my-efl-blog

[my-efl-blog]

Duolingo Sucks, Now What?: A Guide

Now that the quality of Duolingo has fallen (even more) due to AI and people are more willing to make the jump here are just some alternative apps and what languages they have:

"I just want an identical experience to DL"

Busuu (Languages: Spanish, Japanese, French, English, German, Dutch, Italian, Portuguese, Chinese, Polish, Turkish, Russian, Arabic, Korean)

"I want a good audio-based app"

Language Transfer (Languages: French, Swahili, Italian, Greek, German, Turkish, Arabic, Spanish, English for Spanish Speakers)

"I want a good audio-based app and money's no object"

Pimsleur (Literally so many languages)

Glossika (Also a lot of languages, but minority languages are free)

*anecdote: I borrowed my brother's Japanese Pimsleur CD as a kid and I still remember how to say the weather is nice over a decade later. You can find the CDs at libraries and "other" places I'm sure.

"I have a pretty neat library card"

Mango (Languages: So many and the endangered/Indigenous courses are free even if you don't have a library that has a partnership with Mango)

Transparent Language: (Languages: THE MOST! Also the one that has the widest variety of African languages! Perhaps the most diverse in ESL and learning a foreign language not in English)

"I want SRS flashcards and have an android"

AnkiDroid: (Theoretically all languages, pre-made decks can be found easily)

"I want SRS flashcards and I have an iphone"

AnkiApp: It's almost as good as AnkiDroid and free compared to the official Anki app for iphone

"I don't mind ads and just want to learn Korean"

lingory

"I want an app made for Mandarin that's BETTER than DL and has multiple languages to learn Mandarin in"

ChineseSkill (You can use their older version of the course for free)

"I don't like any of these apps you mentioned already, give me one more"

Bunpo: (Languages: Japanese, Spanish, French, German, Korean, and Mandarin)

[my-efl-blog]

not to be a pain about this but not only should you learn to read scientific papers but you should know how to find out if a journal is reputable and some basic statistics (i promise it's not as scary or hard as it seems). these are key adult life skills, not niche academic pursuits. science journalism is abysmal, and researchers misrepresent their own findings all the time.

[my-efl-blog]

"contract grading" "only 4 absences or you drop to an F" "in this class we will be teaching about disabilities. attendance is mandatory and i do not accept late work" "please respond to at least two of your peers in this discussion post" "people with autism need time to decompress in a classroom environment. your class is four hours long with a 7 minute break." "we like to let students learn the way THEY want to learn. please buy our 150 dollar textbook."

[my-efl-blog]

debating if it would be funnier to have a bumper sticker saying "my other ride is a [exact make and model of the car the sticker is on]" or "my other ride is a [equally shitty but different car]"

[my-efl-blog]

This is the magic lucky word count. Reblog for creativity juice. It might even work, who knows.

[my-efl-blog]

got a major pest problem this year actually

[my-efl-blog]

[my-efl-blog]

WE NEED TO KILL HIM

[my-efl-blog]

The research I’m currently doing involves going through articles in an academic journal manually — article by article, issue by issue, in reverse chronological order.

In 2010, an in memoriam article was posted about a notable scholar in the community who was trans. It shared about her life and work, and mentioned her birth name and her chosen name.

Just got to 2005, and she published in issue 2 under her birth name. And then in issue 3, she published under her new name for the first time.

She was in tech, and a lot of what she wrote about is obsolete by now. I have no need to read her publications. But it’s just so touching to see our history laid out like this — an unremarkable name-change in a journal followed by an obituary five years later. It really is never too late.

[my-efl-blog]

Anybody else got that Evergiven sized writers block

[my-efl-blog]

The almost overnight surge in electricity demand from data centers is now outstripping the available power supply in many parts of the world, according to interviews with data center operators, energy providers and tech executives. That dynamic is leading to years-long waits for businesses to access the grid as well as growing concerns of outages and price increases for those living in the densest data center markets. The dramatic increase in power demands from Silicon Valley’s growth-at-all-costs approach to AI also threatens to upend the energy transition plans of entire nations and the clean energy goals of trillion-dollar tech companies. In some countries, including Saudi Arabia, Ireland and Malaysia, the energy required to run all the data centers they plan to build at full capacity exceeds the available supply of renewable energy, according to a Bloomberg analysis of the latest available data. By one official estimate, Sweden could see power demand from data centers roughly double over the course of this decade — and then double again by 2040. In the UK, AI is expected to suck up 500% more energy over the next decade. And in the US, data centers are projected to use 8% of total power by 2030, up from 3% in 2022, according to Goldman Sachs, which described it as “the kind of electricity growth that hasn’t been seen in a generation.”

21 June 2024

[my-efl-blog]

Y'all I know that when so-called AI generates ridiculous results it's hilarious and I find it as funny as the next guy but I NEED y'all to remember that every single time an AI answer is generated it uses 5x as much energy as a conventional websearch and burns through 10 ml of water. FOR EVERY ANSWER. Each big llm is equal to 300,000 kiligrams of carbon dioxide emissions.

LLMs are killing the environment, and when we generate answers for the lolz we're still contributing to it.

Stop using it. Stop using it for a.n.y.t.h.i.n.g. We need to kill it.

Sources:

[my-efl-blog]

Took a year to complete this quilt! Pattern is by NASA Astronaut Karen Nyberg called Cupola View. Fabrics used were also designed by Karen, the collection is called Earth Views.

[my-efl-blog]

my linguistic anthropology textbook getting pretty excited about this Chomsky knitting AU

[my-efl-blog]

(Thanks Rachel!)

[my-efl-blog]

Assessment Terminology

There are many, many words used in discussing assessment in education. Here is a non-exhaustive list of them, with definitions to contextualize them and clarify their meanings. The other glossary is full of other TESOL terms, so if you encounter a term here that you don’t recognize, check there.

General Terms:

The Purpose of The Assessment:

Diagnostic – A diagnostic is an assessment given at the beginning of a course or term that gives specific information about a learner’s abilities or knowledge, which allows an educator to focus on enhancing the student’s strengths and supporting the student’s weaknesses. For example, a diagnostic lets an educator know that a student may be good at the past tense but has difficulty with academic vocabulary.

Formative – Formative assessment is assessment done during the program that influences the program moving forward. Formative assessments during a semester allow an educator to know what the students have mastered, what they are still learning, and what they have not yet learned, which allows the teacher to adapt their lessons to address what the students need more help with. For example, formative assessments about the past tense may highlight which irregular forms should be emphasized in the next unit.

Summative – Summative assessments are conducted at the end of the term, and they are used to show what the student has learned by the end of the program. Final exams are summative assessments.

Performance assessment – a performance assessment asks a student to complete a task of some sort, and their ability to complete that task demonstrates their abilities. Presentations and group projects are examples of performance assessments. These are generally evaluated with a rubric or other rating scale.

Knowledge test – these assessments measure what information a student knows. This information can be facts, models, or processes, usually as learned in content classes.

Skill test – these assessments measure a student’s ability to perform a skill. Second Language acquisition often refers to the Four Skills – Reading, Listening, Speaking, and Writing, so a test of those skills would be a skill test, but an assessment of a specific subskill, like a roleplay that assesses a student’s ability to politely refuse an invitation in the target language, would also be a skill test.

The Contents of the Assessment:

Curriculum-Related Tests:

Admission test – the results of these tests are used to decide if a student is allowed to enroll in a program.

Placement test – these tests are used to determine what course or level a student will be placed into in a program.

Progress test – these tests show how far the student has progressed – how much their knowledge or skills have grown – during their time in the program.

Achievement test – these tests show what the student has accomplished by the end of their time in the program.

Non-Curricular Tests:

Screening test – these tests determine if a student in the US K-12 system is an ELL and if they should be in an ELD program.

Proficiency test – these tests determine how much of the L2 a person understands and can produce. Most people’s levels vary by skill (reading, listening, writing, speaking, grammar, vocabulary), and many people will have small daily variances due to their mood, health, etc. However, a proficiency test should in general explain what the language user can do and what they cannot yet do in the L2.

The Source of the Test:

Classroom-based – assessments in the classroom based on the content, curriculum, and practices of that specific classroom. Not shared between classes or teachers.

Common – assessments used by all teachers in a specific level or program. If all of the instructors of the class use the same assessments, these are common assessments.

Large-scale – these assessments are used in many contexts beyond a single school or course/program. National exams and college entrance exams like the SAT are large-scale.

Teacher-made – these assessments are made by the instructor(s) of a course, usually for that specific class.

Published – these assessments are made by groups or corporations for use by many instructors in many different contexts. Tests that come with a textbook are published.

Standardized –these are large scale tests which are given to test-takers in settings that are as identical as possible to each other. All test-takers receive the same questions and the tests are proctored and invigilated in the same ways. The IELTS and the ACT tests are standardized tests.

Other Important General Terms:

Formal – formal assessments are planned. Tests and homework assignments are formal.

Informal – informal assessments are incidental and unplanned. An impromptu conversation that allows an instructor to gauge the speaking proficiency of a student is an informal assessment.

Low Stakes – these assessments have little to no repercussions or consequences for students. No major decisions are based on low-stakes assessments. An in-class speaking activity or a class exit slip are examples of low-stakes assessments.

Medium Stakes – these assessments may have some consequences for students. Graded assessments, portfolio pieces, or unit-tests are medium-stakes.

High Stakes – these assessments have serious consequences for students. Exit assessments for an ELD program, college entrance exams, and proficiency tests used for employment purposes are high-stakes assessments.

Criterion-referenced – the scores on criterion-referenced assessments are based on lists of criteria, such as Student Learning Outcomes or curriculum standards. Math tests, in which correct processes and answers lead to high scores, are criterion-referenced.

Norm-referenced – the evaluation of a norm-referenced assessment compares the students in a given population. Students’ work is assessed against the work of other students. For example, a norm-referenced college entrance exam would allow a university to only admit the top students in the testing pool but would not give information about how one group of students compares to another group that completed a different test. A norm referenced test would also allow a program to evenly split students into levels based on relative proficiency, which would be useful if the program had ideal class sizes but a flexible curriculum.

Assessment – assessment encompasses any of the work done to gather and analyze data from students in order to know about and report about their learning. Interviews, surveys, tests, journals, and roleplays are examples of typical language proficiency assessments.

Measurement – this is the work of assigning qualitative or quantitative information to observable phenomena, in order to report about and analyze said phenomena. For example, L2 proficiency is an observable phenomenon, and we measure it by describing standards and then assessing language-users’ attainment of those standards.

Test – a test is a systematic, planned way to gather data about a student and their skills, knowledge, or proficiency. Testing is a very common assessment procedure.

Evaluation – evaluation is the step beyond assessment; evaluation takes the result of assessment and uses it to judge the outcomes of a program. End of Course exams can be used to assess students’ learning, and the results of EOCs can be used to evaluate the effectiveness of teaching methodologies used in the course.

Universal Design – this is a philosophy that is followed in many or most fields that create products. Universal Design promotes designs that take into account any accommodations that may be necessary in order to make a product usable for anyone and incorporates those changes into the original product instead of requiring users to supplement or modify the end product. For example, if a can opener isn’t usable by people with low grip strength, then universal design would modify the product so that it requires less mechanical advantage for all users, rather than requiring that people with low grip strength use after-market modifications in order to use the can opener. For assessment and materials design, universal design principals include incorporating scaffolding activities, glossaries, and models for all students, not just for ELLs.

Terms about Scoring:

Discrete-point test – each task or test item measures a different skill or piece of knowledge. For example, a task on a speaking test that requires saying a single word in order to assess pronunciation of a single phoneme.

Integrated test – tasks or test items may measure many skills simultaneously. An item on a speaking test that requires a long turn and assesses pronunciation, functional language, vocabulary, and grammar simultaneously is an integrated test.

Selected response – these tasks give students both the “question” and the “answer”, and students need to recognize the correct answer for the questions. Examples include multiple-choice, matching, true-false, and ordering tasks.

Limited production – these tasks provide the question that require students to supply a small amount of information (small could be as little as one letter or morpheme, or as much as a few sentences). Examples include fill-in-the-blank, graphic organizers, sentence combining, and short answer tasks.

Deletion-based – these tasks delete information from a text and ask students to accurately or appropriately replace that information. For example, a cloze task deletes words in a text and students must recognize metalinguistic information such as part of speech, plurality, and register in order to reconstruct what the missing word might be.

Translation – translation requires knowledge of grammar, vocabulary, register, discourse, and pragmatics in both languages in order to accurately and appropriately convey the same meaning. Translating to the L2 typically demonstrates productive proficiency, whereas translating to the L1 typically demonstrates receptive proficiency.

Extended production – these tasks ask students to generate a large amount of language, which can assess higher order language concerns such as discourse-level organizational language (like signposting and transitions) or genre-appropriate conventions (such as using past tenses to set up a scenario and present tenses to describe the action when narrating an anecdote). Other examples of extended production tasks include recall tasks, summaries, dictation, spoken responses, monologues, role plays, and oral interviews.

Rubric types:

Holistic rubric – a holistic rubric is one in which results are given on a single scale – rubrics that simply label a product as “needs improvement/satisfactory/excellent” are holistic. These are less labor intensive for a teacher and clearer to understand for young learners, but they don’t provide a lot of specific feedback for older or more advanced learners.

Analytic rubric – an analytic rubric breaks the assessment down into components and then rates each of those components separately. These rubrics may assess components such as discourse, grammar, vocabulary, function, content, or participation; a student who produces a grammatically correct essay that doesn’t answer the prompt could get a high score for grammatical accuracy and a low score for content, allowing them to understand where their work can improve. Analytic rubrics that are not task-specific can be used for a variety of assessments, which reduces teacher workload and improves student understanding of the rubric.

Task-specific rubric – these rubrics allocate points for the different criteria of a task. For example, building a volcano in science class may be assessed on a rubric that includes creating the proper shape of a volcano or explaining the chemical reaction in the presentation. The rubric can by highly specific and act as a checklist for the students, but it can only be used for a single task.

The 6 Principles of Assessment are ways to analyze the efficacy of assessments.

Validity – does the assessment measure what it purports to measure? For example, a grammar test that requires specific content or vocabulary knowledge may not be a valid assessment of grammar.

Reliability – if the assessment is repeated, are the results consistent? For example, if two graders would give wildly different scores to the same product, the assessment lacks inter-rater reliability.

Practicality – are the required resources feasible for the number of and frequency of assessments given? For example, a 1-hour interview may be a very accurate proficiency measure, but if a program has one interviewer for 2000 students who are assessed every month, the assessment isn’t practical.

Equivalency – does the assessment match the curriculum? For example, if students only do multiple choice worksheets in class and the assessment is an essay, the assessment isn’t equivalent.

Washback – do task types on the assessment change teaching practices? Examples of negative washback include increasing multiple-choice worksheets and decreasing group projects just because an EOC is multiple choice; examples of positive washback include increasing focus on academic language because summative assessments includes an academic presentation.

Fairness – is the assessment equitable for all students? For example, an assessment that requires knowledge of pop culture may not be fair to a student who doesn’t consume popular media.

US K-12 School Terms:

US – United States of America.

K-12 – Kindergarten to 12th grade. These are the years that the majority of Americans attend school, with kindergarteners typically 5-6 years old and 12th-graders typically 17-18 years old. Students may go to school before kindergarten (pre-K, preschool, or Head Start, for example), and depending on the state, students may be able to drop out at 16. Education in the US is compulsory; public schooling is free; and home schooling or private schooling are widespread but much less common forms of schooling.

Content Classes – content classes are the traditional school subjects like math, science, or language arts.

Annual Subject Testing – these are the tests that all students take in elementary and middle school. Many states have state-wide tests on subjects like math, science, and language arts. These tests are often NOT part of students’ grades – you don’t fail 6th grade science due to your state-wide annual science test score.

End Of Course exams – these are tests taken by high schoolers at the end of the semester or school year, and they ARE part of the students’ grades. For example, 25% of your Algebra I grade may come from your score on the state-wide EOC.

EL or ELL or EB – English Learner or English Language Learner or Emergent Bilingual. These students are in the process of learning English whose proficiency is in the range that they qualify for ELD programs.

ELD programs – English Language Development programs. These are programs with the goal of increasing academic English language proficiency in ELLs, with the goal of them learning subject content in US schools. In non-Bilingual contexts, students in the US study in English, so ELD programs intend to give ELLs the level of English proficiency necessary to learn the required curriculum for their current grade-level.

LUS – Language Use Survey, sometimes HLS Home Language Survey. This is a survey that intends to identify ELLs. It includes questions about parents’ and students’ language use, in order to know if a student needs to be screened for English proficiency or if they qualify for ELD services.

LIEP – Language Instruction Educational Program. The program(s) that a school district provides for ELLs. In the US, a school district may provide service to a city or a county. They are controlled to some extent by the state government, which is controlled to some extent by the federal government. They control to some extent each of the schools and programs in their district. ELD programs are generally controlled at the district level, but each school may have its own ELD programs and policies – some schools in a district may have dual immersion programs, others may have dedicated ESL classes, others may have ESL Specialist co-teachers. The benefit of a system like this is the flexibility to provide the most appropriate services for the local population – a school where Spanish-L1 students make up 50% of the population has different needs than a school where 7% of the students are ELLS but they all have different L1s. The drawback to a system like this is a lack of uniformity for migrant students and difficulty in evaluating all of the different programs.

Migrant – Migrant students are students whose families’ work jobs that require them to move often during the school year; often these are workers in the agricultural sector. These students will move from school to school and therefore will be identified and placed into ELL programs, if they are not L1 speakers, over and over again. Plans for sharing information between school systems are important, and differences in LIEPs and ELD programs in different areas can make these transitions difficult.

BICS – Basic Interpersonal Communication Skills. These are the language skills needed to hold a conversation, go shopping, use transportation, etc.

CALP – Cognitive Academic Language Proficiency. These are more complex, academic language skills that students need to acquire in order to be successful in content classes.

Are there more terms than this? Of course! But this is a list of terms to get you started.

[my-efl-blog]

Potential Summer Reading Class Project: Collaborative Rubric Design Project

This project was inspired by Dong-Shin Shin and Tony Cimasko’s presentation at AAAL 2024, “ESL Students’ self-assessment of multimodal writing for metalanguage development”, as well as the ELT Journal article “Implementing Rubric Co-Construction in ESL Writing Teaching” by Tong Zhang and Zhenjie Weng. Both projects, one for international graduate students and one for advanced undergraduate multilingual writers, focused on the benefits of rubric co-construction for student understanding and use of rubrics in the classroom.

In my context, I teach undergraduate and graduate international students in the same classroom. The undergraduates exit one term earlier than the graduate students. I also teach reading instead of writing. However, I do work with rubrics a lot in my classes, as I focus my reading instruction and assessment on critical thinking and demonstration of understanding, as well as active participation and discussions.

One of my assignments, which students complete 2-3 times per course, was adapted from an assignment by Alicia Ambler, an ESL instructor at the University of Iowa. Her “Reading Circles” assignments puts students in charge of discussions of shared readings – they complete the readings outside of class and then are responsible for facilitating their own discussions of the material, minimizing teacher-led and teacher-focused class time.

Here is a version of the rubric I made for this assignment:

The points above the rubric are task-specific, but the rubric itself is analytic. The rubric could be used for any assignment that had both a written component and group discussion component. In fact, I use variations on this rubric and these descriptors for other critical thinking activities in my classes. Although I understand the use of a holistic rubric for lower level and/or younger students, I do not think that a holistic rubric provides enough information about assessment criteria for university-age B1-B2 learners.

The rubric has been adapted over the past few semesters in which I have used this project. In its original form, the scale only contained scores of 1-4, which weighed the participation and critical thinking elements of the project as less important than completing the pre-discussion tasks, which wasn’t my intention. This change increased the content validity of the assignment, as well as the equivalency, by assessing students on what I intended to assess and making the focus of the assignment align with the goals I express during teaching. As I am the only person doing this project, and as the only instructor for this course at this level, I have not established inter-rater reliability of this rubric. I don't think there has been any washback from this assessment, as I've changed the rubric to reflect my intentions instead of changing my teaching to reflect the rubric.

I always share the rubric with my students before the first Reading Circles assignment, but after attending the presentation and reading the journal article, I realized that I don’t know how much my students understand the rubrics or how invested they are in meeting the goals of the assignment as measured by the rubric.

Student self-assessment is a great way for students to more clearly understand what they are being evaluated on, and rubric co-construction is a way for students to have a say in the skills and products on which they will be evaluated. I do not often use self-evaluation in my reading class precisely because it is a reading class – much of self-evaluation is based on comparing a product to a set of standards, but there is rarely a product in reading. (I tend to avoid assigning too many reports or reflections in my class because I don’t want to assess my students on writing – they have a whole other class for that; at the same time, I find reading skill and reading comprehension tests, which do measure reading ability, to lack the kinds of critical thinking and creative thinking that reading should foster at this level. I want my students, who will be entering university after leaving my class, to know that they will probably not simply be asked to read and memorize passages but will instead be asked to demonstrate their reading and understanding in a variety of ways.)

I also usually allow my students to very strongly affect our course, giving them quite a bit of agency. We choose topics and assignments together. We negotiate due dates and tasks. They often have options for what assignments they want to complete and how and when they will complete them. Allowing my students to create their own rubrics would be a fitting and lovely addition to my course design, as I believe that students in my context know their goals and what they need to achieve them, often better that I do as an outsider. Having them decide on the important parts of the assignment and how they should be rated would be another way to increase their participation in the course.

The first part of the collaborative rubric design projects that Shin & Cimasko (20204) and Zhang & Weng (2024) did in their courses was familiarizing their students with the final products of the project and asking them to evaluate those projects. Shin & Cimasko, who were working on multimodal composing, showed examples of infographics and slideshows and asked students to decide which ones were effective or ineffective and why. Zhang & Weng had an annotated bibliography project, and they asked their students to describe what made example annotated bibliographies good or bad. This is one of the drawbacks to potentially implementing this project in my course – the product that students create at the end of the project is only one half of the full Reading Circle, acting as prep work for the discussions. It would be more difficult, and take up more of our limited class time, to evaluate the hypotheticals of a class discussion that the students haven’t yet participated in.

The next steps, collaboratively building the rubric, would be interesting, but again, would take up class time that I already do not have. The Reading Circles discussion projects are based solely on outside-of-class reading and work,  wherein only the project presentation and the 30-minute discussions take place during classtime. Adding the rubric creation would add considerably more time that I do not currently have access to.

One part of the project that I haven’t yet implemented in class is the reading summary and reflection, again due to time constraints. However, a summary and a reflection are products that students could evaluate and create rubrics around. If I were to give students a holistic rubric for their fist summary and reflection, such as

Then, after they have the experience of their first (marked, but not for class points) summary and reflection assignment, we could transform this holistic rubric into one that adequately and analytically addresses the features that make a good summary and a good reflection. This would allow the students to learn and display their understanding of the genre, and would allow them to self-assess their own work and their peers’ work before submitting their assignments.

Classes start in one week, so I may not have time to fully implement this project for the summer session, but I will strongly think about doing this.

References:

Ambler, A. (2022, September 23-23). Small group reading:  An easy method for engagement [Conference presentation]. MIDTESOL 2022, Kansas City, MO, United States.

Shin, D.S., & Cimasko, T. (2024, March 16-19). ESL students’ self-assessment of multimodal writing for metalanguage development [conference presentation]. AAAL 2024, Houston, TX, United States.

Zhang, T. & Wen, Z. (2024) Implementing rubric co-construction in ESL writing teaching. ELT Journal 78(1), pp. 32-41. https://doi.org/10.1093/elt/ccad033