In recent years it has been increasingly difficult to reach younger target audience for national representative studies due landline phones primarily being used by older respondents and also lower response rates from respondents in younger age groups.
Additionally landline penetration has been decreasing or stagnating while mobile phone penetration in most of Western European countries has reached almost full coverage. Nevertheless it remain difficult to cover the younger age groups (18-24,25-34,35-49) sufficiently.
A fist step has been increasing the share of mobile phone numbers in a dual frame sample. Additional steps can be taken to further screen the mobile phone sample for younger age groups by matching mobile phone numbers for various attributes:
➡ Activity on Social Media accounts (Twitter, Instagram, Facebook etc)
➡ Content matching via AI & Machine Learning
Each of the steps improves the accuracy for younger or specific age groups, at the same time it introduces various biases in the sample due to non-coverage of the entire smartphone / messenger or social media account population.
📰 Methodology
For targeting younger audiences, Sample Solutions only makes use of mobile sample. A multi-step approach is used based on the various levels. The generation of mobile phone RDD sample can be taken from our standard RDD methodology for each country and will not be repeated here.
The starting point is the entire numbering plan that exists for a specific country.
2.1 Raw RDD Mobile Generation
2.2 Provider Lookup for active SIM Cards
Operators around the world are connected by a network known as the “SS7” or “Signaling System #7” network. This network is used for exchanging information related to setting up a telephone call, number translation, prepaid billing systems, and also forms the backbone of the SMS (short message service) text messaging system.
The “Location Register” is at the core of every GSM mobile network worldwide. It is a central database of all customers authorized to use the network, and is used by the network in the authentication of a phone when registering with a transmitter, either on the home network, or when roaming abroad.
By sending the appropriate message over the SS7 signaling network, it is possible to query the home location register of any mobile subscriber using their phone number. This process is known as performing a “Provider Lookup” or “Provider Query”, and is a fast, effective, and accurate way to determine which network operator currently serves a given mobile telephone number and also the status of the mobile phone.
2.3 Link to Social Media profiles
The next step involves linking up phone number with Social Media profiles. In many cases social media profiles do not just make use of a unique email address but also a mobile phone number which allows linking up Social media profiles with mobile phone numbers using the API which is used where available.
From the Social Media profile, information such as images, content, names and profile information can be derived in case they are available publicly. All the generated data is only taken in case there is public information available and lastly linked back to the unique mobile phone number. In most cases this info is also available in google search with profile preview.
Additionally we analyze whether there are other secondary websites where the mobile phone number is listed on together with location information. The analysis of the other websites or social media profile can assist as well in determining the primary language.
2.4 Face Recognition
By analyzing the various pictures posted, we can also find estimations for the age based on face recognition, analyzing the various pictures post and publicly available. In addition to the age, a gender can be derived. The whole analysis only provides an estimate for an age range with a specific probability – by including multiple pictures it is possible however to increase accuracy.
Get a free trial Sample for your next project by filling out our form or contact us via sales@samplesolutions.euf
Carsten Broich
Carsten is the founder of Sample Solutions and Lifepanel with over a decade of sampling and social research experience. A trained aerospace engineer who discovered his love for random phone numbers.
