Data collection in the Caribbean has traditionally been a clunky exercise, heavily dependent on phone calls and focus groups – but ever since mSurvey  (a mobile surveys company which started in Kenya) set up shop in Trinidad and Tobago, the region has started to experience data collection very differently.
The company  recently undertook its largest project to date, “interviewing” more than 11,000 people across Trinidad and Tobago, all via mobile phones and SMS , in less than three weeks. The survey was commissioned by the Telecommunications Authority of Trinidad and Tobago  (TATT), the regulatory body for telecommunications in the country, in an effort to assess the gap that exists between people in the twin island republic who have access to basic telecommunications and broadcasting services and those who do not. In the words of the report  itself:
The Digital Divide Survey is related to the economic development of a country through technological means, and looks at persons with access to technology, the utilization of such modes of technology, and the ease of access to such.
Once the extent of the digital gap became known, TATT would have access to tangible information that identifies which communities and groups are under-served, including those with disabilities, with a view to helping the authority improve the reach and quality of online services. (TATT is reportedly planning to implement infrastructure to the value of over $20 million US.)Kristal Peters , Director of Business Development and Strategy for mSurvey's Trinidad and Tobago office, to find out more about the survey and how it could help to make the virtual playing field more level for people across the country. This is the first part of the interview; a second post will follow.
Global Voices (GV): Why do you think this survey was important to TATT?
Kristal Peters (KP): The digital divide is defined in two ways:
1. The Organization for Economic Co-operation and Development (OECD) defines the digital divide as the gap between individuals, households, businesses and geographic areas at different socioeconomic levels with regard both to their opportunities to access ICTs and to their use of the Internet for a wide variety of activities;
2. The Academic definition of the term digital divide refers to the disparity in accessing to the technologies and resources of the information and communication.
In part, the solution for bridging the digital divide is to enable universal access and service – making available basic telecommunications services at an affordable price for all in society.
A significant requirement of the Authority’s mandate speaks to ‘promoting universal access to telecommunications services for all persons in Trinidad and Tobago, to the extent that is reasonably practicable to provide such access’. In accordance with this, the Authority has drafted a Universal Service Framework and accompanying Regulations which outline targets as well as the administrative system to define and execute particular initiatives to provide telecommunications services to [the] widest possible population. The determination of these targets and initiatives, which will change over time, will be influenced by the results of this Digital Divide survey.
Also contained within the Universal Service Framework and Regulations are Universal Service Obligations where service providers are requested to provide basic telecommunications services to selected communities, identified by the Authority. [This survey was] the mechanism required to identify [and provide information about] these ‘underserved’ groups in accessing or affording basic telecommunication services.
GV: Can you summarize the findings of the survey? What is the reality of the digital divide in Trinidad and Tobago?
KP: Trinidad has a high mobile phone penetration of 140% which can be leveraged for ICT innovation, [but] there is still a barrier to Internet penetration at home with a 45% penetration. Over 70% access the Internet; however, that is through a hybrid ecosystem of cafes, mobile phones, schools, libraries, and home use.
GV: Do you consider access to the Internet as a right or a privilege? Why?
KP: Access to information can bring about creativity, innovation, and develop a deeper understanding; and the Internet can be used as a portal to that access. The Internet was a privilege that is in transition to becoming a right, in the same way that education was once a privilege but is now a right. Access to the Internet has, over time, become inextricably linked to accessing resources relevant in numerous areas of life – from education, to healthcare, to government services, and the list goes on. Lack of access to the Internet can stymie a person’s (and by extension, his/her family’s) capacity to grow their personal and professional networks, enhance their skill sets, be informed about employment and other market opportunities…in a nutshell, with the Internet being a wellspring of information, lack of access to it means deprivation from many opportunities to improve quality of life.
GV: We understand that your company provided TATT with more than 720,000 datapoints in a real time data feed. That sounds pretty impressive in terms of scope, but can you break it down?
KP: A datapoint relates to each piece of information gathered from a survey respondent. That is, each answer submitted by a respondent to a particular question is a datapoint. Each survey ranged between 17 to 34 questions, which equates to between 17 – 34 datapoints for each survey. There were about 24,000 folks who accessed the Digital Divide Survey.
GV: Did mSurvey do TATT's first Digital Divide Survey, which the authority states was completed some time back? Did mSurvey have to bid for this new project? If so, what do you think gave your company and its approach the edge over other tenderers? And how did the landscape change between the time the first survey was conducted and now?
KP: The 2013 Digital Divide Survey (DDS) was the first to be conducted by mSurvey in Trinidad and Tobago. mSurvey responded to an Request For Proposal for this project, and was awarded the contract. mSurvey creates a unique value proposition: our mobile technology allows for the collection of data in real time, in a shorter space of time (conceivably, thousands of persons can be surveyed simultaneously), [it] is less labour-intensive (survey deployment and data tabulation is automated), economical, and our technology integrated directly with the mobile network operators to improve data integrity and the ease and flow of data from multiple sources.
From the respondents’ perspective, mobile surveys maintain absolute anonymity (there is no person-to-person interaction, and mSurvey applies patented security algorithms to reduce data vulnerability); mobile surveys are flexible (can be taken at any time, and anywhere); communication via mobile surveys is free (respondents are not charged for the SMSs they send or receive during the course of the survey); and as in the case of the DDS, respondents received $10 top-up credit automated upon completion of the survey regardless of whether they were pre-paid or post-paid [cell phone] clients.
With mobile penetration rates in TT exceeding 100%, virtually everyone in the country has a mobile phone. With such high mobile penetration rates, we felt that the environment was ripe for conducting a nationwide survey – the first nationwide mobile survey for Trinidad and Tobago, and the first nationwide survey for mSurvey. During the period of time between 2007 (when the previous DDS was conducted) and 2012 (when we were awarded the contract for the project), there was a significant increase in the number of mobile phone subscriptions in [the country]. Furthermore, the mSurvey platform is SMS-based, therefore any kind of mobile phone (be it a Smart Phone , or a feature ‘Me Too’ phone ) could used to access the survey.
GV: How was the survey actually conducted? What types of questions were included? How was the information gathered and assessed and why did you choose to do it by mobile phone?
KP: [First] mSurvey collaborated with TATT to produce the survey questions. We examined the indicators associated with each index and then drafted questions that mapped directly to the indices which would extract the necessary information from respondents. There were multiple-choice questions, open-ended questions, and ‘select all that apply’ questions.
[Then], a nation-wide marketing campaign was launched to inform the general public about the survey, and to give directions on how to participate. These directions were necessary because prospective respondents needed to initiate the survey themselves by texting the word ‘digital’ to the number 8288. mSurvey’s policy is not to spam mobile subscribers with a broadcast message inviting them to participate in the survey for two key reasons: we are not in possession of a master list of phone numbers from service providers and allowing prospective respondents to ‘opt-in’ to the survey puts the respondent in charge, and avoids bombarding them with additional text messages. Research has shown that the response/uptake rate to spam SMS is quite low – around 2%. How many times to you respond to spam SMS from your mobile service provider?
Once the respondent sent the joincode (digital) to the shortcode number (8288/TATT), he gained access to the survey and a welcome message, along with the first question [that] would be sent. Every time a user submitted a response, it would automatically be analyzed, once the answer was valid. If [it wasn't], the same question would be resent to the respondent with an additional prompt on how to enter a valid response. There were limits on the number of times a respondent was allowed to submit an invalid response, after which he would be locked out of the survey.
The first five questions were demographic in nature and allowed us to ensure that the sample of participants matched national demographic distribution according to national census. Prior to launching the survey, quotas were set for the number of participants allowed per region according to their age and gender. If a survey participant entered data indicating that he/she was of a certain age and gender, and from a particular region for which the quota had been met, they would not be allowed to complete the survey. [No-one was] allowed to access the survey more than once.
Out of the 24,000 people who accessed the survey, just over 11,000 actually completed it. [Those] who did not complete the survey either stopped answering questions, or were bumped out of the survey because their profile quota had been met, or they had exceeded the number of attempts allowed per question.
The remaining questions covered a range of topics such as cost of monthly mobile service, [the quality of] Internet accesses (e.g. at home, work, Wifi hotspot, etc.), education level, and monthly household income. There was also an additional sub-survey which gathered information about people with disabilities.
With mobile phone penetration estimated at 140% in Trinidad and Tobago, the mobile phone is ubiquitous and an ideal tool for data collection. Compared with other modes of data collection such as paper surveys via enumerators, or even online surveys, mobile surveys have the potential to access many more respondents, over a shorter period of time (many respondents can participate simultaneously; shorter relay between respondent and final recipient of data), and can be more economical.
GV: Explain the ramifications of not having access to this kind of information in terms of a country's productivity, economic development and sustainability.
KP: Decisions related to a country’s productivity, economic development and sustainability are driven by data. In the absence of data, or reliable and valid data, effective decision-making can be extremely challenging and/or misguided.
Look out for Part 2 of this post, in which we discuss the four main objectives of the Digital Divide survey, the challenges the company faced once the process began, and of course, the results.