[x_section style=”margin: 0px 0px 0px 0px; padding: 45px 0px 45px 0px; “][x_row inner_container=”true” marginless_columns=”false” bg_color=”” style=”margin: 0px auto 0px auto; padding: 0px 0px 0px 0px; “][x_column bg_color=”” type=”1/1″ style=”padding: 0px 0px 0px 0px; “][x_text]AIQ has recently launched our second generation ethnic predictor, Ethnic-IQ™.
As is the case with many other marketing fields that are widely viewed as being ‘known’, ethnicity is actually an inferred field. While we cannot speak as to how other companies gather this data, we can state that this data is often wrong. We see numerous cases (in standard compiled data) where individuals are obviously misidentified. No doubt, anything can happen – but you should not see many cases of individuals identified as ‘Asian American’ having names such as ‘William Johnson’ and ‘Brandon Smith’.
Excluding the small number of ‘hand raisers’ on marketing files who have self-identified as a particular ethnicity, the primary characteristics used to infer ethnicity are:
- US Census (2010)
- First Name
- Last Name
- Religious affiliation (in some cases)
- Purchase and behavioral data
It is important to note that none of the above characteristics are a silver bullet – large doses of analytics are needed to accurately predict ethnicity.
Does our approach work? An AIQ client with a large ‘truth’ file (customers who had self-identified their ethnic origin) compared our predictions and found that we were over 90% correct. Given that our ethnic universe is significantly larger (particularly in individuals that are predicted to have Hispanic or African-American ethnicity), this is a major achievement.
No ethnic predictor is perfect. AIQ can only claim that ours is a lot better.
| Ethnic Group | AIQ% |
|---|---|
| African American | 10.7% |
| Asian American (Chinese) | 1.2% |
| Asian American (Indian) | 0.7% |
| Asian American (Other) | 1.5% |
| Caucasian (Jewish) | 2.6% |
| Caucasian (Other) | 64.3% |
| Hispanic | 13.2% |
| Middle Eastern | 1.1% |
| Native American/Pac Islander | 0.4% |
| Other | 0.0% |
| Unknown | 4.4% |
[/x_text][/x_column][/x_row][/x_section][x_section style=”margin: 0px 0px 0px 0px; padding: 45px 0px 45px 0px; “][x_row inner_container=”true” marginless_columns=”false” bg_color=”” style=”margin: 0px auto 0px auto; padding: 0px 0px 0px 0px; “][x_column bg_color=”” type=”1/1″ style=”padding: 0px 0px 0px 0px; “][x_custom_headline level=”h2″ looks_like=”h3″ accent=”true” class=”mts”]About AnalyticsIQ[/x_custom_headline][x_text]AnalyticsIQ is a consumer marketing analytics company founded in 2007, with a focus on building the best-engineered, most reliable data products to drive better analytical and marketing performance for our clients. Using a combination of analytical expertise, sophisticated modeling techniques and thousands of data elements, AnalyticsIQ has built a highly predictive suite of data products specifically designed to improve analytics and deliver results.[/x_text][x_custom_headline level=”h2″ looks_like=”h4″ accent=”true” class=”mvn”]Contacts[/x_custom_headline][/x_column][/x_row][x_row inner_container=”true” marginless_columns=”false” bg_color=”” style=”margin: 0px auto 0px auto; padding: 0px 0px 0px 0px; “][x_column bg_color=”” type=”1/2″ style=”padding: 0px 0px 0px 0px; “][x_text class=”plm” style=”border-left:1px solid #a7c8d8;”]
Anne Smith
[icon class=”mrs” type=”envelope”]annes@analytics-iq.com
[icon class=”mrs” type=”phone”](678) 773-8798[/x_text][/x_column][x_column bg_color=”” type=”1/2″ style=”padding: 0px 0px 0px 0px; “][x_text class=”plm” style=”border-left:1px solid #a7c8d8;”]
Dave Kelly
[icon class=”mrs” type=”envelope”]davek@analytics-iq.com
[icon class=”mrs” type=”phone”](770) 778-3056[/x_text][/x_column][/x_row][/x_section]