Failure to Disrupt, Justin Reich Flashcards

1
Q

1997, Ken Koedinger and colleagues published “Intelligent Tutoring Goes to the Big City” - the blended model

A

Students learned in traditional settings for most of their class time and then spent about one day a week using math tutoring software

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

new technologies will not reinvent existing school systems

A
  1. new technologies are not wholly new; they build on a long history of education innovations.
  2. there are certain basic obstacles that time and time again have tripped up the introduction of large-scale learning systems.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Large-scale learning environments can be classified into three genres based on who creates the activity sequence for learners

A

instructor-guided
algorithm-guided
peer-guided

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

four kinds of obstacles that large- scale learning systems have encountered repeatedly over recent decades

A
  1. the curse of the familiar
  2. the edtech Matthew effect
  3. the trap of routine assessment
  4. the toxic power of data and experiments
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

INSTRUCTOR-GUIDED LEARNING AT SCALE

A

Massive Open Online Courses (MOOCs)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Higher education researchers sometimes describe the challenges of improving postsecondary education by referring to the “iron triangle”

A

cost, access, and quality

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

New Technologies, Old Pedagogies

A

one of the most useful dispositions in evaluating edtech is to regularly ask the question, “What’s really new here?”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

John Dewey

A

Dewey famously argued, “I believe that education, therefore, is a process of living and not a preparation for future living,” and he advocated for an approach to education that emphasized apprenticeship, interdisciplinary learning, and connections to the world beyond schools. Social constructivism—the idea that individuals construct new understandings from prior understandings in the context of learning communities—is one term used by education researchers to capture this family of pedagogies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Edward Thorndike

A

Thorndike believed that learning could be precisely measured, and he was an early developer and advocate of standardized tests and intelligence testing. With these measures of learning, best practices in the direct transfer from experts to novices could be standardized and scientifically evaluated. Instructionism is a useful label for these ideas. MOOC developers have been overwhelmingly instructionists

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

The Key Components of MOOCs

A

Learning Management Systems (LMS)
Storefronts
Autograders

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

essential purpose of a Learning Management Systems (LMS)

A

organize instructional content online:
1. provide a set of authoring tools that allow faculty with no programming experience to create course sites
2. feature convergence, where every innovative feature that a company develops is rapidly copied by competitors
3. online components of courses within an institution are usually extremely similar to one another
4. moves every student along a linear, instructor-guided pathway by default

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

storefront

A

a business operations innovation that allowed the public to sign up for a course

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Automated assessment/Autograders

A
  1. quantitative disciplines, including cs, physics and math (and early reading), have developed automated assessments when the steps to solve a problem are well defined
  2. computers cannot validly assess complex human performance. Under most conditions, computers cannot effectively evaluate unstructured text in essays or short- answer assignments
  3. Since teaching how to reason from evidence is one of the main purposes of higher education, and since the main way of demonstrating this reasoning is through essays and
    similar written performances, the inability to autograde this kind of work is a critical limitation for instructor- guided online learning
  4. MOOCs are better suited to credential learning in those domains where knowledge is more amenable to computational assessment
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Two problems with taking single courses through MOOC providers and earning a credential

A
  1. relatively few students who signed up for courses actually completed them
  2. the value of non-degree credentials remained ambiguous; bypassing the bureaucracy of admissions also meant giving up on the legitimacy associated with formal relationships with an academic institution
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

MOOC providers pivoted away from individual courses and toward more comprehensive programs

A

attracted primarily working professionals who were not applying to residential degree programs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

edX created a new program called the MicroMasters

A

allows students to earn an online credential by completing a series of MOOCs and sometimes passing a proctored exam.
Students can then apply those credentials as course credit toward an on-campus or online degree

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

MicroMasters appear to be an extension for the already educated

A

equality issue, educational gap

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

MOOC providers also make their courses available for companies

A

provide through their internal professional development offerings;
codevelop new programs for use within companies.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

what these MOOC-based degrees, MicroMasters, and corporate training programs have in common

A

they recognize the value of a formal, bureaucratic connection between a learner and a university or employer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

MOOC providers have come to look less like a disruptive force

A

more like a well- established player in higher education called online program managers (OPMs)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

people who succeed in this kind of instructor-guided, self-paced online learning are

A

typically already- educated, affluent learners with strong self-regulated learning skills;
MOOC created new opportunities for the already educated more than they have created new pathways into higher education;
they are most likely to effectively serve the already educated pursuing advanced postsecondary learning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Autograded courses allow people to proceed at their own time and pace

A

but the scale of the enterprise means that students need to press on alone or find their own sources of academic support

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Two very strong predictors of registering for and succeeding in a MOOC

A
  1. socioeconomic status (measures of access to social and financial capital)
  2. proficiency with self-regulated learning strategies
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

People develop self-regulated learning strategies through direct instruction and practice

A

often through a long apprenticeship in formal educational systems

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
a recent study comparing learners in the Arab world taking HarvardX and MITx courses to learners enrolled with the Jordanian MOOC provider Edraak shows that learners in Edraak courses have better gender balance, include more people with lower levels of education, and have higher completion rates
A few specific features of Edraak may explain this success: their courses have instructors from Arab universities, are targeted to regional needs, and support the right-to-left writing of the Arabic language
25
Terabytes of Data, Little New Insight
information is largely limited to descriptive evidence of MOOC learners, not how they learn
26
Rather than building lots of MOOC courses and never make changes, and hoping that data-driven insights appear downstream
a far more promising approach is to invest in online courses that are designed from the beginning not just for student learning but also for conducting research about learning; refine the instructional materials and assessments to continuously improve those outcomes
27
K–12 institutions have mostly avoided instructor-guided, large-scale learning environments
K–12 educators (and parents and school boards) recognize: 1. the inherently social nature of learning 2. the limits of young people’s self-regulated learning skills
28
ALGORITHM-GUIDED LEARNING AT SCALE
Adaptive Tutors and Computer-Assisted Instruction
29
algorithm-guided learning: adaptive tutors/computer- assisted instruction (CAI)
the next action in a sequence is determined by a student’s performance on a previous action rather than by a preset pathway defined by instructors
30
Adopting adaptive tutors would constitute a major initiative
requiring investing in hardware, scheduling computer lab or laptop cart times, selecting software, training teachers, communicating with parents and family, and tinkering with many other elements in the complex ecology of a school. (opportunity cost)
31
factors that play a role in determining the efficacy of the initiative of adaptive-tutoring technologies
technology availability; teachers’ willingness to adopt new practices; adaptive-tutoring technologies are well developed only in a few subject areas, including math and early reading
32
Will adaptive tutors help my students in my school? (4 sets of questions)
1. What is the pedagogical model? What are the fundamental principles of the underlying technology? 2. how similar technologies have been integrated into schools elsewhere 3. what the accumulated research evidence says about which kinds of schools and students are most likely to benefit 4. consider all this history and evidence in the light of one particular, idiosyncratic school: yours
33
Item Response Theory (IRT)
create a mathematical model of the relative difficulty of test item; paved the way for computers to automatically generate testing and learning sequences that could adapt to the performance of individual students
34
Eric Taylor, among teachers using CAI, the variance of learning gains from teacher to teacher was lower than the variance among teachers not using CAI
time on computers was a boon for students who had the weakest teachers, but a hinder for students who had a proficient instructor
35
Cognitive Tutor: Algebra
two days a week for individualized practice (blended mode); learning outcomes were better in schools where teachers most fully allowed students to proceed on practice problems at their own pace
36
ASSISTments
mostly a homework helper: students get immediate feedback on hw, and have the option to do some additional “skill- builders” that incorporate some adaptive elements of CAI. Most student work was probably on non-adaptive teacher-assigned homework problems, so the intervention probably wasn’t really testing adaptive learning environments. most of the gains were among low- achieving math learners; played a role in closing achievement gaps.
37
effects of both interventions were about the same (Cognitive Tutor: Algebra and ASSISTments)
suggests that all the complex machinery of the full CAI system may be unnecessary, and a lightweight online homework helper could perhaps be just as good as a complex adaptive tutor
38
early reading - adaptive tutors
On average, studies of adaptive tutors in early reading have not shown positive impacts on learning
39
math - adaptive tutors
math is the only domain where adaptive tutors have consistently shown some evidence of efficacy
40
The best way to understand CAI is as one possible tool among many for improving mathematics education
other options too: 1. human tutors to provide more support for the students struggling the most 2. professional development for teachers in rich mathematical discourse or a deeper understanding of fundamental math content 3. new software that facilitates new kinds of visualization in mathematics, like Geometer’s Sketchped or the Desmos graphing calculator
41
PEER-GUIDED LEARNING AT SCALE
Networked Learning Communities
42
The consensus about the great potential of personalized learning depended on a stark disagreement about what the term actually meant
1. personalized pace (advocates of adaptive tutors and blended learning) 2. personalized content and learning experience (students would be able to leverage online networks to explore their own interests)
43
twin visions of personalization— personalization as algorithmically optimizing a student’s pathway through established, traditional curriculum and personalization as students choosing topics for study and communities for participation
not only very different, in some sense irreconcilable
44
distinction in how these camps (twin visions of personalization) view the notion of “scale”
for most systems of instructor-guided and algorithm-guided learning at scale: scale is a hurdle to be overcome through technology (try to substitute human tutor) for peer-guided learning at scale: scale is a crucial resource for creating powerful learning experiences. Scale means knitting together a community of learners to teach and share
45
two types of MOOC
1. instructionist MOOCs (xMOOCs) 2. connectivist MOOCs (cMOOCs), peer-guided might coexist side by side in an online learning ecosystem
46
legitimate peripheral participation
a novice hangs out on the edge of a community of experts, looking for opportunities to move from the edge toward the middle
47
Situated learning and connectivism are pedagogical approaches that are attentive to the social and cultural dimensions of learning
they encourage designs that let people move from the periphery to the core of a learning experience or learning community
48
instructional technologists, George Siemens and Stephen Downes argues that knowledge exists in networks
the way to increase knowledge is to generate richer, denser networks. In this model, the best learning happens when learners connect with other people and resources that support ongoing inquiry the content of a course is a kind of trick designed to bring people together into conversation, and it is through this conversation— rather than through direct instruction—that the learning happens
49
In the connectivist vision for peer-guided learning at scale
learners need to develop a variety of online learning skills as a precursor to learning about particular topics or subjects (high entry barrier)
50
Scratch
a block-based programming language developed by Mitch Resnick, Natalie Rusk, and their team at the Lifelong Kindergarten lab at MIT
51
There is no right way to program in Scratch or right pathway to learning how to program
so the site generally stays away from the kinds of linear instruction provided by MOOCs or adaptive tutoring systems
52
Connected learning is interest-driven and peer-supported, but crucially,
it also provides opportunities for academic connections
53
Many of the core expectations of schools—that students produce their work independently, that all students complete a project in a similar amount of time, that all students study topics regardless of their interest level—conspire against
a pedagogy that seeks to empower students as leaders of their own creativity and learning
54
People with very different pedagogical proclivities pursued a similar vision of “deeper learning,”
a set of interrelated competencies that include traditional disciplinary knowledge as well as the skills of communication, collaboration, problem solving, and self-regulation
55
When students experience deeper learning
they develop mastery of deep content knowledge in a domain. They also experience a shift in identity where the learning activity is a part of who they are rather than something that they do
56
Traditional instructionist educators believe that content mastery is a necessary precursor to shifts in identity and opportunities for creativity, while social constructivists observe
that most motivation for learning comes from opportunities to be creative
57
formal educational systems are overwhelmingly organized around the mastery-first models
room for create more opportunity for creativity-first and identity-first learning
58
Part of the challenge of measuring peer-driven learning
in many of these learning environments, goals are determined by individual students and the networked community, not by teachers or evaluators the concerns are less about whether an entire class of learners is developing new capacities in one subject
59
Researchers in the Lifelong Kindergarten lab have largely studied the Scratch community
intensive qualitative research — thick descriptions of the lives and practices of individual Scratchers
60
a vigorous commitment to methodological pluralism
We need learning environments that shift whole distributions to the right, and we need learning environments that enable deep learning for a self-selected few
61
Formal education systems mandate that all students learn certain fundamentals
students in schools are expected to tackle many of their most consequential assessments alone so that their individual competency can be measured. These expectations create an inhospitable institutional climate for peer-guided learning environments to take root.
62
The most adaptive approach in the near-term is probably for creative educators to
find more spaces where peer-guided large- scale learning can be woven into the periphery of schools— in electives, extracurriculars, and untested subjects—so that learners can have some practice in navigating these new networks with a community of local peers and mentors to support them
63
The core question of learning games is one of transfer
do people flexibly deploy game insights back in the humdrum of everyday life
64
near transfer (contrast to far transfer)
situations that are only slightly novel
65
A recent meta-analysis examined studies of chess training, music training, and working-memory training, and found
little compelling evidence that any of these three practices improved people’s general cognitive performance
66
most skills are actually quite domain specific
'domain- independent', 'broadly useful' skills like critical thinking, collaboration, and communication are actually quite domain specific, primarily depends upon an encyclopedic knowledge of the domain
67
games that are more effective
play over multiple sessions, with more advanced and theoretically informed features
68
even experiments using games with simple mechanics
showed modestly better outcomes than control conditions without games
69
Instructor-Guided Learning Games
Chocolate-Covered Broccoli: pouring behaviorist chocolate over instructionist broccoli is often described as “gamification” these practices have a broad foothold in schools have the advantage of aiming to bridge a problem of near transfer
70
Algorithm-Guided Learning Games
Duolingo includes adaptive features that allow for personalized spaced repetition benefits of spaced repetition: can implement complex, personalized schemes of spaced repetition for each student limitations of autograding
71
Peer-Guided Learning Games
Vanished and Minecraft