Maths Through SMS
11:01 PM
Mustafa
,
In a developing country like Pakistan, it would be a great service if the masses could be educated without the binding of a formal institute. In 2004, Pakistan's literacy rate was recorded at 46.6% with male literacy rate at 60.6% while female literacy at a mere 31.5%. This can be attributed to acute poverty, lack of education facilities in far flung areas and a lack of skilled teachers.
Our Proposed Solution
Education in Pakistan can be divided into five levels
- Primary, grade one through five
- Middle, grade six through eight
- Matriculation, grades nine and ten
- Intermediate, grades eleven and twelve
- University programs leading to graduate and advanced degree
In Pakistan, a number of people abandon their formal education at some point in their lives, mostly due to economic factors. They are still eager to learn and educate themselves if given a chance, however, since they are the breadwinners of their families, it makes it all the more difficult for them to spare time for such an effort. It is the need of the hour that innovative and flexible solutions are brought in to help educate the common man on his terms.
Learning through the cell phone is an answer to the problem. Mobile phone is the most prevalent, readily available and affordable technology
To judge the effectiveness of the system, an objective was set to improve the mathematical skills in particular and learning skills in general of a control group. We selected a control group in which some of the members were educated upto the Secondary School level, while others were Higher Secondary School Certificate holders. The idea was simple and exciting, an SMS Based, Mathematics Centered, Quiz Game.
The Solution (SMS Based Game)
>SmsMathQuiz is an SMS based game that participants can play in groups. Making this game group based has a twofold effect. Research has shown that people tend to learn better when they are competing amongst themselves. Group dynamics in players introduces an element of competition and hence enhances learning. Secondly, being a group based game it has the same effect as the world is experiencing with social games. Experiments have revealed that the game is indeed effective and interesting to play.
Following is a step by step description of the game.
- A group of people send a request via SMS to start a 10-15 question math quiz.
- Every player is sent the same question through SMS. He/she replies via SMS.
- The person who answers first and correctly is announced after every question through an SMS message to all the participants.
- Every participant moves on to the next question at the same time.
- At the end of the game, the highest scorer is announced via SMS to all the participants.
It has been observed during the development and experimentation of this game that giving a lecture to the participants before a game could be of great help in polishing their math skills.
Why Math
In the non-English speaking world, language is one of the barriers in effective learning. This was the reason we chose math over any other subject. Teaching math, as opposed to teaching something else like the English language, requires minimal knowledge of the language and the questions can be designed in such a way as to minimize the language barrier.
The
We decided to experiment with people who have at least primary school education and have some fundamental mathematical skills. So, it was decided to contact the peon staff of our university (Lahore University of Management Sciences) because most of them were right within the range that we were considering for these experiments.
Experiments
Three experiments were conducted keeping different variables in mind. All questions were multiple choices. Players were required to choose the option they consider to be the best and send it via SMS to the server.
Experiment 1
Interested individuals from the administrative staff of LUMS were gathered under one roof. They were given an introductory presentation on how to join a game via SMS and how to answer a question in the specific format. At that point in time, simple multiplication/addition questions were considered for the game. Participants were interviewed to get feedback.
Observations from Experiment 1
- Competitive nature of the quiz was thrilling for the audience, everyone tried to answer first.
- Players were committing errors when they were replying in a specific format. This happened mostly on two occasions. Firstly, when they were trying to join the game and secondly, when they were answering a question.
- Since questions were only mathematical like 543 + 34, there was no problem faced in understanding the questions.
- Difficulty level of the questions was very low and players were quite confident that they could face challenging problems.
Experiment 2
In response to the observations of experiment 1, it was decided that with each question, a reply form will be sent as well. Players were supposed to fill the form with their desired option (a, b, c or d). This approach successfully resolved the problem of sending replies in the wrong format.
The second issue identified in the experiment 1 was the nature of the questions. In experiment 2, it was decided to test the players on slightly difficult mathematical concepts like operator precedence (10 * 10 + 2), and single variable algebraic equations (4 * X = 32, X =?).
Observations in Experiment 2
- Players were a little rusty in their concepts when it came to slightly advanced mathematical concepts like single variable equations and operator precedence.
- It was decided that the word problems should be introduced to get a hint of how players would respond to this new format of questions.
Experiment 3
In experiment 2, it was noticed that players were having trouble grasping slightly involved mathematical concepts. To resolve this problem, brief tutorials were given at the start of experiment 3. These tutorials covered concepts required to solve the questions. Precisely speaking, tutorials were given on arithmetic series and basic geometry. When experiment started, it was observed that the tutorial in the start was a good idea. Results after the experiment 3 were really encouraging.
In this experiment, word problems were given in plain English and Roman Urdu (Urdu Written in english Alphabets). It was observed that, for the players, some questions were understandable in plain English while some made more sense in Roman Urdu. For example, the players easily understood the word “Rectangle” but they were having trouble with understanding the Roman Urdu translation for the word “Rectangle”, that is “Mustateel”.
Observations in Experiment 3
Thirteen areas of interest were chalked out to judge every player. The following table shows the score a player received in each area.
- Out of 10 how many questions were understandable in English?
- Out of 10 how many questions were understandable in Urdu?
- Did a question in Urdu prove helpful in understanding the question in English?
- Did a question in English prove helpful in understanding the question in Urdu?
- Average time taken to solve a problem
- Questions answered
- Questions Correct
- Answers counted
- Interest to have questions counted on question bases
- Number of failed joining attempts (people attempting to join a group) in the system on first use
- Number of failed joining attempts in the system on second use
- Number of incorrectly formatted answers if Questions are not “Fill in the blank kind”
- Number of incorrectly formatted answers if Questions are like “Fill in the blank format”
Abbas | Zulfiqar | Zahid | Zulqurnain | Ejaz | Majid | Mean | Variance | |
1 | 9 | 8 | 8 | 9 | 9 | 8 | 8.5 | 0.3 |
2 | 9 | 9 | 8 | 9 | 9 | 8 | 8.66 | 0.266 |
3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
4 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
5 | 3 | 5 | 6 | 4 | 4 | 4 | 4.33 | 1.06 |
6 | 9 | 9 | 9 | 9 | 9 | 9 | 9 | 0 |
7 | 8 | 5 | 6 | 6 | 6 | 6 | 6.16 | 0.966 |
8 | 6 | 0 | 0 | 1 | 2 | 1 | 1.66 | 5.066 |
9 | 1 | 0 | 0 | 1 | 0 | 0 | 0.33 | 0.266 |
10 | 3 | 4 | 2 | 0 | 1 | 1 | 1.8 | 2.16 |
11 | 2 | 3 | 1 | 0 | 0 | 0 | 1 | 1.6 |
12 | 3 | 2 | 1 | 0 | 1 | 2 | 1.5 | 1.1 |
13 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
A visual representation of the above data is given below. Each question is numbered along the y-axis. One color represents one player. Simple integers are listed along x-axis.
Two important graphs are also presented here.
The following graph shows the time vs. efficiency of each player in experiment 3.
As the graph suggests, every player got at least 50% answers right if not more. And these included somewhat complex word problems. This (along with feedback from the players themselves) proves our point that (1) a tutorial before the quiz and (2) questions in both English and Urdu proved really helpful for the players.
This second graph shows the number of failed join attempts (players trying to join a group to play a game) in experiment 1 and experiment 3. Very few failed join attempts (in experiment 3) shows:
- The players were beginning to develop a level of comfort with the game
- The format of the text message to join a game was made easier in experiment 2. The positive results in experiment 3 proves it to be a good step.
A video interview with a couple of players is given below:
Conclusion
We can conclude on the basis of the results of multiple experiments that a lecture before a game improves the performance and interest of the participants in the game significantly. A participant can then focus on applying a concept to real problems. For effective usage, this system should accompany lectures on a public medium like TV broadcast, followed by SMS questions.
Experiments have shown that participants were far more comfortable with multiple choice questions than questions requiring descriptive answers. This observation is a direct implication of the fact that a cell phone keypad is not appropriate to enter long text.
To overcome the language barrier, questions were presented both in the English and the Roman Urdu format. Participants found questions written in both the languages to be quite helpful. This establishes that the language barrier can be overcome if questions are presented in both English and Urdu.
To overcome the language barrier, questions were presented both in the English and the Roman Urdu format. Participants found questions written in both the languages to be quite helpful. This establishes that the language barrier can be overcome if questions are presented in both English and Urdu.
As a final remark, we strongly claim that this SMS based teaching system can greatly aid in teaching the masses. People who have left education due to some reason can have a chance to refresh their already learned concepts and acquire some new concepts through this system.
Source Code
To access the source code of this application, kindly use this link: http://code.google.com/p/smsmathquiz/
Posted By
Awwab Us Sabah Siddiqi (awwab84@gmail.com)
Sheikh Muhammad Mustafa (sheikh.muhammad.mustafa@gmail.com)
Raza Shah Hamdani
Hasan Maqbool
