the countries. The coverage of telephone directories, of Internet, and of mobile cells provides the feeling of the need to adopt a mixed‐mode approach. A World Bank study reports that, in the 2018, Euro area fixed telephone subscription for 100 people is 44.4, mobile 122.6 and Internet 83.8. Table 1.2 shows the same indicators by country. Only some countries are shown in the table since the objective is just to evidence that there is a relevant difference across countries.
Table 1.2 Penetration of fixed and mobile phone and of Internet (year 2018)
Source: Data from International Telecommunication Union. World Telecommunication/ICT.
| Country | Fixed telephone subscription (% of inhabitants) | Mobile cellular subscription (% of inhabitants) | % of individuals using the Internet |
|---|---|---|---|
| Austria | 42 | 125 | 88.0 |
| Denmark | 19 | 125 | 97.6 |
| Finland | 6 | 132 | 88.9 |
| France | 59 | 108 | 82.0 |
| Germany | 52 | 129 | 89.7 |
| Greece | 47 | 116 | 72.9 |
| Italy | 34 | 137 | 74.4 |
| The Netherlands | 35 | 124 | 94.7 |
| Norway | 11 | 107 | 96.5 |
| Portugal | 50 | 115 | 74.7 |
| Romania | 19 | 116 | 70.7 |
| Slovenia | 33 | 118 | 79.7 |
| Spain | 40 | 116 | 86.1 |
| Sweden | 24 | 125 | 92.1 |
| Switzerland | 39 | 130 | 90.0 |
Note that Internet users are individuals who have used the Internet (from any location) in the last 12 months. Internet can be used via a computer, mobile phone, personal digital assistant, games machine, digital TV, etc. Fixed telephone subscriptions refer to the sum of active number of analogue fixed telephone lines, voice‐over‐IP (VoIP) subscriptions, fixed wireless local loop (WLL) subscriptions, ISDN voice‐channel equivalents, and fixed public payphones. Mobile cellular telephone subscriptions are subscriptions to a public mobile telephone service that provide access to the PSTN using cellular technology.
1.2.5 WEB SURVEYS AND OTHER SOURCES
Current digital environment and technology trends are providing a huge amount of data about most phenomena. These data are available on the web. Often they are free of charge, if not protected for privacy.
Examples of data available in digital format are credit card transactions, tax data, social chatting, telephone use (calls details: time, location, length of the call, etc.), social security payments, GPS, videos. Using this type of data for statistical purposes is appealing and challenging. The term big data is currently used to characterize data with high volume, velocity, and variety. There is a debate on the definition and on the use of big data for statistical purposes.
Roughly speaking, big data are based on the automatic collection on everything that people do; they are not subject to statistical classification criteria and to statistical treatment for representativity. Also administrative data, i.e., information that are collected for registering units (people, businesses, sales, and so on) into an activity process, might be included into big data.
Practitioners and researchers are now wondering if big data could substitute web surveys to provide information for social and economic decision making. For a discussion about that, see Couper (2013).
It should be emphasized that conclusion is that big data and web surveys are complementary data sources, not competing data sources. The availability of big data to support research provides a new way to approach old questions as well as an ability to address some new questions that in the past were not considered.
Web surveys may be run as stand‐alone surveys. However, source integration is a major trend for the future of the next 10 years of web surveys.
It is a new area of research to achieve the three following goals: (1) Minimize the cost associated with surveys. (2) Maximize the information, i.e., the findings based on big data generate more questions, and some of those questions could be best addressed by web surveys or other traditional survey methods. Moreover, information from one source could be useful for improving data to be estimated from a survey. (3) Minimize the respondent burden. Integration alleviates the burden of duplicating data gathering efforts and enables the extraction of information that would otherwise be impossible.
Therefore, it is necessary to work in the direction of using big data and integrating them with the survey results. It is important to face experimental applications having in mind the characteristics, nature, and the limitations of big data as statistical sources and the methodological soundness of the survey results.
At the time being, market research and private/public businesses have great interest in trying to use big data to investigate markets and individual behavior. The use of this data as exploratory source is the most plausible application, whereas using this data for statistical purposes and integration with web survey requires still a lot of effort around definitions, classifications, and estimation methodological problems.
Official statistics producers are investigating how to use other big data sources and how to produce estimates in a multisource framework. Some experiments have been already undertaken, with contrasting results. Most successful applications consist in the integration of web survey data and administrative data (i.e., administrative data could be considered a type of big data according to many authors). Administrative data have been used:
To generate a survey frame or to supplement/update an existing frame. When surveys are run on the web, administrative data integration could help in applying the adaptive survey design (see Chapter 8) to improve the data collection process. An ultimate task could be the replacement of data collection (e.g., use of taxation data for small businesses instead of seeking