Chapter 2: Creating a nationally representative sample

This is the second chapter of our ongoing blog series on U.S. Hispanic tracking research. As I mentioned in our previous post, more and more brands are starting to integrate Hispanic sample into their ongoing tracking studies and this present a host of new issues sample companies have to deal with.

Creating a nationally representative sample is one of the most critical components of setting up a successful tracking program. Because many trackers have been in implementation for many years, nationally representative quotas on the general market side are taken for granted. Thought went into setting up the tracker and ensuring the quotas are representative on a national basis or representative of the particular target market that the brand or company is looking to track.

The same type of care (if not more) needs to be put into creating a nationally representative Hispanic sample. One of the most common pitfalls we see in setting up a tracker among U.S. Hispanics or integrating Hispanic sample into an existing general market tracker is the use of the same quotas for the Hispanic portion of the tracker.

The U.S. Hispanic population is distributed differently across the nation than the rest of the population. On a high level, the U.S. population is concentrated in the Midwest and Northeast, whereas the Hispanic population is concentrated in the West and Southwest. These regional concentrations are linked to many other important factors that if ignored could lead to skewed, unrepresentative sample.

For example, if you created a nationally representative sample based on general market regional distributions, you’d have Hispanics that are primarily of Cuban, Dominican, and Puerto Rican decent as those countries of origin are over-represented in the Northeast. This skew is the inverse of the actual country of origin representation of the U.S. Hispanic population which skews overwhelmingly Mexican origin and primarily reside in the West and Southwest.

These country of origin differences are crucial to capture correctly because world views and cultural differences are vast among different Hispanic countries of origin. If you get the mix wrong, you could potentially cause your client to make decisions on marketing messaging that resonates with a small percentage of the Hispanic population, ultimately leading to less than stellar sales results.

To create a nationally representative sample of U.S. Hispanics, the key elements to get right up front are the following:

  • Census Region
  • Country of Origin
  • Language
  • Age

Age seems like a quota that would appear in any properly set up tracker, but Hispanics skew almost a decade younger on average than their non-Hispanic white counterparts. If you quota the same age breaks as general market, you will end up with an older Hispanic sample that is not representative of Hispanics at large.

If your client approaches you to help them set up a Hispanic tracker or integrate Hispanic sample into their existing study, it is important you take the time to research how the Hispanic population is different than the general market. I touched on key points in this article that will help you have a high-level framework but understanding your client’s specific needs and the distribution of the Hispanic population will be necessary to make your tracking successful.

Some resources to get your started would be Pew Hispanic and the Census website. As always, we are here to help as well and are happy to answer any questions you may have. Look forward to sharing our next Chapter on Hispanic tracking, mixed mode vs. online only.