|
|
|
|
 The following describes the process E-Tech uses to encode the three codes: Ethnic, Religion, and Language Preference. This is important to know, it differentiates your Ethnic, Religion, and Language Preference selects from others whom you may compete for business.
E-Tech has compiled more than 893,880 surnames by ethnicity, more than 79,180 first names common to more than one ethnicity, more than 86,660 first names unique to a given ethnicity, more than 350 middle names unique to a given ethnicity or religion, and more than 3,595 expert system rules.
We base these things upon what we call "onomastic variables,” which are certain prefixes and suffixes of surnames which came about being unique to or common to given ethnicity. These surnames were formed incorporating certain historic name forming variables such as, language specific phrases for "Son of" and geographic locators like "southern" or "by the linden trees", etc. E-Tech also has a group of geographic reference tables. It utilizes these tables to determine the proper codes to apply to an individual. Each record passed through the E-Tech process uses the above files as references so it can examine the individual's first name, middle name, last name and the zip code of the record.
|
E-Tech looks at the first name first, to see if the first name is one that matches an ethnically unique first name on its reference files of more than 165,840 first names. If it does match, the analysis is complete, the ethnic code and religion code from the reference file are used, and the Language Preference code is set. This reasoning is important to understand. A person named Nguyen O'Brien will be coded Vietnamese, not Irish. Based upon the person having a first name unique to Vietnamese, E-Tech will code this person as Vietnamese Speaking for the language preference. This person might have married an Irishman but he/she probably still speaks Vietnamese. For example the name, Marisol Koswalski, would be coded as Spanish Speaking not Polish, because her first name is of Spanish origin.
When the E-Tech process does not find unique first names, it looks to see if the name is on its surname reference file. If E-Tech matches the name, E-Tech assigns the ethnic and religion codes and the person is marked as English Speaking (although they may not speak the language of that ethnicity). For example, John Chen will be coded Chinese but as he was named John he probably does not speak either Mandarin or Cantonese. If still no match, E-Tech looks for a match against the thousands of expert rules it stores to see if the name can be tagged using this approach. An example is that any surname ending in the four characters "oglu" is Turkish, no matter what creative variations in data entry preceded these letters. Slightly misspelled or hyphenated names as well as uncommon or varried spellings can be correctly determined.
 For names that are common to two or more ethnic groups, E-Tech's next step is to look at the middle names to be used as a "tie breaker". An example of this would be the name, Peter Yu. Peter Yu could be either Korean or Chinese. Therefore, E-Tech uses his middle name to determine his ethnicity. For example, E-Tech codes Peter Mei Yu as Chinese and Peter Hak Yu as Korean. Another example would be the name, John Smith. Depending on his geographic location, John Smith could be English or African American. The name John Big Eagle Smith, from his middle name, E-Tech would code his name as Native American regardless of his geographic location. E-Tech also uses middle names to define different religions within a group, for example Sikh or Hindu. In E-Tech's Version 7.2, we have further addressed the many requests to solve the Non-Unique Surname Name phenomena. For example, the surname LEE if not identified by first name can be English, African American, Chinese, or Korean. By identifying the dominant ethnicity within a Zip+4, E-Tech can identify these individuals by their correct ethnicity. E-Tech 7.2 has over 2,525 surnames it analyzes during the ENA stage of the process. By identifying the dominant ethnicity within a Zip+4, E-Tech can now identify these individuals by their correct ethnicity.
Next, E-Tech applies geographic locators (the latest census data converted to Zip+4) to certain ethnic groups to determine if an individual is actually African American. E-Tech also searches for surnames that are common to English, Scotch, Irish, Welsh, French, Dutch, or Islamic Groups to determine a person's ethnicity. E-Tech records these previously matched as African American. This method is now improved with new research and data in our E-Tech version 7.2 release to better identify the African American, Hispanic & Asian groups we offer.
Using our first, middle and surname methodology in conjunction with geographic locators, E-Tech's version 7.2 can also identify, Asian and Hispanic Assimilation Levels and Hispanic Country of Origin. This enables you to target your message to those individuals from Cuba, Mexico, Puerto Rico or 18 other Hispanic Country of Origins.
E-Tech's version 7.2 makes one final pass of the "commonly used" first name file by ethnicity to identify additional "language speaking individuals" not previously identified by the unique first names.
Any list or database encoded with the E-Tech system gives up to 95% of your total database submitted as an ethnic or religion select. We believe the accuracy to be comparable to the best direct response lists. You can now offer both quantity and quality. You will have more ethnicities and have a code that tells you whether an individual speaks a given language. In addition, our three-code system enables you to exclude or include certain groups not possible otherwise. For instance, if you want Russians, but not Russian-Jewish, you select Ethnic Russians with Religion as Eastern Orthodox.
If you have any other questions about the E-Tech system our phone number is 866.333.8324
|
|
|
