stone

“ITZ GD 2 TXT”How To Use SMS Effectively In M-Learning : 

'ITZ GD 2 TXT' How To Use SMS Effectively In M-Learning Andy Stone Learning Technology Research Group Kingston University, UK a.stone@kingston.ac.uk

Outline of presentation: 

Outline of presentation Introduction: LTRG/M-learning The experiment Results Analysis of responses Further observations

Introduction (i): 

Introduction (i) LTRG: Learning Technology Research Group Set up 2001, within School of Computing And Information Systems, Kingston University LTRG research interests include: the effective use of multiple media in teaching and learning evaluation from a user-centred perspective actively engaged in the effective use of mobile telephony as a medium to support the above

Introduction (ii): 

Introduction (ii) Current m-learning work includes: Grounded Theory as a means of user-centred research – used for MLEs, now m-learning Supporting students on placement year Optimising interactivity and response (TODAY) What combinations of technologies work best? Provision of timely, relevant support services Just-in-time, just-enough 'cue card chunks' Short tests, revision timetables, more to come

The Experiment: Optimise Interactivity and Response using SMS text messaging: 

The Experiment: Optimise Interactivity and Response using SMS text messaging Late 2001: experiment undertaken with a major UK youth brand Objective: test the effectiveness of different styles of two-way SMS campaign Partners: KU and The OTHER Media 4 main objectives of experiment: evaluate the effectiveness of SMS as a data collection mechanism for the youth brand compare different data collection processes produce guidelines for future SMS activities compare email, SMS and the web as ways of announcing SMS campaigns

Outline of the experiment : 

Outline of the experiment 1000 mobile phone users involved Had to send 4 pieces of data by SMS Sample group was divided into 4 subgroups to test the effectiveness of different types and lengths of SMS interaction One other subgroup included: those who followed a link to the campaign on the website

Outline of the experiment: 

Outline of the experiment The subgroups in the experiment were as follows: Group 1: Announce campaign via email, request all info at once Group 2: Announce campaign by SMS, request all info at once Group 3: Announce campaign by SMS, 2-step process Group 4: Announce campaign by SMS, 3-step process Group 5: Campaign announced on website, 2-step process

Research Questions to be Addressed by Experiment: 

Research Questions to be Addressed by Experiment How well and how quickly do the selected audiences respond to the SMS campaign? Does increasing the number of steps in the process improve the quality of the data? Does increasing the number of steps in the process reduce the number of responses? How does the method of announcement affect the participation, quality of the data and the speed of response?

Replies obtained: 

Replies obtained Relative failure of email vis-à-vis SMS SMS replies much higher than other forms of media used in direct marketing practice Education isn’t the same as direct marketing… …but results support other research work in progress within LTRG, supporting feasibility of using SMS to support our students.

Correct Responses: 

Correct Responses a 'correct' response may require human intervention to 'clean' the data: Group 2 (all info at once) was surprisingly high Keyword usage Essential for more complex interactive applications – at least 93% for all groups Correct punctuation Also important for machine processing of data Examples were given in outgoing SMS messages Again, group 2 (perceived as 'hardest task' group) produced highest quality responses

Speed of response (i): 

Speed of response (i) Median for all groups – about half an hour Median more useful here: This indicates the time by which half the respondents had responded Helps us estimate expected response times where interactivity is required Can discount 'outliers' that would greatly increase the mean statistic. Such anomalies may be due to illness, lost phones, or lack of interest in completing the exercise. Students may be more motivated to support their own learning, so median is the most useful 'average' to use

Speed of Response (ii): 

Speed of Response (ii) we anticipate SMS m-learning applications will be those which are quite time-sensitive; e.g. placement support applications short revision mini-tests reminder/notification messages (time management) other scenarios where the exchange of information within a narrow time-scale is essential. Distribution of responses over time also useful

Distribution of Response Times – SMS much faster!: 

Distribution of Response Times – SMS much faster!

Other observations : 

Other observations A few respondents clearly believe that their messages are being read by a person because they include chatty comments and punctuation; e.g. 'thanks!' 'thanx' 'who is this?'

Other observations: 

Other observations Multi-part dialogues do not appear to reduce respondents' willingness to participate. Mobile users are prepared to co-operate with fairly complex interactions and to supply accurate information. Despite constraints of SMS, effective applications may be developed using current technologies

Other observations: 

Other observations As this generation of users moves on from education to work… … they have grown up comfortable with a notion of m-learning… …as technologies improve, the sophistication of applications can increase Managed properly, this could lead to a strong m-learning culture supporting e-learning and ubiquitous computing.

Contact Information : 

Contact Information We welcome feedback, collaborations, funding, and visits!  EMAIL: a.stone@kingston.ac.uk TEL: +44-208-547-2000 ext 2838 Feel free to contact me for more information! Thanks for listening – enjoy the rest of the workshop… .. and the England-Brazil match (thanks Mike!)