Computer skills development by children using
'hole in the wall' facilities in rural India

Earlier work often referred to as the "hole in the wall" experiments has shown that groups of children can learn to use public computers on their own. This paper presents the method and results of an experiment conducted to investigate whether such unsupervised group learning in shared public spaces can improve children's performance in school examinations. The experiment was conducted with "hole in the wall" (minimally invasive education, or MIE) kiosks in the rural Sindhudurg District of Maharashtra State, India. 103 children of the Grade 8 level, across 3 villages, were administered the curricular examination for 'Computer Science' for that grade.

The results show that children who had learned at MIE kiosks were able to complete this curricular examination without being taught the subject. They scored only marginally lower than children who had been taught the 'Computers' curriculum in school throughout the school year. The results of this study throw new light on pedagogy for bridging the digital divide. It poses the question that similar learning may well be observed in whole or part in other subjects of the school curriculum.

Introduction

The experiments were first conducted in Kalkaji, a suburb of New Delhi, India. A computer was connected to the Internet and embedded into a brick wall near a slum. It was reported that most of the slum children were able to use the computer to browse, play games, create documents and paint pictures within a few days. The results have, since then, been reported in detail, elsewhere (see, for example, Frontline World 2002, Education Guardian 2000, Businessweek Online 2000, Mitra 2000, Mitra 2003 and Wullenweber 2001).

The experiments were described by Mitra and Rana as "minimally invasive", a term borrowed from surgery (Mitra and Rana 2001, Mitra 2003), and indicative of minimal human intervention. Mitra further explained the term in an interview to Businessweek Online.

I'm saying that, in situations where we cannot intervene very frequently, you can multiply the effectiveness of 10 teachers by 100 - or 1,000 - fold if you give children access to the Internet.... This is a system of education where you assume that children know how to put two and two together on their own. So you stand aside and intervene only if you see them going in a direction that might lead into a blind alley. That's just so that you don't waste time... (Mitra, Businessweek Online 2000)

A design for kiosks meant for use by children in outdoor environments has been developed and tested in various experiments, and is considered robust.

In what follows, we are testing the following hypotheses:

If given appropriate access, connectivity and content, groups of children can learn to operate and use computers and the Internet to achieve a specified set of the objectives of primary education, with none or minimal intervention from adults.

Background for this experiment

Three kiosks were connected to the Internet via VSAT in June 2004, when the final version of this paper was written. Access to Internet content is also provided in an offline form. The offline content includes educational games and freeware from the Internet, mostly in English.

The kiosks in Sindhudurg are fitted with TOBU mice and keyboards as input devices. The design for these has evolved over the years since the first 'hole in the wall' in 1999. At the time, a touch pad was the input device embedded in the wall next to the monitor. Between 1999-2000 these off the shelf touch pads were replaced with custom made, joystick style mice. Although this did significantly increase the mean time between failures, the mechanical failures in these were not entirely eliminated. This led to the development of a six-button mouse, deployed in many kiosks. However, in humid and corrosive air, such as near tropical oceans, the switches in these mice developed faults. Finally, the development of the capacitative TOBU mouse in 2003 by the laboratories of NIIT's Center for Research in Cognitive Systems solved the problem of an outdoor pointing device. This mouse is entirely solid state and has no moving parts. Eighty-eight of these have been in operation for over a year in all parts of India.

Figure 1: Photo of the TOBU Mouse

Also, the first experiment of 1999 did not provide for keyboard access, as architectural designs for protecting standard keyboards in public access environments were not developed at the time. Since then NIIT has developed a sturdy cowl to protect keyboards with enough space for a child's hands to go through and access. The touch button mice are now embedded within the cowl keyboard structure. Over the 4-year period from 1999-2003, research in engineering design has made the present input devices rugged and tamper proof.

Each machine contains a remote monitoring software program to track usage of various software applications in the machines, as well as environmental conditions.

The kiosks are placed in playgrounds or close to schools (Figures 2 and 3). Rural children between the ages of 8-14 are presently using these kiosks, many with no previous exposure to computers. The local language is Marathi. This is also the language of instruction in the village schools. English is learned as a second language from Grade 1.

Figure 2: Photo of an MIE kiosk, Sindhudurg

"Computers" is offered as an optional subject in some village schools at the Grade 8 level along with leatherwork, sewing and carpentry. Only a few are selected into the computer class at school, due to limited capacity. Selection depends on merit and past academic performance.

In June 2002 a local computer teacher reported a saving of 10% of his teaching time in the computers class because his students had been exploring at the kiosk (Attar, June 2002). This gave us reason to conjecture that students who were not part of the school computer class may well be able to attempt some part of the curricular examination without being taught. To probe this further, in November 2002 we conducted a short study, to see how children who have learned at MIE kiosks fare on the Grade 8 examination for the Computer Science syllabus, without being taught the syllabus.

Figure 3: Photo of children at the kiosk

Six children aged thirteen years (three boys and three girls) were tested on a 2 hour 45 minute curricular examination in computer science. The exam was worth a total of 100 marks, split between 'theory' (60 marks) and 'practicals' (40 marks). The minimum score required for passing the examination was 35%. All six students passed the practical exam with very good scores. Two students were close to passing the theory examination, while four did not. The results were intriguing.

We decided to examine larger numbers of children through an experimental study of MIE children's performance on two aspects of the Grade 8 'Computers' exam:

1. The 'practical', or hands on examination, given no instructional inputs.
2. The theory examination, if the learning material, in the form of the prescribed textbook, was given to them.

Sub hypothesis

Groups of children with appropriate public access to the computers and the Internet can meet the minimum levels required by the Maharashtra State Education Board computers syllabus for Grade 8, with none or minimal intervention from adults

Population

School going rural children aged 12-13 years, of Maharashtra State, Sindhudurg District, with access to MIE kiosks.

Profiles of groups

Group	Villages	Number Male Female	Age	MIE kiosk learning	Classroom computer teaching	Hours of computer handling/ exposure	English language teaching in school	Medium of instruction
Experimental Group	Kalse + Shirgao	N=30 12F+18M	12-13	Yes	No	At least 20 hours	Yes	Marathi
Control Group A	Kuvle	N=31 16F+15M	12-13	No	No	0 hours	Yes	Marathi
Control Group B	Shirgao	N=42 38F+4M	12-13	Yes	Yes	At least 20 hours	Yes	Marathi

Method

• Experimental Group - MIE students (N=30) drawn from villages Shirgao and Kalse
• Control Group A - computer illiterate students (N=31) drawn from village Kuvle
• Control Group B - classroom computers students (N=42) drawn from village Shirgao

The first phase tested the students on the practical examination. The 70-minute practical examination was conducted on all 3 groups at the computer lab facilities of Shirgao High School and Kalse High School. The student to machine ratio was 1:1. Each student was handed a question set of exercises.

In the second phase, the theory examination was administered to all groups. Prior to the examination, each of the students in the experimental group was handed the Grade 8 'Computers' textbook and was told that the exam would follow in a month. Control Group A was not handed the book, to check the results given no inputs whatsoever. The book is a standard text used for curricular teaching for Control Group B.

All 3 groups were administered the 90-minute written theory examination in their respective schools (Table 1). The examination was conducted in Marathi, the language of instruction. This was followed by an oral examination of 5 minutes per student, conducted in Marathi.

A schoolteacher from Shirgao who teaches computers set the examination papers, monitored the examination as well as graded all students. All of these activities were conducted as per Maharashtra State Education Board norms. The computer teacher is a Maharashtra State Education Board examiner as well as moderator. The author was present as observer of evaluation methods.

Type of exam	Content	Evaluation criteria	Comments
PRACTICAL Duration 40 minutes 20 marks	Students were given 2 of a bank of 4 exercises. Each to be done in 20 minutes. Exercise 1: Using the mouse Select the My Computer icon Open the My Computer icon Close the My Computer opened icon Right click on desktop and open properties Drag the My Computer icon Exercise 2: Shut down the computer Click on 'Start' Choose 'Shutdown' option From the Shutdown window choose the 'Shutdown' option Click on the 'OK' button in the Shutdown window Shut off power supply Exercise 3: Opening Date time and changing time, year and month Open clock Choose month of May Choose year 2007 Choose time 9.10 Apply changes Exercise 4: Handling programs: Open the Program called Paint Minimise it Maximise it Draw a pretty picture Close the program	Each exercise worth 10 marks. 2 marks for each step executed.	1 to 1 student machine ratio Students have not used a mouse before.
WORKBOOK PROCESS NOTE Duration 30 minutes 20 marks	Students were given 1 of a bank of 2 exercises to do as per instructions given to them. They are supposed to follow some of the steps, execute some others by themselves and write down their observations of the process/method on paper. The 2 exercises: Using numeric keypad with cursor control keys and numeric keys Using the Start Menu	8 marks for the written note. 12 marks based on teacher's observation of the method employed by child, help taken, and task completion.
THEORY Duration 90 minutes 50 marks	Fill In the Blanks 6 marks; Match the following 4 marks; True or false 6 marks; Describe process 4 marks; Describe the differences 4 marks; Name the following based on the given description 6 marks; Short notes 6 marks; Describe method 4 marks; Short answers 10 marks	Model answer paper
ORAL Duration 5 mins 10 marks

Observations during the practical exam

1. None in the Experimental Group or Control Group A had any exposure to the computer syllabus or teaching. Hence, computer vocabulary was largely unfamiliar to them. They were not familiar with questions like "open the program WordPad". However, the experimental group interpreted the questions in the context of their own experience and responded.

2. Both Experimental Group and Control Group A were unfamiliar with the examination pattern. Hence the process of the examination was novel to them. The act of following one instruction at a time in a given set of questions was a pattern they were not prepared for, but had to follow. While this confounded the Control Group A, the Experimental Group grappled well with it.

3. Both Experimental and Control Group A had not used a conventional mouse prior to the examination. At MIE kiosks children use a TOBU mouse device with six buttons (Figure 1). Hence the Experimental Group, although not familiar with using the conventional mouse, was conceptually familiar with mouse operations, and was quickly able to adapt to using the conventional mouse for clicking, double clicking and dragging.

4. In the 'Process Note' examination, students were given some steps to do and had to write down the process and its results. It was seen that the Experimental Group needed to be reminded to write down their observations of what they were doing. They would often execute all the steps and forget about writing them down. They were untrained in this method, but did well.

5. Children in the Experimental Group sometimes understood a question as a task and accomplished the task in alternate ways. One child in the experimental group found an alternative way of opening Microsoft Word. However the instruction for the Process Note was to 'open' through "Run". These alternate methods were not given marks in the practical examination where the emphasis was on following instructions.

Results overview

• 87% of the Experimental Group passed the practical examination.
• 67% of the Experimental Group scored over 60% (classified as 1st Class) in the practical examination
• The highest overall score (theory+practical) in the Experimental Group was 62%, the same in the Control Group B was 69% and in the control group A was 14%
• 88% of the Experimental Group passed the complete examination
• All of the Control Group B passed the complete examination
• None of the Control Group A passed the examination
• The average score for the Experimental Group was 47%

Comparing group results

Figure 4: Cumulative group scores

Figure 4 shows the average group scores of the Experimental Group at 47%, Control Group B 56%, and Control Group A 9%. The higher overall average for Control Group B was due to higher scores in the theory examination. Control Group A attained 9% despite having no exposure to computers prior to the examination. There could be several reasons for this:

• Learning gained from the first 40 minutes of being exposed to computers during the practical examination
• Marks gained from randomly chosen answers to multiple choice questions
• Some rudimentary prior knowledge gained from hearsay or media

Figure 5: Practical exam group averages

Figure 6 compares the average group scores on the theory examination. The Control Group B leads at 33 marks out of 60, the Experimental Group is at 25 and the Control Group A is at 7.

Figure 6: Theory exam group averages

Discussion and conclusions

Objectives of primary education

It is also quite likely that this mode of unsupervised group learning through the computer is impacting subject knowledge for other subjects. There is considerable learning material as well as games stored on these machines. Three of the MIE locations mentioned in this paper were connected to the Internet through VSAT in June 2004. The impact of this is an area for future research.

Curriculum delivery

It is postulated that curriculum can be divided into three bands, in order for learners to acquire the necessary competence (Mitra 2003):

1. A band that needs a human teacher who is conversant with the subject matter and teaching methodology
2. A band that needs an assistant who is somewhat more knowledgeable than the learner
3. A band that needs resources and a peer group alone

Serving the economically and academically disadvantaged

On kiosk learning environment design

Next steps

Acknowledgements

• I wish to thank Sugata Mitra for his guidance and comments
• Financial assistance from NIIT Limited, the Government of Delhi, the ICICI bank and the International Finance Corporation is gratefully acknowledged.
• I wish to thank schoolteacher Shamsuddin Attar of Shirgao High School, Sindhudurg, India for conducting the examination and grading the papers.
• This experiment would not have been possible without the support of the school principals and the use of school premises during the course of the examinations in Shirgao, Kalse and Kuvle village.
• I wish to thank my colleagues at the Center for Research In Cognitive Systems - Sanjay Gupta, Ravi Bisht and Ritu Dangwal - for their cooperation and support.

References

Bytesforall (2002). Village kinds [kids] find computers a useful, simple toy. http://mail.sarai.net/pipermail/bytesforall/2002-July/000111.html [verified 10 Aug 2004]

EducationGuardian.co.uk (2002). Slum children get free Internet access. Education Guardian, 17 Oct 2000. [verified 10 Aug 2004] http://education.guardian.co.uk/Distribution/Redirect_Artifact/0,4678,0-383459,00.html

Frontline/World (2002). Kids-eye view: Looking through the hole in the wall. FRONTLINE/World, October. [verified 10 Aug 2004] http://www.pbs.org/frontlineworld/stories/india/kids.html

Mitra, S., Tooley, J., Inamdar, P. and Dixon, P. (2003). Improving English pronunciation - an automated instructional approach. Information Technologies and International Development, 1(1), 75-84, MIT Press.

Mitra, S. and Rana, V. (2001). Children and the Internet: Experiments with minimally invasive education in India. The British Journal of Educational Technology, 32(2), 221-232.

Mitra, S. (2000). Minimally invasive education for mass computer literacy. Presented at the CRIDALA 2000 conference in Hong Kong, 21-25 June.

Mitra, S. (2003). Minimally Invasive Education: A progress report on the "Hole-in-the-wall" experiments. British Journal of Educational Technology, 34(3), 367-371.

Padmakar, P. and Porter, H. (2001). The hole in the wall machine. Time Magazine, Asian edition, 3 September, p.16.

Sweatshops and Butterflies (2001). India: Hole-in-the-Wall. [verified 10 Aug 2004] http://www.greenstar.org/butterflies/Hole-in-the-Wall.htm

The "Hole-In-The-Wall" (2003). [verified 10 Aug 2004] http://www.niitholeinthewall.com/

Wullenweber, Walter (2001). Das loch in der wand. Stern Magazine, No. 42, 11 October, pp 97-102.

Author: Parimala Inamdar Center for Research in Cognitive Systems, NIIT Ltd Corduroy Building, Aditya Textile Compound Safed Pool, Andheri East, Mumbai 400072, India Email: parimalai@niit.com Website: http://www.niitholeinthewall.com/ Please cite as: Inamdar, P. (2004). Computer skills development by children using 'hole in the wall' facilities in rural India. Australasian Journal of Educational Technology, 20(3), 337-350. http://www.ascilite.org.au/ajet/ajet20/inamdar.html