Innovation Blog

Introduction

Research and development of software to harness artificial intelligence for scoring student essays has many significant obstacles. Using machine learning techniques requires massive amounts of data and computing power far beyond what is available to the typical secondary public school. The cost and effort to devise such technology does not seem to be juice worth the squeeze, since it is still more time efficient and cost effective to just have a human do the job. However, the potential exists to devise AI-assisted grading software whose purpose is to increase the speed and accuracy of human raters. AI grading that is “assisted” applies natural language processing strategies to student writing samples in a narrowly defined context and operates in a mostly “supervised” fashion. That is, a human rater activates the software and may make scoring judgments with the advice provided by the AI. A promising area for this, more narrowly contextualized application of artificially intelligent natural language processing, is in scoring summaries and short answer tests. This also poses interesting possibilities for automated coaching for students while they write. This study examines a set of algorithms that derives a suggested score for a secondary level student summary and short answer test response by comparing a corpus of model answers selected by a human rater with the student work. The human rater stays on duty for the scoring process, adding full credit student work to the corpus such that the AIs is trained and selecting student scores.

Features of Text for Comparison

The AI examines the following text characteristics to evaluate a student work by comparison to one or more models:

• “readability” as determined by the Flesch-Kincaid readability formula
• the percent difference in number of unique words after pre-processing. “Preprocessing” refers to text that has been scrubbed of irrelevant characters like HTML tags and extra spaces, has been lemmatized, synonymized, and finally stemmed.
• intersecting noun phrases
• Jaccard similarity
• cosine similarity of unigrams
• cosine similarity of bigrams
• cosine similarity of trigrams
• intersecting proper nouns
• cosine similarity of T-score
• intersecting bigrams as percent of corpus size
• intersecting trigrams as percent of corpus size
• analysis score2

The program first compares the student text to each model using cosine similarity of n-grams. The most similar model in the corpus is then compared to the student work. Four hundred and twenty-six short answer questions that had been scored by a human rater were compared using the algorithm. From these results was developed scoring ranges within each text feature typifying scores of 100, 85, 65, 55, and 0. Outliers were removed from the dataset. Next, sets of student summaries were scored using the ranges for each text feature and the program’s scoring accuracy was monitored. With each successive scoring trial, the profiles were adjusted, sometimes more intuitively that methodically, until over the course of months the accuracy rate was satisfactory.

When analyzing a student writing sample for scoring, the score on each text feature is compared to the profiles and the program keeps a tally of matches for each scoring category (100, 85, 65, 55, and 0). The “best fit” is the first stage of suggested score. Noun phrases, intersecting proper nouns, and bigram cosine were found to correlate most highly with score matching the human rater, so an additional calculation is applied to the profile scores to weight these factors. Next, a set of functions calculates partial credit possibilities for scores in the category of 94, 76 and 44 using statistics from the data analysis of the original dataset of 426 samples. Finally, samples where analysis are important in the response have their score adjusted one final time.

Analysis score” is a metric devised to evaluate the “analytical richness” of a student writing sample​.

The development of the scoring ranges for text features proceeded somewhat methodically and at times more intuitively or organically. Over the course of months, when error patterns in AI scoring became apparent, adjustments were made to improve performance. Natural language processing, even at this basic level, is very demanding on computer memory and processing resources. At this writing, the server running this software has 6GB of RAM and work is often being done on the code to reduce processing time. One strategy is to store both “raw” and processed versions of the student work products as they are written so that processing time can be shortened at the end. the corpus of model responses is also saved in this way.

Training the AI

Upon creation of an assignment, the teacher can save model responses to the corpus. Once students have completed the assignment, the teacher can begin by reviewing and scoring the work product of students who usually score full credit. Upon confirming that these are indeed full credit models, the teacher can click a button to add the student sample to the corpus of model answers. The software limits the teacher to five models in short answer testing and seven models in composition assessment.

Once trained, the teacher can run the scoring algorithm on each student submission. At this writing, processing takes about nine seconds on average per sample, depending on the text size. This program works best for assignments where there is a narrow range of full credit responses. Its primary purpose is to score writing samples by comparing to a limited number of full credit responses. Its strength is in recognizing similar meaning across texts in varying ways to say the same thing. This program does not assess spelling or technical / mechanical writing conventions, although it does rely on student accuracy for scoring to the extent that adherence to certain conventions are necessary for the program to operate. Examples: proper noun count requires that students capitalize them; sentence count requires that students apply standard rules of punctuation.

The debriefing is a powerful tool for teaching to which students readily respond. I have had students tell me they really felt they benefited from these activities.

In general, the debriefing is a lesson that consists of analyzing student errors and offering corrections. Naturally, this is done anonymously so as to avoid embarrassment. It is particularly useful in teaching writing, computer programming, and similar complex tasks that can be broken down into smaller skill sets for training.

For example, when I teach French composition, I select errors from student compositions and present them anonymously to the class. I explain the error, I correct the error, and students then proceed to practice recognizing and correcting the error themselves. Innovation has a number of these lessons for sale at one of our online stores.

By way of another example, when teaching social studies, I help students develop skills for analyzing historic documents using constructed response tasks. This assignment calls upon students to provide historical or geographical context for a document and then to analyze its reliability and relationships with other documents such as cause-effect, turning point, or to compare and contrast. Especially for the reliability element, it is useful to display student work, both strong and weak, for commentary and analysis.

If you’ll indulge a final example, when I teach persuasive writing I like to display student samples in class and we can practice together identifying claims, warrants, rebuttals, and so forth. We can weigh the strength of arguments and of writing style.

21st century learning spaces are designed to facilitate debriefing for all sorts of tasks. Since this is a key feature of my own teaching practice, it is really baked in to the Innovation platform:

• Multiple-choice: Teachers can start up a “live session” after a test to review. In the live session, the host displays the question and students join the session from their own devices and interact. (Kahoot! is a well-known example).
• Short Answer: Teachers can initiate a “live session” for short answer that works the same way.
• Jeopardy-Style Review: It is easy to select questions from a set of recent tasks such as quizzes or short answer prompts and then generate a Ventura game.
• Analytics: Innovation has a complete set of analytics tools for all online tasks. This includes multiple-choice and short answer item analysis, standardized (“curved”) grading functions, and statistical analysis tools to evaluate and compare assessments. Analytics tells teachers what to debrief; what has priority for review and remediation.
• Item Analysis: The test “master” for each assessment presents an item analysis of student work and a ready-to-display version of the test.

Debriefing lesson planning can be very arduous. It can be time-consuming to create a slide show or document with copy-pasted elements of student work submissions for analysis. The Innovation platform facilitates this in multiple ways with a few clicks, in true form to a strong 21st century learning space.

Marc Prensky’s 2001 article “Digital Natives, Digital Immigrants” sparked a lot of conversation, even debate, about the use of computers in education. Mr. Prensky proposed that students who grew up with digital devices integrated into their lives “think and process information fundamentally differently from their predecessors” and he posited that it’s “very likely that our students brains have physically changed” as a result of how they grew up (Prensky, 2001). Prensky characterized the older generation as populated by digital immigrants, whose more limited command of computer use was an obstacle to teaching these digital natives. His recommendations focused on what we would now call “gamification” of learning; “edutainment”.

In the two decades since the coining of the term, information and communication (ICT) technology has changed and many challenges legitimately arose to Prensky’s depiction.

I need a term for young people with extensive experience and skill with the digital world of commerce, entertainment, and social media. I would like to borrow Marc Prensky’s term “digital native”.

ICT Skills

The digital native is said to “possess sophisticated knowledge of and skills with information technologies” and to “have particular learning preferences or styles that differ from earlier generations of students”. (Bennett, et al. 2008).

Prensky, in my view, was asking teachers to bend their lessons away from sound educational practice to match the entertainment that students were used to experiencing when using computer technology. I believe this was his mistake.

The ensuing decades saw some challenge to these notions. Scholars posed legitimate challenges for the basis on limited and anecdotal evidence (Bennett, et al. 2008). Further research in the first decade after Prensky’s papers, while confirming the near ubiquitous use of digital devices by adolescents, found that “a significant proportion of students had lower level skills than might be expected of digital natives” and that “only a minority of the students (around 21%) were engaged in creating their own content and multimedia for the Web” (Kvavik, Caruso & Morgan, 2004, as cited in Bennett, et al. 2008).

I would propose that computer software, especially at this early period in the 2000’s but even still today, has been designed primarily for commerce, entertainment, and socializing. This is what Prensky and his supporters were seeing students use; skills in using this software was what students were developing. I submit that software designed to sell, entertain, and socialize has some features that are not supportive of an effective educational tool. Education needs platforms that engage students in what we know to be good learning practices. Prensky, in my view, was asking teachers to bend their lessons away from sound educational practice to match the entertainment that students were used to experiencing when using computer technology. I believe this was his mistake.

But we know that something is different. Those of us who were born before 1980 can attest to changes in adolescents that is associated with the digital age. In my view, this shift is better understood in less dramatic terms: as a cultural change of the sort that has been going on in civilization for millennia. The point is well taken that educators should adapt to this kind of cultural shift.

Thinking and Information Processing

Prensky posits that the extensive use of digital devices has changed how students think and process information. Digital natives are characterized as “accustomed to learning at high speed, making random connections, processing visual and dynamic information and learning through game-
based activities” as well as multi-tasking (Bennett, et al. 2008). Not only do these assertions lack evidence, but I would argue that none of them lend themselves to right learning practice. Multi-tasking does not work for human beings (Napier, 2014), since it interferes with encoding in long-term memory and increases cognitive load.

In addition to the lack of evidence to support the assertion that digital natives thinking is significantly different from previous generations, Bennett et al note that “the claim that there might be a particular learning style or set of learning preferences characteristic of a generation of young people is highly problematic.”

While there is some overlap between engaging in digital experiences for commerce, entertainment, and social media and for education, experience has taught me that there are some fundamental differences. Digital teaching platforms should reflect sound educational practice and practical application to classrooms.

But we know that something is different. Those of us who were born before 1980 can attest to changes in adolescents that is associated with the digital age. In my view, this shift is better understood in less dramatic terms: as a cultural change of the sort that has been going on in civilization for millennia. The point is well taken that educators should adapt to this kind of cultural shift. However, where I diverge from Prensky is here: effective teaching practice does not mean that we adopt the ways of commerce, entertainment, ans social media in wholesale fashion.

Beyond the myth of the “digital native” (2019) by Carlos A. Scolari

What are young people doing with media? An alternative framework for understanding how adolescents use technology is one which maps out the skills adolescents possess across different digital media. “[T]ransmedia skills are understood as a series of skills related to the production, management and consumption of digital interactive media” (Scolari, 2019).

“[T]ransmedia literacy turns [the] question around and asks what young people are doing with the media. Instead of considering young people as consumers taken over by the screens (television or interactive screens, large or small), they are considered ‘prosumers’ able to generate and share media content of different types and levels of complexity” (Scolari, 2019).

Let us redefine the “digital native” as the typical adolescent who has an uneven skill set for media, having come from immersive digital experiences in games, commerce, and social media. Let us recognize that, while there is some overlap, the approach students need to be taught to cultivate toward digital devices as learning tools has some important and very fundamental differences from what they have done before.

“[T]he concept of “digital native”, understood as a young person who “comes with a built-in chip” and who moves skillfully within digital networked environments, shows more problems than advantages” (Scolari, 2019).

In terms of transmedia skills, we still have some things to teach students. Strong skill sets are not evenly distributed (Scolari, 2019). Students come to us skilled in the areas they use most, engaging in entertainment, commerce, and informal social interaction. Skills linked to production are usually strong in adolescents, but those associated with related to ideology and values are more limited.

Scolari writes: “at an individual level, a young person who demonstrates that they have advanced photographic production skills (creation of memes) or audiovisual management skills (a YouTube channel) can, at the same time, have less developed abilities in, for example, detecting stereotypes or managing privacy.”

A Revised Definition

We who have been teaching for decades know that there is a cultural shift going on that is related to information and communication technology. The term “digital native” coined by Prensky lacks a firm foundation and may well be more a reflection of the time it was conceived than anything else. In addition, it seems to me that the current generation of beginning teachers are in no way digital immigrants, having themselves grown up with extensive digital experience.

Adolescents use technology extensively and this does affect their starting point for education. We teachers can capitalize on the skills they possess already in the classroom while refining those they may lack and discouraging adolescent practices that are detrimental (like attempting to multi-task).

Let us redefine the “digital native” as the typical adolescent who has an uneven skill set for media, having come from immersive digital experiences in games, commerce, and social media. Let us recognize that, while there is some overlap, the approach students need to be taught to cultivate toward digital devices as learning tools has some important and very fundamental differences from what they have done before. The platforms we use to teach them in the digital world should reflect this. I would term these apps “21st century learning spaces“.

Sources

• Bennett, S.; Maton, Karl A.; and Kervin, Lisa: The ‘digital natives’ debate: a critical review of the evidence (2008).
• Napier, Nancy: The Myth of Multitasking (2014).
• Prensky, Marc: Digital Natives, Digital Immigrants (2001).
• Scolari, Carlos A. Beyond the myth of the “digital native” (2019).

Striking a Balance: Leveraging Innovation Teaching Platform for Effective Pedagogy

In the ever-evolving landscape of education, the debate between teaching to the test and teaching to the standard persists, presenting educators with a perpetual challenge: how to strike a balance between preparing students for standardized assessments while also ensuring they acquire the essential knowledge and skills outlined in educational standards. Fortunately, innovative teaching platforms like Innovation Assessments offer a multifaceted approach to pedagogy that can help educators navigate this delicate balance and promote best practices in teaching and learning.

Utilizing Comprehensive Learning Management:

At the heart of the Innovation teaching platform lies a comprehensive learning management system designed to streamline content creation, assessment, and student engagement. By leveraging this robust platform, educators can seamlessly align their instruction with established educational standards while also incorporating targeted test preparation strategies.

Balancing Test Preparation and Conceptual Understanding:

Innovation Assessments provides educators with the tools to integrate test preparation within a broader framework of standards-aligned instruction. Through features such as customizable assessments, practice drills, and automated grading, educators can ensure that students receive targeted support in mastering both the content knowledge and test-taking skills necessary for success on standardized assessments.

Fostering Inquiry-Based Learning:

One hallmark of effective teaching practice is the promotion of inquiry-based learning, which encourages students to explore, question, and construct their understanding of the world around them. With its diverse array of multimedia resources, interactive activities, and collaborative tools, the Innovation platform empowers educators to create engaging learning experiences that foster curiosity, critical thinking, and problem-solving skills in students.

Personalizing Instruction to Meet Diverse Needs:

Recognizing that every student is unique, the Innovation platform offers personalized learning features that allow educators to tailor instruction to meet the individual needs, interests, and learning styles of their students. From adaptive assessments that adjust to students’ proficiency levels to customizable learning pathways that provide targeted remediation and enrichment, educators can ensure that each student receives the support and challenge they need to succeed.

Empowering Educators with Data-Driven Insights:

In addition to facilitating student learning, the Innovation platform equips educators with valuable data-driven insights into student performance, engagement, and growth over time. By analyzing student data and trends, educators can identify areas of strength and weakness, inform instructional decision-making, and implement targeted interventions to support student success.

Promoting Continuous Professional Growth:

Lastly, the Innovation platform serves as a catalyst for educators’ continuous professional growth by providing access to a wealth of resources, professional development opportunities, and collaborative communities. Through ongoing training, peer collaboration, and reflective practice, educators can refine their pedagogical approaches, stay abreast of emerging best practices, and ultimately elevate the quality of instruction in their classrooms.

In conclusion, while the debate between teaching to the test and teaching to the standard may persist, innovative teaching platforms like Innovation Assessments offer a holistic solution that transcends this dichotomy. By harnessing the power of technology to integrate test preparation within a broader framework of standards-aligned instruction, foster inquiry-based learning, personalize instruction, empower educators with data-driven insights, and promote continuous professional growth, the Innovation platform emerges as a potent tool for promoting best practices in teaching and learning. As educators embrace this multifaceted approach, they can navigate the delicate balance between preparing students for standardized assessments and equipping them with the knowledge, skills, and dispositions needed for success in the 21st century and beyond.

During the pandemic, many office workers moved to remote work from home. This precipitated a rise in monitoring software that companies could use to ensure that, being at home, workers were productive. An article in Forbes Magazine from 2021 reports that “[d]emand for worker surveillance tools increased by 74% compared to March 2019.” This rush to monitor and micromanage turned out to be unnecessary, as fears of a loss or productivity proved unfounded and “94% [of companies] reported that worker productivity either stayed at the same levels or improved.”

But this is not the case with adolescents.

The traditional classroom had to be a “very supervised” place because, by virtue of the fact that they are immature, most of our charges need guidance to get back on track. It is one reason why remote learning went so badly for many youngsters: it is not in the nature of most to be focused. The executive functioning needed to ignore distraction, set goals and reasonable timelines for work, even to break a longer task up into smaller, achievable segments is rarely present in adolescence. Until this develops, the role of the instructors includes teaching this skill and guiding students to follow the right course. Teaching with digital devices at present has reduced much of this supervisory ability. 21st century learning spaces would come with an array of monitoring and accountability features.

Data […] promotes accountability, but it also puts the student potentially in the driver’s seat and that is what developing executive functioning is all about.

I recall an instance where a student of mine was completing the essay portion of an examination remotely. I was able to monitor his examination in real time using software that shared his screen with me. When I noticed that he was typing sentences that appeared beyond his ability, I was able to google those phrases and find the source he was plagiarizing from online (he had his phone with him to cheat). This monitoring software allowed me a virtual way to simulate normal classroom supervision and to take the natural step of concluding the examination and award no credit.

After the pandemic, I continued using digital tools for student work. My students all had ChromeBooks. I had a student who was clever in taking advantage of a certain doubtfulness about technology by some adults around him. Faced with an incomplete assignment, he would claim he did it and that the app must have “lost his work”. He would claim that it “did not save”. In a traditional classroom, I would have seen his paper and whether it was written on, but the digital work did not include this monitor yet. I adjusted the software for his writing assignments to report when a response was deleted, when a student left the browser page for another, when students pasted text in, and even double-check the server to ensure an answer was saved. These application features returned important accountability assurances that were initially lost when moving to digital devices.

As time went on, my colleagues and I devised further modifications to the software at Innovation. I developed the “proctor” on important apps for testing and writing.

The proctor records data about the page and the students’ interactions with the assignment. Depending on the particular assignment, it records when work has begun, when an ancillary resource like a video has successfully loaded, when a student leaves the page and for how long, when text is pasted in, and when answers are saved. The proctor is visible to students (see illustration above) so they know their work is being monitored.

My colleague in the science department uses a flipped classroom technique. He made a great suggestion for the development of an app to monitor student interaction with a video assignment. As a student watches a video assignment, proctor records events like start video, stop video, how long between pauses, when the video ended, and how long the student was there.

The tracking monitor helped maintain a system of accountability for students.

Besides the proctor, Innovation tracks student activity around the site. The auditor maintains a record of logging in, accessing a course, starting a task, saving work, getting a score, etc.

The critical work of developing executive functioning in adolescents can be enhanced by providing youngsters the kind of data that, if they attend to it, can inform their decisions about what they should do. The proctor and other reporting tools are available to all students. Although consequences for missing the mark on attention to task can and should be part of the program, it is not great practice to be all sticks and no carrots. Objective data on what a teenager is actually doing (rather than what they remember they did or want you to think they did) can be the focus of discussions about on-task behavior and how the individual can take responsibility for it. We can take a look at performance on as assignment and examine on-task behavior related to its production. Could on-task behavior have improved the final product?

Data like this promotes accountability, but it also puts the student potentially in the driver’s seat and that is what developing executive functioning is all about.

My first experience using computers to teach was in 1993 when I was teaching French at a small, rural school in the Adirondacks. When the US Air Force base in Plattsburgh, New York closed, it donated its old computers to regional schools. They were “286s” that basically only ran, well, BASIC! Fascinated, I taught myself BASIC and started writing programs to drill vocabulary and verb conjugations. I really have not stopped coding educational apps since. It turned into a very stimulating hobby and very useful for my teaching practice. (When we were doing remote learning during the pandemic, my students were already operating in a digital classroom and remote learning was easy!)

Those old “IBM Compatible” computers were designed by computer engineers for the business world. The input interface was a keyboard (for typists and secretaries who, in the old days, were the only ones in the office who needed to learn to type). The big, boxy device was designed to sit on an office desk. The software ran programs like simple word processors and spreadsheets. These are also office utilities. You begin to see where I am going here?

Computers like these were initially devised to increase the efficiency of offices. They were for business.

I saw my first computer game at my cousin’s house when I was in middle school. His family had an Atari system. That was also where I first saw coding in BASIC. I saw the early computer game, Pong, at a restaurant when I was in my early teens. And there was Space Invaders at the arcade in Old Forge… And there was an arcade at the Fairmount Fair mall … You begin to see where I am going now?

Computers like those were devised for entertainment. Whether for entertainment or for commerce, the whole paradigm was intended for purposes other than education.

I was computer coordinator in my school at the time when computers and internet first migrated into schools. I was there in the heated discussions over whether we use Mac or IBM. I helped wire our school for internet and networking. I ran cables through crawlspaces and attic spaces in the 70-year old school. Talk was about what kids will need in this computer age and mainly we felt they needed skills associated with business, so when we adopted devices that were primarily designed for offices and plunked them in classrooms we figured it was good. Computer labs were de rigueur in the late 90s and early 00’s. Each classroom in my school had four or five PCs, which we built in a basement workshop. We were running Windows, Microsoft Office, PowerPoint, and so forth. We were trying to bend a device and its software that was designed for business and entertainment to classroom use.

By 2012, smartphones had become ubiquitous among students and this led to a number of other problems. Young people mainly play games and socialize on their devices. Socializing mediated by social media platforms has made changes and caused problems we are only beginning to unravel.

People use digital devices to engage in commerce, participate in entertainment, and to socialize. These being the principle purposes of the devices, they shape the course of design not only of the physical device itself but the software and features that the devices host.

My thesis is that when we brought these devices into schools, right from the beginning, we were trying to repurpose things meant for commerce, entertainment, and socializing into an environment where none of those was our pedagogical purpose. Sometimes it fits, sometimes it does not.

What do apps look like that are devised for education?

I use the phrase “21st Century Learning Spaces” to refer to a digital device and its software that fit education well because it is designed for that purpose and not repurposed from some other setting. I don’t build digital devices, but I do write software. For the past ten years, my colleagues and my students and I have been dissatisfied with bending apps to serve an educational function and having it not quite fit. Every try to use Google Forms to give a quiz? Is that really easier than what we used to do?

I hope you will join me in the next few posts and permit me to unpack the 21st Century Learning Space concept with its implications, limitations, and applications to education. I hope teachers will try out Innovation Assessments and see whether I have managed to meet some of the criteria for educational apps in a 1:1 device classroom.

Play is an important part of growing up. Trying out roles, acting out adventures… imagination! Here are some improv activities adapted to the classroom with some rubrics.

I first saw these on a TV show in the ’90s called “Whose line Is It Anyway?” where comedians performed improv scenes. I adapted them to teaching French at the time in order to develop conversation skills and improvised speaking in a fun lesson. When I switched to social studies in ’04, I sometimes used these at the end of a unit.

Stranger in Town

This activity must have a real name among improv comedians, but I just called it this. Three volunteers come to the front of the room. One goes out into the hallway briefly, out of earshot. The two who are left quickly agree on a scene they will perform for 3 minutes. The person in the hall, the “stranger”, is invited in and the scene commences. After 3 minutes, the stranger has to guess who they are based on the scene, in which, by the way, they had to participate. Imagine walking into a room and finding out you’re Henry VIII!

Here’s a sample setup for a social studies class where students had chosen their activity in advance and I had prepared it. To save time, I would sometimes generate the scenes for stranger in town and have teams roll the dice to see which to play. The identity of the person playing “stranger” was always a surprise!

Press Conference

The volunteer goes to the front of the room, preferably to a lectern, and pretends to be some famous person in history. The class are reporters whom she will call upon to ask questions. Pretend you’re Hannibal and you’ve just gotten your army with your elephants over the Alps. Can you answer some questions from our reporters before you disembark?

Although I often had students prepare for these if they weren’t confident, it can be an improv exercise for the brave and bold. Let it be a surprise to the volunteer whom they will play!

Newscast

Two volunteers for newscast come to the front of the room to play reporters who are to report on a scene from history. They don’t know what they scene is until they get there! (Presumably this is at the end of a unit and the scenes are from the current topic of study). One student plays the anchor at the desk which the other is “in the field”, a reporter on the scene. A third student may participate as a bystander whom the reporter will interview.

Artifact

In artifact, students create a quick construction paper cutout of some object associated with our unit. It could be an Egyptian scroll or a Greek sword or Thomas Jefferson’s quill. The student presents the artifact to the class as an archaeologist at a conference. For teaching world language, the artifact is something from our current vocabulary or reading.

I used all of these to teach French as well. For teaching world language, these scenes would not be about history, but of everyday life. Stranger in Town could be a scene in a restaurant and the stranger is a waiter serving a fussy customer. Press conference or newscast could be an event of current interest.

Improv is fun, but may not be for everyone.

Improv is fun. If you have time for it, dive in! Some kids are uncomfortable with this and I never made them play. I confess that in my later years teaching social studies, I was forced by time constraints to abandon these for my older kids. The demands of curriculum and remediation and state tests were such that the minimal content reinforcement provided by these activities, well, the juice was just not worth the squeeze.

Not all classes are right for this kind of thing. If you have a middle school class that has trouble self-regulating or is over-excitable, this may not be a good idea. It also does not work well with students who feel very uncomfortable in ambiguous situations or in performing. You will know who can benefit from this and who might not because you know your kids.