My World 2015 and mobiles - post 6 of 6 - concluding thoughts and questions

These 6 posts aimed to give a summary of the groundbreaking My World 2015 survey and explore how mobiles have been and are being used to promote and distribute the survey.

My World 2015 aims to survey individuals across the globe with regard to their priorities in public policy. The survey allows respondents to choose 6 out of 16 pre-selected priorities or to submit their own priority in a 17^th ‘fill-in-the-blank’ option. Respondents have participated in the survey via pen-and-paper ballot, via a central website, and through mobile technology (SMS, IVR, and a mobile application).

To review the posts from the beginning, here are the links: 

1. My World 2015 and mobiles - post 1 of 6 - an introduction to the posts
2. What is My World 2015? My World and mobiles - post 2 of 6
3. How are mobile phones used to distribute the My World survey? My World and mobiles - post 3 of 6
4. Interactive voice response (IVR) and My World 2015 - My World and mobiles - post 4 of 6
5. biNu and My World 2015 - My World and mobiles - post 5 of 6
6. And this one...

Some questions I am still looking to answer when it comes to how My World is getting done...

1. How were the 16 priorities were determined? My understanding as described in the paper is 1) the United Nations Millennium Development Goals (MDGs) Campaign made 15 thematic categories based on the MDGs and existing related research, 2) A Uganda mobile survey asked users to name one policy area of priority and these were cross-checked with the 15 thematic categories, 3) A network of MDG and development experts were consulted and the 16 topics were finalised. 
2. Any specifics on dropout rates available (per SMS, IVR, biNU)? 
3. Did the positioning of the priorities in the survey (e.g. healthcare offered as an early option to respondents) possibly impact decision, or were priority positions random in the different survey methods? 
4. Can anyone find the link to Yemen video seen by 3 000+ viewers? Was this also actively distributed on Facebook, Twitter, etc or was there no point given it was on Yemeni TV?
5. Where exactly was interactive voice response (IVR) offered as a survey option? I can find specific mentions of Yemen, Rwanda, and India, but a My World ‘How To’ document suggests that additional countries (Bolivia, DRC, the Republic of the Congo, Nigeria, and Bangladesh?) may also have included IVR options. 
6. Is this link to the SMS version of the survey distributed in Yemen and only SMS in only Yemen or all the types of the survey done in Yemen? 
7. Any study comparing SMS vs. IVR (dropout rates, response rate, etc.)? 
8. How many SMS messages in the survey? Was it one text per priority or one text per four priorities (as with IVR)? Were these four priorities grouped together randomly?  
9. A cost analysis or proposal would be great to see to have an idea of overall costs for GeoPoll's partnership or biNu's promotional work. 
10. biNu has a number of unregistered users – could these individuals fill out the survey? 
11. How does biNu offer advertising, e.g. is it similar to Facebook ads or is it just Facebook ads? I Googled this but could find no clear examples of biNu ads. 

biNu and My World 2015 - My World and mobiles - post 5 of 6

biNu and My World 2015

The final approach to mobile distribution of the My World 2015 survey is a partnership with biNu.

biNu is a cloud-based application that allows “almost all types of [feature – specifically Android or Java-based] mobile phones to access Internet applications and services running in the cloud with near instant response times, even on slower or congested 2G (GPRS / EDGE) networks”. biNu claims that users access 10 times less bandwidth for an experience that is 10 times faster than standard mobile browsers.

biNu has had an active user base of 3-5 million during the first six months of 2014. The user base fluctuates as more and more people get smart phones or change providers.

How is the survey done via biNu?

My World 2015 partners with biNu to both distribute and promote the My World 2015 Survey. Users with accounts on biNu can find the survey in the survey section of biNu (which also offers free credit to users that fill out surveys). biNu also supports survey distribution with advertisements and reminders for account holders.

Most biNu account holders use the application to chat, share content and to access largely text-based news, sports updates, and information.

Approach of the biNu My World 2015 survey

My World 2015 noted the popularity of biNu amongst mobile users in developing countries. biNu acts as a cheap means by which to directly target the owners of inexpensive feature-phones for surveys (as little as 20 cents a question, according to biNu’s website). Users receive credits for answering surveys, allowing them to access more features as well as more time on biNu.

The My World 2015 survey allows each biNu account holder (registered user) to fill out the survey once. However, there are some unregistered users, and it’s unclear as to whether or not these users could successfully access or fill out the survey.

Implementation of the biNu My World 2015 survey

In addition to being offered as one of many surveys made available to biNu users, My World 2015 received promotional placement from biNu including Facebook ads, automated reminders sent to registered users, and some video promotion. Moreover, clickthrough rate to the My World 2015 survey via biNu increased with the added incentive of 10-20 biNu credits for completing the survey.

Results from the biNu My World 2015 survey

By early June, biNu survey distribution resulted in over 100 thousand votes from over 180 countries. About 84% of all respondents are male, and 80% of all respondents are between the ages of 16 and thirty. Over 80% are in or have finished some sort of secondary education. The largest number of votes has come from India (about 18%), perhaps due to strong promotion about the survey via high-profile Bollywood stars.

The top three priorities for biNu respondents include better education, better job opportunities and better healthcare. “An honest and responsive government” came in fourth overall.

Click here for up-to-date data from biNu and My World 2015. 

Future plans for and lessons learned from the SMS My World 2015 survey

biNu continues to distribute the survey. The largest concerns in the biNu distribution include the disproportionate amount of votes from men and a high dropout rate.

To increase female participation, My World 2015 has increased promotions targeting women, using female celebrities and offering additional biNu credit incentives to women. My World 2015 is considering targeting only women via biNu, but this has not yet been done given the survey’s main goal of remaining non-exclusive.

To decrease the dropout rate, My World 2015 has asked biNu to distribute automated reminders to those that began but did not complete the survey.

Interactive voice response (IVR) and My World 2015 - My World and mobiles - post 4 of 6

IVR and My World 2015

Interactive voice response (IVR) versions of the My World 2015 allowed respondents to call a toll-free number and respond to automated voiced questions, usually in a local language, by pressing buttons on their mobile phones in response to automatically voiced questions.

How is the survey done via IVR?

As with the SMS and other options by which to take the survey, local partners and campaigns promoted both the survey and the local options by which to take it to potential respondents. For countries where IVR was offered, this includes promoting the toll-free number by which to take the survey in all relevant My World 2015 promotional media, from print articles to fliers and billboards to promotional SMS messages.

To set up the IVR platform in each country where it was offered, My World 2015 worked with local NGOs to record the survey in local languages. To set up the technical platform behind the IVR survey, My World 2015 and partners worked with an international company specialized in mobile technology as well as national and local mobile service providers.

Approach of the IVR My World 2015 survey

To make sure the My World 2015 survey was non-exclusive, distributors wanted to ensure that there was a survey option for illiterate individuals or individuals who did not have their own mobile phones. For this reason, My World 2015 offers an IVR option, where respondents call a toll-free number, pick their preferred language, and then respond to automated verbal questions by pressing buttons on their phone.

This survey option, when available in a specific country, was promoted in all relevant media and local outreach.

Implementation of the IVR My World 2015 survey

Working with Kirusa, an “international mobile technology provider”, connected to mobile service providers in Bolivia, the Democratic Republic of the Congo, the Republic of the Congo, Nigeria, and Bangladesh, as well as with local NGOs and telecom companies in different target countries, My World 2015 encourages partners and interested parties to record the survey in widely-spoken languages and distribute a toll-free number allowing respondents to call in and take the survey.

In a ‘How To’ document, My World 2015 explains how the IVR option works once set up:

•      The user will receive an SMS message on their mobile phone introducing them to MY World through a simple message and giving them a toll-free number to call to take the survey. They may also be offered a non-financial incentive to participate.
•       When an individual calls the toll-free number, an automated telephone system will allow them to choose their preferred language. They will then be asked four multiple-choice questions about the issues that are most important to them (each time selecting one from a choice of four MY World options). At each stage they will dial into the telephone keypad the number corresponding to their choice. Finally, they will be asked their gender, age and education level.
•       [My World 2015] estimate[s] that a caller will need a maximum of two minutes to complete the automated survey.
•       [Callers] will not incur any charges for making the call.

To record the survey in local languages, My World 2015 relied heavily on NGOs and other in-country partners. My World 2015 provided the prompt and script (usually in English), and then the partners translated, recorded, and edited the local transcripts.

Results of the IVR My World 2015 survey

The data portal for My World 2015 does not offer a clear look at all IVR results. Instead, those interested can access the results from the on-going IVR survey in Yemen, where Y-Telecom, local UN Volunteers, and the NGO Y21Forum have heavily promoted the toll-free number. MenaVAS and a local UNV, Mr Ahmed al- Ashwal, have taken care of the technical side and reporting for the Yemeni IVR and SMS version of the survey.

As of early June 2014, over 100 thousand Yemenis responded to the survey via IVR. (Only 46 thousand have responded via SMS in Yemen.) About 31% of these respondents are female (versus 23% in the SMS version of the survey). Fifty-nine percent of all respondents are between the ages of 31 and 45 while 41% are between the ages of 16-30.

As with the Yemeni SMS version of the survey, respondents ranked a better education, better job opportunities, and an honest and responsive government as their top priorities. Better healthcare ranked 7^th collectively (though it did rank 5^th amongst the largest responding age group, ages 31-45).

Click here for up-to-date IVR results from Yemen. 

Future plans for and lessons learned from the IVR My World 2015 survey in Yemen

IVR response rate, in Yemen at least, is more equitable with regard to gender and slightly more equitable when it comes to education level. More people have chosen to respond to the survey via IVR vs. SMS, perhaps because it is easier to call in and respond to the survey via phone, generally costing the respondent only about two minutes of his or her time, instead of responding to several text messages.

IVR seems to be, based on the Yemen case study, a more effective means of engaging respondents to the My World 2015. The dropout rate appears to be lower and the overall response rate higher and slightly more equitable with regard to gender. Offering the survey via IVR is more costly with regard to time and technical set-up on the side of the distributor; however, results suggest it is time well spent.

Further analysis was not possible because the data was not disaggregated to view IVR versus SMS survey data.