latina de Costa Rica

Saturday, July 27, 2019

Post 333 / DS Log 5

As mentioned before, the data scientist teacher is someone who works with data in three different fronts simultaneously. That teacher is a miner, a cleanser, and an analyst. Let’s see his/her role in the following case scenario, a language school whose graduating population is not interested in participating in the institution’s graduation.

1	Problem Identification	In this case scenario the problem is not that only 53% of the possibly graduating learners want to participate in the end-of-the-course ceremony; the real problem to discover is why this is happening. The teacher has to find out the real question to ask to mine and explore data.
2	Data Collection	At this point, the teacher has to start collecting data from as many sources as it is possible including the school’s database, any economic report issued by the Ministry of Economy, printed press and news programs on TV/radio or social media.
3	Data Exploration	With all possible data coming from public or institutional databases, the teacher can start finding the pieces of the puzzle. Information can then be organized for the teacher to start finding commonalities, categories, or names for the reasons that are found.
4	Analysis of Data Collected from Various Sources	The teacher cannot only base the analysis of data from institutional databases. Some sort of triangulation is needed to match data to reveal connections never thought before but that may have a say on why some 43% of possibly graduating students don’t feel encouraged to attend a graduation ceremony.
5	The Storytelling behind the Data	Once the teacher lets the data speak by themselves, s/he listens to what all these is stating as to why learners are uninterested (or interested) in going to a graduation ceremony. Reasons can be triggered by different societal phenomena.
6	Taking Action	As soon as the data tell their story, the teacher has to talk to the institution’s stakeholders, directors, or departmental heads to discuss what needs to be done to either motivate learners to join the graduation celebration or to reconsider if the way the graduation ceremony is actually held should continue to be celebrated in the same way.

Though it is not yet clear why only 53% of learners who finish the institution’s program are willing to go to the graduation ceremony, the case is a good example as to how the “investigation” has to be carried out. And for a better understanding of these notes of mine, please take a look at the infographic that matches what each step of the data science procedure contains along with the case scenario.

Saturday, July 27, 2019

Teachers Exploring Data: The teacher as a data scientist

Big Data, Data Science, Data-Driven Teaching, The Data Scientist 0 comments

Hyde Park, The Long Water, London, UK

Photo by Jonathan Acuña (2018)

Teachers Exploring Data

The teacher as a data scientist

By Prof. Jonathan Acuña-Solano, M. Ed.

School of English

Faculty of Social Sciences

Universidad Latina de Costa Rica

Thursday, July 25, 2019

Post 331 / DS Log 4

Can teachers mine and explore data? A resounding yes as an answer needs to be uttered. Data Science (DS) is not exclusively connected to business management decisions; it can be perfectly manipulated by educators, who -in the search for answers- can make use of information to decide on various education issues. Though DS is not meant to be the cup of tea of every education professional, institutions ought to count with some to help in the visualization and explanation of behavior among students.

Educators interested in Data Science should look for a role in data visualization. These teachers need to encompass a set of traits that can make their mining of data easier:

a) The ability of manipulate data,

b) The talent to analyze data,

c) The capacity to come up with innovative solutions,

d) The skill to present findings, results, and suggestions,

e) The cleverness to tell the story behind the data, and

f) The knack to deal with other stakeholders.

Due to the importance of these traits, teachers playfully manipulating data to make sense of their reality also possess curiosity, analytical strength, excitement about data, and some sort of technical skills to work with data from a database.

Copy the following Google Drive address to see the original infographic and to enlarge it:
https://drive.google.com/file/d/1Jm1_QcrT48if_vWkB_LGSO-s2nTByDCW/view?usp=sharing

Are educators ready to explore data? No doubt that there are many teachers who already work with data empirically. And this empiricism is perfect because it is the basis for helping them develop their competence in this area. Conclusion: as Professor Murtaza Haider stated (2018), he can train someone to use algorithms and software to work with data, but curiosity and the desire to explore data cannot be taught.

References

Haider, M. (2018). Specific Skills to Hire a Data Scientist. [Video File]. Retrieved from Laureate Edcuation, Inc. at https://dtl.laureate.net/webapps/blackboard/content/listContent.jsp?course_id=_165016_1&content_id=_801204_1&mode=reset

Thursday, July 25, 2019

What's Needed to be a Data Scientist?

Big Data, Data Science, Data-Driven Teaching, The Data Scientist 0 comments

Hyde Park, The Serpentine Lake, London, UK

Photo by Jonathan Acuña

What’s Needed to be a Data Scientist?

In language learning and education

By Prof. Jonathan Acuña-Solano, M. Ed.

School of English

Faculty of Social Sciences

Universidad Latina de Costa Rica

Sunday, July 21, 2019

Post 330 / DS Log 3

After putting together all pieces of information in the infographic I created to see what’s needed to be a data scientist, lots of thoughts came into my mind as to how this relates to me as a language instructor and educator. What’s really behind all this new knowledge linked to Data Science (DS) and Big Data (BD)? Let me explore some of my ideas regarding DS and BG and their connection with education.

With a bit of curiosity, all teachers can become data scientists of their own data. Each of our courses is a unique situation that will generate data that must be comprehended to cater for student learning and coaching needs. Teachers have the capacity to analyze, provide solutions to class situations, and spot ways to interpret the data, e.g., exam results, platform grades and performance, etc. Basic statistical analysis is needed to go into mining data for analysis and interpretation, and in the end, we can provide better teaching and coaching to learners. Educators need to become data miners and explorers to provide students with ways to develop skills and competencies.

Mining our own data also reflects our ability to tell the story behind what is being analyzed and understood. In a school situation, where lots of data pertaining grades (and other pieces of information) is stored in a data base, a teacher or school administrator can analyze, interpret, and tell the story that data tell him/her. Finding ways to present discoveries among the analyzed data is something that with a bit of exercise any educator can do. The identification of new revelations in data can be a happy, joyful moment since it can be the identification of a path to follow to help students in their learning process.

Dr. Murtaza Haider (2018) stated that a great sense of humor is needed by anyone working as a data scientist, and I bet it is the same principle for any educator. The teacher can laugh at him/herself because of the “crazy ideas” that can arise from the interpretation of data in the school data base. Seriousness in data mining, interpretation, and analysis is not always good because humor can be a great part of team building that can erase worries from the face of educators. The teacher has to be good at laughing at his/her “odd” ideas that can help comprehend what is happening in a class, course, or school year.

There are more traits a teacher data scientist needs to embody. What was introduced here was a good account of characteristics needed by a miner in data science. It is our task to see how they become relevant to our teaching profession and how we can use these data to help the learners, the institution, and the educators with ways to correct paths and provide guidance towards success.

References

Sunday, July 21, 2019

Potentials for Data-Driven Instruction: What is in Store for ELT?

Blended Learning, Data-Driven Teaching, LMS, Teacher Training 0 comments

The Chess Fountain, Gran Estación Mall, Bogotá, Colombia - Photo by Jonathan Acuña

Potentials for Data-Driven Instruction:

What is in Store for ELT?

By Prof. Jonathan Acuña-Solano, M. Ed.

School of English

Faculty of Social Sciences

Universidad Latina de Costa Rica

Sunday, May 6, 2018

Post 317

As a curious traveler who intends to explore new places, I’ve been too several intriguing but fascinating non-conventional sites in my life. And while strolling down the many shopping places, museums, and other tourist attractions in Bogotá, my wife and I came across this amazing sight, a chess fountain. And since I am the kind of person who always wants to understand things beyond what is really perceived by the naked eye, here I was faced with these gigantic chess pieces, like the mammoth issue linked to the “effective, meaningful” usage of an LMS (learning management system) in a language institution interested in using data coming from it to refine its language programs.

Whenever I think of chess, I always picture those Muslim noblemen from a different era making their right plays towards a chess mate to win a game. And as you must be certain, in order to win a chess game, several moves are made to achieve that goal. However, when I relate data-driven instruction in ELT to winning a chess game, I often wonder if we can ever get to win this game we are playing now to have learners profit from their use of an LMS. How far have we come along this road to take advantage of an LMS aligned with an institution’s curricula? Well, let me share a bit of what I have experienced and mull over for quite some time …

The LMSs in the Lookout

At this point in our history, most publishing houses in the ELT business have come to devise their LMS to provide schools, language institutes, universities, etc. with a digital way to track student coverage of thematic units from the coursebook they have chosen for their programs. I refrained myself from using the word learning (to replace coverage of thematic units) since the act of learning is a very personal moment of self-discovery that is not exactly measured through an LMS devised by a publisher. For me, learning is an intimate moment when each individual discovers s/he is now in possession of new knowledge that can be used otherwise. And the “possession” of new information can be attained through a platform such as the LMS.

What worries is not the presence of an LMS in a language institution, but the way it is being used by the school’s instructors. Based on my personal experience with teachers having to assign content in the institution’s LMS and memoranda that I have written to myself over the years after talking to my peers in various university settings in my home country, the school’s platform is more likely to be used as the substitute of the paper workbook we used to have before; it has become an e-Workbook full of exercises that can be assigned or de-assigned by the instructor. The statistical evidence that a system like an LMS can generate is not being used profitably to enhance the learning process by educators or school policy makers. All this situation resembles again as if we were in this chess game against one of the Muslim noblemen facing a “check,” and we are close to discover that any effort is not enough to win the game because we are not really using the LMS to help plan for learning.

The question now is, is this abandonment of statistical evidence taking place because of lack of teacher training or dearth of understanding? Are we about to lose our chess game because of lack of expertise in the game? Being someone dealing with some sort of administration of an LMS, I see the lack and dearth in teachers’ platform tasks. Based on my reflections regarding this abandonment of statistical evidence, educators are not grasping the real use of an LMS for learning purposes. We have not trained our educators/players with the right “plays” (moves) to use a platform for the sake of student learning. Additionally, instructor supervisors are not providing an accurate and meticulous follow-up of teacher work on the platform. And all this panorama makes us wonder whether we really overlooked the real potential of Data-Driven Instruction (DDI) in ELT or not. Did we also overlook the fact that chess players are also taught or self-instructed to be good in the game, even when it comes to be playing against a computer software?

Pinpointing Problem Areas

The one question to ask over here, in this chess game-like situation, is: “are we teachers planning around troublesome areas when teaching and then using an LMS as part of our blended learning approach for student language development? If we all use the data that LMS platforms generate to enhance our teaching, we “teachers can more accurately pinpoint the problem areas that most of the students have and then spend class time on those” (Baber, 2013). However, we tend to plan around the textbooks rather than around the troublesome areas that our institution’s platform is revealing but that are not being taken care of “accurately.” By far the LMS can help us spot “problem areas” by analyzing data connected to student performance while working on the exercises (or tasks) on the platform. But, are we doing it? If we are not exactly doing this, not spotting the language learners are struggling with, planning cannot be geared towards aiding pupils to improve and master vital contents in their language development.

Baber (2013) suggests being more creative in the use of a data-driven language class. For him, “being perhaps a bit more creative, there’s scope for different classroom constructs altogether” (Baber, 2013). What about teaching a class based on data from the LMS that has been analyzed; the analysis can tell us what needs to be taught and who needs to be guided and instructed. Suppose we have a group of students dealing with past perfect, and there might be a section of the class that with the platform work they are able to master the topic quickly and accurately. What do we do with these pupils? Baber (2013) suggests that “instead of shepherding them all into one room three times a week,” we can “have three different classes, each focusing on a different problem area that a subset of those students have.” And the benefits? These can be much better for all students by having the ones with very particular troublesome areas to overcome practice work with the instructor, and those who do not need that much instruction can focus their attention on other language contents that require more work for them.

A Shift in Class Constructs

Based on what Baber (2013) comments about “shepherding” students, why do we have to make learners attend class? As educators we have all experienced the situation where a subset of learners in class already know the content we are to study. Depending on the ages and maturity of students, this particular scenario can trigger boredom and class disruption affecting classroom management. Isn’t it better to ask pupils to come when they need to? “While the total classroom time per student is less, it meets the needs of those particular students far better, and performance can be increased” (Baber, 2013). Language instruction does not always happen in the classroom; many of the LMSs do include this instruction with inductive and deductive tasks for learners to come up with their own rules, or simplified versions of the content (especially grammar) that needs to be mastered by the student. The class needs to become a place to practice the language being studied for those pupils who have areas that they must strengthen. “The online component is also delivering teaching, not just consolidation exercises or assessment, so students benefit via both mediums” (Baber, 2013).

As the chess players of these teaching scenarios where an LMS can provide us with information about our learners, are we profiting from these learning platforms by using the right moves? “Blended learning has long been heralded as the Holy Grail but I don’t think we’ve actually seen the benefits yet” says Baber (2013). Learning management systems can provide us with data to drive our teaching to help students where they really need; it can provide as with very punctual information about where students are experiencing a problem area. We are at a “check” point on our LMS chess game because we are not really instructing our educators (and ourselves) to use every datum to potentiate student learning and mastery of the target language.

Data-Driven Performance Improvement

Are we really meeting learner needs in the classroom? If we think we are fully fostering student learning in our classes, we can be surprised by what students can say about that. That we are not trying to attain the correct deployment of LMS use is not being stated here, but that we need to try to redirect our teaching to meet learner needs is by far a fact of teaching and language mastery goals.

Data-driven teaching can be the way to cater for learner needs and language mastery goals. Can student performance be really improved with this new approach for playing this chess game-like new way of planning? Why not!?! If we teach what data coming from the LMS states, we are bound to discover that learning materializes in different ways. Since “Data Driven Instruction and Inquiry (DDI) is a precise and systematic approach to improving student learning” (New York State Education Department, n.d.), the usage of the inquiry cycle of data-driven instruction that “includes assessment, analysis, and action and is a key framework for school-wide support of all student success.” And as data-driven instruction is conceived, assessment is already covered by our students when working on the LMS, but analysis is what may be missing in the correct “play to attain a check mate;” that is, when we analyze the data, the proactivity in action we are to embark ourselves in our teaching and planning is linked to the problem areas students must improve to master the piece of language they are studying with us.

Taken from the Engage New York New York State Education Department’s web page at https://www.engageny.org/sites/default/files/ddi-arrow-chart-small.png

The Need for More Training

“Teachers who are trained in using [the LMSs] and evaluating students’ performance data can make a real difference to their students’ learning” (Baber, 2013). But the fact of life is that we are not there yet, and we must work towards teaching our instructors how to use the DDI teaching model. And this has to be done through closer supervision. We cannot pretend that our language trainers will learn to do this overnight, especially when we think that the trainers are the first ones that must be trained to coach their supervisees.

The first step towards the use of data from the LMS to teach is to help educators understand and analyze reports. All teachers must have their eyes open to see how to make their next “chess-playing movement;” once instructors can identify the red flags, they have to analyze time spent on tasks to spot troublesome areas, the grades students are getting, the number of attempts registered in the platform, the frequency of sign-ins to platform, and so on. But more than identifying the red flags, they must read between the lines to see what is actually happening with pupils and their language learning.

Concluding Remarks

For those of us newbies with LMSs, data-drive instruction, the DDI teaching model, and so on, we must all agree that:

1) It’s important to get trained to profit from all data coming from a platform and convey more significant learning to and for our pupils in the classroom;

2) It’s sensible to expand our understanding of all these new elements that are becoming “the next big thing” as Eric Baber, former IATFL President called all this back in 2013. The more we get familiarized with the data functionalities in the LMS, the better for our teaching and for our pupils’ learning; and

3) It’s our responsibility to have us deepen ourselves into data-driven instruction to find more meaningful ways to teach our students, so they can really profit from any time they invest in their language learning.

Let’s turn our teaching more DDI-ish to really provide language learners in our classes with more accurate instruction and more memorable learning moments for our students. By doing all this, and many other “chess plays” or tricks we can learn along the way, we can become better players aiming at winning the game and materializing student language learning.

References

Baber, E. (2013, May-June). Data-Driven Teaching: The Next Big Thing? (IATEFL, Ed.) Voices(232), p. 3.

New York State Education Department. (n.d.). Data Driven Instruction. Retrieved from Data Driven Instruction and Inquiry: https://www.engageny.org/data-driven-instruction

Sunday, May 06, 2018

Data-Driven Teaching: A Shift in Blended Learning Education

Data-Driven Teaching, DDT, Hybrid and Blended Learning, LMS 0 comments

Photograph taken in Honduras, CA and contributed by Fernando Carranza

Looking Through Casement ELT Window

Data-Driven Teaching:

A Shift in Blended Learning Education

By Prof. Jonathan Acuña-Solano, M. Ed.

School of English

Faculty of Social Sciences

Universidad Latina de Costa Rica

Saturday, October 15, 2016

Post 300

For several years now, blended teaching in language learning is an integral part of many programs in universities or at language schools around the world. That is, with the incorporation of learning management systems (LMSs) for language development and mastery, teachers are now in much control of what students are doing away from the classroom. Though the LMSs has come to substitute the traditional print workbook of yesteryear, data now coming from the platforms are not really being used to plan instruction and learning focused on the students. A shift in blended learning education has not yet been accomplished, and it is a real need nowadays.

Let’s Understand Data-Driven Teaching (DDT)

“Data analysis can provide a snapshot of what students know, what they should know, and what can be done to meet their academic needs. With appropriate analysis and interpretation of data, educators can make informed decisions that positively affect student outcomes” (Lewis, Madison-Harris, & Times, n.d.). In terms of language teaching and learning, DDT must then be focused on relevant areas of instruction for learners; DDT does not focus on teacher-centered instruction, but quite the opposite. Its main reason to exist is to help educators to create activities that guarantee student-centeredness in language training. While using data to guide one’s teaching, planning is then targeted to strengthen student weak, developing areas and not to just cover course textbook content due to the suggested pacing for a course.

Instructor-Led Online Hours do Count in DDT

Are language instructors really aiming at using DDT while teaching a course? Based on my experience with LMS administration and usage mostly recorded in memoranda, this has not materialized yet in my language teaching contexts, at the university and the language school where I work. The LMS is being loosely used by instructors and colleagues to basically assign content on the platform to somehow practice what is covered in F2F class sessions. Somehow the LMSs continue being used as eWorkbooks rather than a system to collect data for the strengthening of one’s teaching to develop student performance weak areas. The effect of using the platform as an eWorkbook is that the time spent online is not consolidating student learning, which is meant to be the reason why LMSs exist. As a consequence, planning needs to be connected to what data on the platform are telling instructors to guide them in class teaching in a blended learning scenario. The one single imperative that is being left out in this new educational scenario is the analysis of statistical reports to create a connection between the classroom, the platform, and back to the classroom. And part of this imperative is to use this blended learning instruction cycle to make instructor-led hours count for student language development.

Why LMS Work Instead of Paper-Based Homework

If my typical, traditional student is like yours, print workbooks for homework are not exactly a priority for them. A typical learner of mine is that one that shows up for class with not homework on his/her workbook, either because they simply forgot or because s/he could not find the time to complete the assignment. Consequently, learning consolidation may not be achieved when this kind of learner decides that homework is not important for him/her due to their other social or educational endeavors. Moreover, for the teacher –when the workbook’s exercises are checked orally, there is no way to know what areas are giving learners a hard time; something that can be easily done now with the statistical reports that an LMS can produce for instructors. To sum up, language teachers with no DDT orientation as part of their planning and teaching fail in LMS correct use. The right usage of the LMS explains why it is essential to use statistical information to use the platform instead of a print workbook due to the amount of information that can be derived from LMS’s reports.

The Need for Making LMS Instructor-Led Hours Count

A mind shift is needed on how LMS work is perceived by students and orientation is needed from instructors. Student independent work is quite good for self-regulated individuals, and many people take their language learning seriously, whether that is a language or something else they are interested in. For more traditional students, the LMS guided hours can be very fruitful if DDT is present. The idea that platform exercises just need to be completed to comply with work for a course is not exactly the expected behavior a real interested individual demonstrates in language learning. That is why it is necessary make the LMS instructor-led hours count for language mastery and performance. Online hours, as it can be seen, need to trigger data to drive the blended learning cycle to practice the areas that must be practiced, and not a random exercise a teacher arbitrarily decides is the right one to join class activities with platform tasks and then back to the classroom exercises. Though the LMS work may good for some individuals, those students who already know the subject-matter by heart do not need to review what they have already mastered; the platform hours need to be guided in such a way that learners just work on the areas they need to continue developing.

The Blended Cycle and DDT

To really make the LMS instructor-led hours count, the blended cycle needs to be based on data-driven teaching. At this point in language blended education, the instructor is practicing content in class to improve student performance in the four skills. Then, practice activities to continue building on class content is assigned as a consolidation task (group of exercises). Then, learners, guided by the teacher, retake the same content to demonstrate the mastery of it in class. Data produced by the LMS reports generated by the platform is analyzed prior the retaking of content for demonstration; this is done to create a lesson plan that is initially based on the LMS trouble with activities learners had and that the data show. Whatever is going to be (re)practiced in class is to help learners who show difficulty in their LMS work understand whatever they have not been able to grasp 100%.

Taken from https://www.cli.org/blog/what-is-data-driven-instruction/

Measurably Better

“If teachers deploy blended learning ‘properly’, students’ results are measurably better” (Baber, 2013). Why is it that we are still striving to get better results with students when we have information right at our fingertips by clicking here or there in the LMS? As Baber (2013) cleverly states it, “a key element of ‘proper’ deployment is that teachers regularly log into the learning management system, view students’ performance, and then adapt what they do in the classroom.” But based on my reflective journaling and personal memoranda, I can barely see any of this Baber is talking about actually happening in any of my two workplaces. Once again, LMSs are being used as eWorkbooks that will not yield the same kind of result similar to the one we could be getting by simply logging into the platform from time to time to see what the whole group as a whole is having trouble with, and from that point on continue building the language they need to develop, but with a more constructivist orientation in our planning process and blended teaching practices.

“If teachers plough on with their pre-determined curriculum regardless of the students’ strengths and weaknesses as visible from their performance date –they may as well go back to old-fashioned homework on paper” (Baber, 2013). Yes, it is true that we have course outlines to follow as well as a coursebook that needs to be covered, but without understanding learners’ “strengths and weaknesses” we are just contemplating what really is happening outside through a casement window; while there are teaching professionals deploying blended learning practices properly and there are students learning a language proficiently, what are we waiting for to log into the system(s) we are currently using and learn from what our students are striving to learn to give them a hand and the correct kind of blended activities to guide them through their construction of the inter-language to speak English as a Lingua Franca (ELF) as it is described in the CEFR.

References

Baber, E. (2013, May-June). Data-driven teaching: the next big thing? Voices, 232.

Lewis, D., Madison-Harris, R., & Times, C. (n.d.). Using Data to Guide Instruction and Improve Student Learning. Obtenido de SEDL.Org: http://www.sedl.org/pubs/sedl-letter/v22n02/using-data.html

Saturday, October 15, 2016