The methodology of the Telefonica Index on Digital Life (TIDL)

Digital Life is a fascinating concept.  It almost defies measurement. Digital Life is also ubiquitous. You find it everywhere, even in the remote jungles of Thailand where life has not changed for thousands of years. The TIDL is a project to capture the complex value of Digital Life by creating an index. We use the GEDI index building methodology to accomplish this. Index building is a statistical method that captures a difficult to measure concept like Digital Life, by using a complex average of many data points to produce a value that can be easily understood by most people.

The framework of Digital Life is composed of three concepts—Digital Openness, Digital Confidence and Digital Entrepreneurship. These three concepts are the building blocks of Digital Life. The concepts are supported by pillars.  These pillars are the important components of each sub-index. For example, in Digital Openness, regulatory conditions that affect market structure, open innovation, and customer choice are important. These are important in all countries.

While Digital Confidence and Digital Openness are market and regulatory focused concepts, Digital Entrepreneurship is about action. It is about people and the digital businesses they start. In order to capture this human action the four pillars of Digital Entrepreneurship education, opportunity start-ups, product innovation and finance was multiplied by the countries millennials attitudes towards entrepreneurship. This informs us about the extent of digital opportunity.

To build a great Index you need three things:  a clear conceptual framework to guide the collection and interpretation of the data; a solid methodology to build the index; and lots of good data on the phenomenon. The important thing to remember in index building is that whatever we put in as data should be something that is broadly considered important.  If it’s not important we leave it out. All inputs are not equally important but it is difficult to quantify what is more or less important so we treat all the variables as equally important. The focus in the TIDL was on collecting original data across the 34 countries, picking the best variables and making sure that they were up to date. Finally, all data and methods should be well documented.

Picking variables is both an art and a science. The pillars of the TIDL are made up of more than 50 Key Performance Indicators (KPI). KPIs are the building blocks of the index. They represent the most basic data. They are aggregated into variables and the variables are allocated to their respective pillars. The index building methodology requires three steps, the capping of the data at the 95th percentile, the normalization process of the raw data between (0-1), and average pillar adjustment. The different averages of the normalized values of the variables imply that reaching the same indicator values requires different effort and resources. Since we want to apply the TIDL for policy purposes, the additional resources for the same marginal improvement of the indicator values should be the same for all indicators. Therefore, we need a transformation to equate the average values of the variables.

The resulting index gives us an easy and reliable way to think about the complexities of Digital Life. The most readily useful aspect of an index like TIDL is that it allows us to rank countries on a scale of (0-1) on their overall Digital Life, as well on the three primary components of that Life – the sub-indices. This is a helpful snapshot of Digital Life. Second, and more important, it gives us a large pool of information about country’s Digital Life that is benchmarked both against other countries and against the pillars in any one country. This benchmarked information makes it easier to understand where a country excels and where it underperforms, and can therefore help guide policy towards the best possible destination. In essence, we have a compass and a road map, to understand how easy or hard it is for someone to navigate Digital Life in a given country and what could be improved.

Data is always challenging, especially if you are measuring something as elusive as Digital Life. While TIDL uses the best data available, in our case, as in many others, not all the data are available for all countries. This missing data results in a potential distortion of the results. To address this, we developed a measure that estimates the potential distortion of missing data. There is no reason to believe that any of the missing data would be an extreme outlier for that country causing a distortion of the final TIDL score.

TIDL could be improved with more and better data. For example, we do not have machine to machine interactions (direct communication between devices e.g. via the Internet of Things). We do not have data on the percentage of business that have a web presence. We do not know the number of digital start-ups to total start-ups. We do not know how many start-ups are growing. We do not know the number of jobs digital business is creating as a percentage of total employment.

As new data becomes available on an almost daily basis TIDL can be made even better.