Discover's Demographic Data

Discover's demographic sample is representative of a variety of traits of the American population aged 18-69, including age, gender, and location. The following article illustrates the composition of this sample and how demographic traits are calculated.

Basic Traits

Gender

If not declared in the user’s profile, our models determine gender from first names, bio and affinities with an accuracy of 94%.

Age

Discover’s age sample ranges from 18 to 69 years. If age is not declared within the profile, our models can estimate age from affinities and bio information with an accuracy of 90% in 3 years ranges.

Generations

  • Generation Z: Born 1997-2010 (only including respondents aged 18+)
  • Millennials: Born 1977-1996
  • Generation X: Born 1965 and 1976
  • Baby Boomers: Born 1946-1964

The Generation trait is a derivative of age. 

Relationship Status

  • Single: Those who don’t have a declared relationship, are divorced or widowed
  • In a relationship: Those who are in a declared relationship or engaged
  • Married: Those who declare they are legally married 

If not declared in the user’s profile, our models can determine people’s relationship status from their social relationships, posts and bio information with an accuracy of 81%.

Education

  • High School: people who are attending, or attended, high school
  • College: people who are attending, or have attended, college
  • Graduate School: people who are acquiring, or have acquired, a masters or doctoral degree

If not declared in the user’s profile our models can estimate people’s education from their social relationships, bio information and affinities with an accuracy of 84%.

Race

  • White: A person having origins in any of the original peoples of Europe, the Middle East, or North Africa
  • African American: A person having origins in any of the original peoples of Africa (excluding North Africa)
  • Asian: A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam
  • Other: a person not falling in any of the previous categories

Our models to determine race are created by a manually validated training set which is applied to the data with an accuracy of 93%.

Ethnicity

  • Hispanic:  A person of Hispanic origin
  • Not Hispanic:  A person not of Hispanic origin

Our models to determine ethnicity are created by a manually validated training set which is applied to the data with an accuracy of 88%.

Location

We determine location based on the user's declared location.

Regions

States are designated into the 4 regions you can select on Discover as follows:

  • Northeast: Pennsylvania, New York, Vermont, Massachusetts, New Hampshire, Maine, New Jersey, Connecticut, Rhode Island 
  • Midwest: North Dakota, South Dakota, Nebraska, Kansas, Missouri, Iowa, Minnesota, Michigan, Wisconsin, Illinois, Indiana, Ohio
  • South: Texas, Oklahoma, Arkansas, Louisiana, Mississippi, Tennessee, Kentucky, Alabama, West Virginia, Virginia, Maryland, Delaware, North Carolina, South Carolina, Georgia, Florida
  • West: Montana, Wyoming, Colorado, New Mexico, Arizona, Utah, Idaho, Washington, Oregon, Nevada, California. Alaska, Hawaii

States

Discover's study includes people from all 51 US States and from US territories.

Metro Areas

You can find 948 metro areas on Discover. The metro areas are delineated based on US Census Bureau standards.

Home and Work

Income

The calculation of income is based on location, and is specifically determined by zip code. The income of each zip code is defined by data disclosed by the IRS. 

Presence of Children

  • With children: people who have 1 or more children actively within the household
  • Without children: people with no children or “empty nesters”

If not declared in the user’s profile, our models can estimate the presence of children from images, posts, bios and affinities with an accuracy of 83%.

Language

Three elements are taken into consideration to determine language, in order of importance:

  • The language used in posts and comments 
  • The language set in the browser
  • The language the person declares that they speak

Our models can estimate language with an accuracy of 98%.

All the languages that can be selected on Discover are listed in the following table:

EnglishSpanishFrenchGermanItalianChinese
JapaneseMandarinPortugueseRussianArabicHindi
HebrewKoreanGreekVietnamesePolishCantonese
DutchSwahiliCzechTagalogRomanianUrdu
DanishCreoleSerbo-CroatianTurkishPersianSwedish
PatoisGujaratiTamilPunjabiIndonesianUkrainian
TeluguAfrikaansSlovakMalayalamHungarianNorwegian
YiddishBengaliEsperantoFilipinoMarathiThai
SanskritArmenianCatalanAramaicAlbanianBulgarian
FinnishTaiwaneseKannadaFarsiHmongHawaiian
LithuanianShanghaineseJawaYorubaIcelandicLatin
SlovenianEstonianWelshLatvianMacedonianGallegan
IrishMalayKhmerSinhalesePushtoFrisian
Kurdish     

 

Employment Status

  • Employed: Has a job
  • Unoccupied: Unemployed, looking for a job, and available for work; are not seeking a job; or are unable to work
  • Student: Main occupation is to attend school or otherwise get an education
  • Retired: No longer actively working

Our models can determine employment status from bio info, posts and affinities with an accuracy of 91%. Check this link for more in depth information

Job Function

  • Management 
  • Healthcare 
  • Other Services 
  • Finance and Accounting 
  • Manufacturing and Production 
  • Sales
  • Marketing and Advertising 
  • Arts, Design, Entertainment, Sports, and Media 
  • IT and Technology 
  • Engineering, Research and Development 
  • Office and Administrative

Our models can estimate job function from bio info, posts, social relations and affinities with an accuracy of 80%.