Mobile Phones and Language Literacy in Rural Developing Regions (By Matthew Kam)
Upcoming SlideShare
Loading in...5
×
 

Mobile Phones and Language Literacy in Rural Developing Regions (By Matthew Kam)

on

  • 1,959 views

Presentation for the VII International Seminar of the UOC UNESCO Chair in e-Learning...

Presentation for the VII International Seminar of the UOC UNESCO Chair in e-Learning

Literacy levels in most poor countries remain shockingly low and formal education is making little progress. MILLEE improves literacy through language learning games on cellphones – the “Personal Computers of the developing world” – which are a perfect vehicle for new kinds of out-of-school language learning. The project focuses on developing scalable, localizable design principles and tools for language learning. The challenges are (i) to integrate sound learning principles, (ii) to provide concrete design patterns that integrate entertainment and learning, and (iii) to understand cultural and learning differences in children in developing regions.

Statistics

Views

Total Views
1,959
Views on SlideShare
1,959
Embed Views
0

Actions

Likes
1
Downloads
36
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

CC Attribution-NonCommercial-ShareAlike LicenseCC Attribution-NonCommercial-ShareAlike LicenseCC Attribution-NonCommercial-ShareAlike License

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment
  • Cellphones are not only low-power devices that can be used at places without reliable electricity, but also have the computational capability to run software, including educational software. The primary need for out-of-school learning arises because there are 101 million children worldwide who do not attend school. They spend most of their time working for their family in agricultural fields or households. Infact, cellphones promise to make education more accessible to children in out-of-school environments without disrupting them from not working for their family.
  • We wanted to find out whether the children would use the cellphones voluntarily The contribution of this paper is to report on how children in rural areas use cellphones in their daily lives over an extended period of time. Accordingly, our research goals were focused on: The opportunities for mobile learning in everyday rural settings The social contexts and challenges to mobile learning in naturalistic settings
  • The field research took place in two neighboring villages in north India. By rural standards, the economic baseline of one village was average while the other was lower than average. (Refer them to the map  make the dots) We chose to work with these 2 communities because we had a successful history of conducting field research. (Go through the points for economic baseline) ----------- Best: $6000-$9000 Good: $4000-$6000 Better, than average: $2500-$4000 Average: $800-$2500 Worse than average: $250-$800 Poor, very poor: $100-$250
  • For this phase, we recruited 18 children. The caste and gender breakdown of the participants reflected the local demographics i.e. 2 upper-caste girls, 7 lower-caste girls, 6 upper-caste boys, and 3 lower-caste boys, as shown in the figure. To track the changes in social relationships between participating children, we also mapped the existing social interactions between them. These relationships are indicated by the lines between participants in the graph. It seemed that children did not interact much across gender and caste boundaries, as you can see that the group at the lower-right corner has no interaction with the rest. Talking about the technology baseline, 14 out of the 18 households owned atleast 1 cellphone. 7 of these households had cellphones that were programmable. But even though the majority of the households owned a cellphone, we loaned out cellphones from our research group to each child, so that we do not have to ensure that our cellphone applications, work across a diverse range of cellphone models that kids have.
  • We constructed detailed accounts of a child’s daily life, in which we observed what the children did from the time they woke up to the time they went to bed. These accounts reflected the existing gender and caste inequalities. From these accounts, we identified 9 scenarios where a child could play mobile learning games. (pause) 2 of these scenarios are..
  • Lower caste boys are not expected to have the time to learn from cellphones in the field, since they would work as hired labors in the field. On the other hand, an upper-caste boy can be using a cellphone to learn when “working” in the fields. That is because unlike his lower-caste counterpart, they were only there to supervise. (*) Explain the photograph
  • Male and female siblings learn from cellphones games at home, after dinner-time and before bed-time In both upper- and lower-caste families, boys have more time to use the cellphone and learn using it, than their sisters because the girls are expected to help with the housework after dinner. But after the girls have completed the housework, they can join their brother to learn using the cellphones. (*) Explain the photograph
  • Our cellphone games focused on teaching English because it is considered a “world language”. And fluency in English can be equated to better socio-economic status, leading to better jobs and higher education. The curriculum for our cellphone games was designed by a local teacher who had 10 years of prior experience teaching children in India. The curriculum targeted a total of 180 word families, which was considered an appropriate target for 16 weeks. The first screen introduces the word – both in written and oral form – along with the image of the word being taught. It also explains the meaning of the word in native language, i.e. Hindi. The second screen asks the learner to identify the image, and pick the corresponding word. If he chooses the correct option, it is counted as a success
  • Since objects in traditional Indian games are everyday objects which are readily available in villages , traditional games do not involve objects such as money, armor and magic items that are found in Western games. Furthermore, since objects in traditional games are physical equipment, these game resources are fixed and are available in highly limited quantities . As such, while they can be eliminated , they are also non-renewable , i.e. new resources cannot be created in a game session. Similarly, it is not possible to create new units during gameplay, unlike digital games that make resource management a central goal. By implication, resources in traditional games cannot be invested to reap more future resources. None of the traditional games involve skills or statistics that enable a character to perform an action better than other characters. As such, character development has no central role in traditional games, unlike videogames (e.g. role-playing ones) in which it is desirable to raise a skill level or gain the ability to perform an action that a character could not do previously. In other words, in traditional games, state variables for player characters do not track skills because skill acquisition is non-applicable. Similarly, since player characters in traditional games do not have attributes that differentiate them significantly from one another, there are no boss monsters for players to defeat.
  • The previous screens focused on introducing the semantics of a vocabulary word to the learner. After that, the game tested the user on his short-term retention of the vocabulary word in a game context. The design of our cellphone games was adapted from the traditional village games that the kids enjoyed playing. For instance, the figure shows one of the games called Marakothi which was transformed into a digital version.
  • After analyzing our data, we had a few key findings. We realized that this rural community is heterogeneous in many different ways. One of these was that their electricity access was correlated with the income levels. At an overly simplified level, we could divide the households into three categories. That is, households that were at the mean income level, those above the mean level and those below it. The households above the mean income level, owned electricity generators at home which were used to charge all the electrical appliances, including the cellphones. The households at the mean income level charged their cellphones using the wall power outlets at their homes. However, even though 16 out of 18 households had such outlets at their home, the challenge was of irregular power supply. Moreover, participants did not leave their cellphones plugged into the wall outlets due to the fear of it being stolen. So, essentially households at the mean income only managed to charge their cellphones at times when electricity was present and they were also home. --- 2 rich 12 typical 4 poor
  • However, the social rituals around electricity differed dramatically for households below the mean income level The real challenge with this group was that their power outlets provided highly fluctuating voltages, which damaged 4 of our cellphone battery chargers. These 4 families were the poorest families in the community, and could not afford their own cellphone. Therefore, up until they enrolled in our study, they were unaware that their electrical outlets could not support cellphone chargers. So, these households relied on neighbors who had access to generators or stable power outlets for charging their cellphones. But they were hesitant to repeatedly ask for favors, because of the implied social obligation. Moreover, the electricity problem became worse when the children made excessive use of multimedia-rich applications, which eventually drained the batteries faster.
  • Despite these electricity challenges, it seemed that the participants still managed to keep their cellphones sufficiently charged for mobile learning applications. An average user spent 2 hours, 23 mins per-week playing mobile learning games. On the other hand, 2 participants – users 17 and 18 in the figure – who had access to electricity generators at their home, emerged as “power users”. With the cellphones charged more often, they were able to spend more time, with the mobile learning games. They also became sufficiently well-versed with the technology to troubleshoot technical problems on other’s cellphones, like adjusting sound volume.
  • According to our log data, the amount of cellphone usage varied a lot for the first 10 weeks. So, in order to analyze this properly, we waited for the novelty effect to worn off i.e. until the 10 th week. An average child covered 46 new words over 16 weeks of usage. On extrapolation, each child is expected to cover 150 new words in one year. The literature on second language acquisition states that a realistic target for children is 500 words per year, given good learning conditions. Therefore, learning 150 words seemed like a reasonable achievement given the existing electricity challenges, and gender attitudes.
  • Even though the promise of mobile learning is to extend learning to various social contexts outside school, our interviews showed that the cellphones were mostly used at home. Children reported that during summers since home was cooler than outdoors, they spent 75% of their time at home. However, there could be a possible bias in the data because it was collected in the summer months. In exit interviews, the participants reported a greater use of cellphones outside home.
  • Cellphone games constituted something that all children had in common. It provided them the impetus to interact with one another, and as a consequence, it appeared that existing relationships became stronger. At the same time, 10 participants who did not know each before the pilot, developed new relationships. A lot of them crossed geographic and caste boundaries. These interactions primarily developed in the process of seeking help from one another on cellphone games or exchanging digital content like audio clips, video clips, wallpapers and ringtones. More importantly, these relationships were not only restricted to interactions around technology but also transferred to real-world, non-gaming settings. For instance, there was one time when a lower-caste boy was accused of writing on the school desk by his teacher, but the upper-caste boy admitted his fault. This was a rare incident since upper-caste boys defending lower-caste boys is not so common in rural India.
  • Now I have a question for the audience. By the end of the study, cellphones given to 6 of 18 participants had swollen batteries. Can anyone guess why?
  • According to cellphone’s manual, the most plausible reason for battery swelling was exposure to extreme heat. Considering that most of the cellphones with swollen batteries belonged to girls, our best guess is that girls tried to hide the cellphones that we loaned to them near the stove in the kitchen. The fact is that even though we loaned the cellphones to the girls, the brothers were continually trying to monopolize the cellphones because of the implicit approval from the parents.
  • But on the other hand, there has been a growing number of initiatives that attempt to improve education in the developing world using educational games Games are not only fun and engaging, but can also
  • For instance, apart from also playing the mobile learning applications, they spent time listening to music clips, which they bought from nearby shops, as well as exchanged with other friends via bluetooth.
  • But on the other hand, there has been a growing number of initiatives that attempt to improve education in the developing world using educational games Games are not only fun and engaging, but can also
  • But on the other hand, there has been a growing number of initiatives that attempt to improve education in the developing world using educational games Games are not only fun and engaging, but can also
  • But on the other hand, there has been a growing number of initiatives that attempt to improve education in the developing world using educational games Games are not only fun and engaging, but can also

 Mobile Phones and Language Literacy in Rural Developing Regions (By Matthew Kam) Mobile Phones and Language Literacy in Rural Developing Regions (By Matthew Kam) Presentation Transcript

  • Mobile Phones and Language Literacy in Rural Developing Regions Matthew Kam Assistant Professor Carnegie Mellon University Human-Computer Interaction Institute October 7, 2010 VII International Seminar of the UOC UNESCO Chair in e-Learning
  • Multidisciplinary Collaborators at CMU
    • Local project manager – Indrani Vedula
    • School of Design – Jodi Forlizzi, Stephanie Meier
      • Visual design
    • Entertainment Technology Center – Hanika Khakhanis
      • Videogame design
    • Language Technologies Institute – Maxine Eskenazi
      • Evaluation of speech technologies for language learning
    • Dept. of Modern Languages – Pooja Reddy, Sue-mei Wu
      • Second Language Acquisition and Reading Science
  • International Collaborators Sesame Workshop Chinese Academy of Sciences ASSET India Foundation Byrraju Foundation Dhirubhai Ambani Institute of ICT IIIT Hyderabad Naandi Foundation Suraksha University of Nairobi
    • Current and previous funders:
    • MacArthur Foundation
    • Microsoft
    • National Science Foundation
    • Nokia
    • Qualcomm
    • Verizon
  • Building Local Capacity [in ACM interactions 2008]
    • Worked with 20+ undergrad researchers from India
      • Alumni are pursuing graduate school in North America
    • Winter school with 130 undergrads across India, hosted by IIIT Hyderabad (December 2009)
    • Develop local pool of talented manpower
      • Human-computer interaction
      • Learning sciences
      • Low-end cellphones programming
  • Needs and Problem Statement
    • Fluency in “power language” e.g. English
    • Public schools in developing regions (e.g. India) are not succeeding
    • 101 million primary school-age children do not attend school
      • 36 million in South-Asia
      • 39 million in Sub-Saharan Africa
    • Cellphones can make education more accessible through out-of-school environments
      • User can learn anytime, anywhere without disrupting work
    • Game-like exercises for enjoyable learning experience
    Solution Overview
  • Project Timeline
    • 10 rounds of fieldwork, >12 months total in India
    • Human-centered design process with 100 children
    2004 2005 2006 2007 2008 Needs assessment (village + slums) ‏ Exploratory study (slums) ‏ Exploratory study (village) ‏ Feasibility study (slums) ‏ Feasibility study (village) ‏ More iterations + testing Classroom study 2009 Out-of-school study Controlled study 2010
  • Selected Media Appearances
    • Cell Phone: The Ring Heard Around the World. In Canadian Broadcasting Corporation television documentary , aired on April 3 and June 5, 2008. http://www.cbc.ca/doczone/cellphones/india.html
    • India’s Cell Phone Tutors. In ABC News , aired on June 16, 2009. http://abcnews.go.com/video/playerIndex?id=7854956
    • Angrezi, the Phoney Way. In Times of India , December 5, 2009.
    • In Rural India, Learning English via Cellphone. In The Chronicle of Higher Education , October 21, 2009.
  • Classroom Study [in IEEE/ACM ICTD 2009]
    • Deployment throughout Spring 2008
      • Three times per week
      • After-school program at private village school
    • Quantitative study with 27 participants (grades 2-9)
    • Demonstrated significant post-test improvements on spelling skills ( p = 0.007, σ = 3.3)
    • Learning gains correlated with grade levels ( r = 0.61)
  • Out-of-School Pilot Study [in ACM CHI 2010 – Best Paper Honorable Mention]
    • First study on rural children’s voluntary use of cellphones in their daily lives over extended time
      • Opportunities for mobile learning in everyday rural settings?
      • Social contexts behind such naturalistic settings, including adoption challenges?
  • Study Context
    • Mango growing district in rural India
    • Economic baseline
      • $500 to $2,500 annual household income
      • Upper-caste households own land on which lower-castes work as laborers
      • All 20 households we visited owned at least one cellphone
    Source: NCAER, 2001
  • Participants
    • 26-week pilot with 18 children
      • 9 boys, 9 girls
      • 10 to 14 years old (mean = 12 years)
    • Loaned out cellphones
      • 14 out of 18 participating households had at least 1 cellphone
      • 7 out of 18 households had programmable cellphones
  • Scenarios for Mobile Learning
    • Participant-observations with 20 rural households
    • Identified 9 scenarios by different times of day
      • Constructed after analyzing “day in the life” accounts
    • Scenarios specific to different caste and gender
    5:30am 7am 2pm 5pm 7:30pm 9pm Morning Afternoon Evening Night Early Morning
  • Scenarios: Learning in the Fields
    • An upper-caste boy learns from a cellphone-based game when “working” in the fields
    5:30am 7am 2pm 5pm 7:30pm 9pm Morning Afternoon Evening Night Early Morning With cellphone
  • Scenarios: Learning at Home
    • Male and female siblings learn from cellphones games at home between dinner and bed-time
    5:30am 7am 2pm 5pm 7:30pm 9pm Morning Afternoon Evening Night Early Morning With cellphone
    • Curriculum developed by local teacher
      • Targeted 180 word families for 16 weeks
    • Vocabulary teaching and testing
    Curriculum Design Explanation of “Father” in Hindi
  • Task-Based Language Teaching
    • Instructional sequence around tasks (Cameron 2001, Nunan 2004, Prabhu 1987) ‏
      • 1. Schema-building exercises that introduce vocabulary, linguistic forms and context for the task
      • 2. Communicative exercises to provide controlled practice
      • 3. Listening to how above linguistic units are used in authentic settings
      • 4. Language development exercises for above units
      • 5. Freer practice in groups, e.g. information gap activities
      • 6. Pedagogic task proper
  • Best Practices in 2 nd Language Teaching [in ACM CHI 2007]
    • Avoid reinventing the wheel
    • Reviewed sample of >35 applications
    • Sample has a balance b/w listening, reading, speaking and writing skills
    • Distilled >50 design patterns (Alexander 1977)
  • Traditional Village Games [in ACM CHI 2009 – Best Paper Honorable Mention]
    • How are traditional Indian village games different from existing Western videogames?
    • Compared game elements against 296 game design patterns documented in Bjork and Holopainen 2005
  • Analysis: Differences in Games
    • Identified 37 non-trivial differences
      • Difficulty based on sub-goals
      • Resource management
      • Skill acquisition
      • Score keeping
      • Rituals associated with space
      • Inter-team interactions
  • Prototypes: Videogame Design
    • Designed videogames based on traditional village games (Kam et al. 2009)
    Playground Digital Version
  • Findings: Electricity Access
      • Recharged cellphones from self-owned generators (2 households)
    • Power outlets in homes (16 out of 18 children)
      • But irregular electricity
    • Despite irregular electricity, did not leave phones plugged into wall outlets due to security concerns
      • Recharged cellphones only when electricity was present and participant was home
  • More Social Rituals Around Electricity
    • Highly fluctuating voltages (4 households)
      • Damage to battery chargers
    • Recharged cellphones from neighbors who had generators or stable power outlets
      • Hesitant to take full advantage of neighbors’ generosity
      • Recharged cellphones at workplaces
    • Excessive use of multimedia-rich applications drained batteries quicker
  • Duration of Mobile Learning
    • Despite electricity challenges, participants were able to keep cellphones charged for mobile learning
      • 2 hours, 23 min per week (on average)
    Electricity generator at home Reliable power outlets at home Unreliable power outlets at home
  • Learning Benefits
    • Average participant covered 46 new words over 16 weeks of unsupervised usage of cellphones
      • At this rate, each participant is expected to learn 150 new words in a calendar year
      • Benchmark is 500 words, given good learning conditions
  • Locations of Cellphone Use
    • Unexpectedly, most learning via cellphones occurred at home (vs. outdoors)
      • Home is cooler than outdoors due to summer heat
      • Participants self-reported taking cellphones outdoor prior to summer
      • Concern about loss of phone
    * % of total time spent at a particular location
  • Community of Gamers
    • Strengthened existing ties (13 pair-wise relationships)
    • Games facilitated new ties across gender, caste and village boundaries (10 pair-wise relationships)
    • New social relationships transferred to real-world, non-gaming settings
  • Other Challenges
    • 6 swollen batteries
    • Why did this happen?
  • Other Challenges – Gender Equity
    • 6 swollen batteries
    • Why did this happen?
      • Extreme heat from kitchen stoves
      • The 6 participants (5 of them were girls) tried to hide phones from brothers
  • Scaling Up in India and Beyond
    • Controlled experiment with 800 low-income children in 40 locations (w/ Nokia grant + 450 cellphones donation)
      • Official 5th-grade English curriculum in India
      • Benchmark against standardized English test in India
    • Partnership development with
      • Cellphone manufacturers
      • Wireless carriers
      • Third-party content developers
      • Education service providers
  • Bridge Curriculum
    • Prerequisites for advanced literacy skills, based on Chall’s stages of reading development (1983):
      • Phonological + orthographic awareness
      • Oral vocabulary knowledge
      • Phonetic decoding
      • Word identification (including fluency)
      • Spelling
      • Lexical inferencing
      • Morphological awareness
      • Sentence-level reading comprehension
    100 lesson plans for entire academic year
  • Adoption Ecology
    • “ One size fits all” approach does not scale
  • PACE Framework [in ACM CHI 2007]
    • Four components of PACE framework
      • P attern – best practices for learning and engagement
      • A ctivity – design of learner interaction
      • C urriculum – targeted syllabus + audio-visual content
      • E xercise – basic unit realizing Activity + Curriculum
    • Facilitates division of labor and local adaptation
  • Acknowledgements
    • Carnegie Mellon University (USA) – Rafae Aziz, Ryan Baker, Rachita Chandra, Manoj Dayaram, Maxine Eskenazi, Ayan Kishore, Alex Kowalski, Anuj Kumar, Derek Lomas, Gino Mancuso, Andrew Ngan, Pooja Reddy, Daniel Rhim, Geeta Shroff, Kyle Sondrock, Raja Sooriamurthi, Le Wei, Sue-mei Wu
    • Chinese Academy of Sciences (China) – Lv Fei, Tian Feng , Ben Rachbach
    • DAIICT, IIT, IIIT, NSIT (India) – Aishvarya Agarwal, Lalit Agarwal, Shilpan Bhagat, Malav Bhavsar, Anshul Chaurasia, Akash Gangil, Denny George, Aakriti Gupta, Chetan Gupta, Mayank Gupta, Amol Jain, Ashwin Jain, Rachit Jain, Siddharth Kothari, Kaustav Kundu, Siddhartha Lal, Manish Lohani, Mohit Maheshwari, Nikhil Marathe, Akhil Mathur, Kishan Patel, Vipul Raheja, Dhruv Shah, Yash Soni, Surendra Survi, Satyajit Swain, Vivek Tripathi, Pallav Vyas
    • University of Berkeley (USA) – Ruth Alexander, Lauren Bailey, Eric Brewer, John Canny, Deepti Chittamuru, Jane Chiu, Varun Devanathan, Asya Grigorieva, Dimas Guardado, Christopher Hom, Glynda Hull, Anjali Koppal, Maksim Lirov, Aaron McKee, Gary Miguel, David Nguyen, Anand Raghavan, Divya Ramachandran, Priyanka Reddy, Vijay Rudraraju, Monish Subherwal, Simon Tan, Anuj Tewari, Jingtao Wang
    • University of Nairobi (Kenya) – Tonny Omwansa
    • Pilot team (India) – Mehnaaz Abidi, Shabnam Aggarwal, Aman Anand, Siddharth Bhagwani, Jatin Chaudhary, Sonal Gupta, Shirley Jain, Alok Prakash, Neelima Purwar, Rolly Seth, Gautam Singh, Kartikey Singh, Kavish Sinha, Indrani Vedula
  • Contact Info
    • Matthew Kam
    • Assistant Professor
    • Carnegie Mellon University
    • Human-Computer Interaction Institute
    • Email: [email_address]
    • Homepage: http://www.cs.cmu.edu/~mattkam
  • Case for E-Learning Games
    • Can incorporate good learning principles (Gee 2003)
    • Demonstrated learning benefits with urban slums children (Banerjee et al. 2005)
      • 2 years, >10, 000 children
      • Math learning computer games twice per week
      • Significant gains on math test scores
    • … and rural children (Kam et al. 2009, Azim Premji Foundation 2004)
      • Significant improvements on spelling skills
  • Everyday Use of Multimedia
    • Listened to music clips
      • Bought at point-of-sales outlets in town
      • Shared with friends via Bluetooth
    • Took videos and photographs of important events
  • Opportunities: Girl Empowerment
    • Promote lifelong learning (e.g. employability skills) after child marriage
    • Enhance women empowerment workshops
    • Educate mothers not to be “stopping channels”
  • Opportunities: Bridge Programs
    • Increase reach of learning for nomadic communities, out-of-school children, etc.
    • Account for community norms
    • Address cognitive effects of schooling (and literacy)
  • Opportunities: Micro Data Collection
    • Accountability monitoring
    • Adaptive learning and artificial intelligence tutors
    • Improve learning theories (e.g. individual differences)