Here are 8 elements of a research program you should always take into account when planning and executing an effort with culturally diverse audiences:

1. Methodology matters: In 2010, the Pew Research Center analyzed pre-election surveys and found that “support for Republican candidates was significantly higher in samples based only on landlines than in dual frame samples that combined landline and cell phone interviews.” The analysis goes on to show that Hispanics and African Americans comprise a larger portion of duals reached on their cell phone rather than their landline. If pollsters only dialed random landline numbers to gauge national interest, the Hispanic and African American populations would have been underrepresented, and the “insights” gained from the research would likely have been skewed.

2. Screen soundly – Hispanics can be Jewish, White, Black, Arab and more. They can be from the U.S. or from Central or South America. Some may be native Spanish speakers while others may barely be able to say “hola.” They may be fully acculturated to life in the U.S., or be new immigrants just learning and adapting to the U.S. culture. And the variations can go on. Bottom line, know who you need to talk to in order to get to the insights you are looking for.

Might a from a small town in El Salvador newly arrived to Washington, DC have a different perspective on public transportation compared to a NYC born Puerto Rican? It’s quite possible that despite both being Hispanics, their different experiences may very much influence their perspectives - and that’s important to take into account when screening people so that your results are now inadvertently skewed.

3. Moderators make a difference – Selecting a moderator who understands the cultural nuances and non-verbal cues given by individuals from diverse cultures is vital. Those nuances could be in the colloquialisms used, the body language or even the subtle and unspoken sense of hierarchies that may exist within a focus group of people from the same country. Moderators who understand the cultural perspective of groups will also take into account the cultural dimensions of individualism versus collectivism, which can directly impact how people respond to questions. Dutch researcher Geert Hofstede noted that people from Indonesia and West Africa rank toward the collective side of responsibility and reward, whereas individuals from the U.S. and Britain fall on the individualism side.

Also, while the moderator's gender may not matter for some focus groups, male chauvinism is a reality in some cultures. Utilizing a female moderator for a group of men from a male chauvinistic culture may result in a situation where the group is not comfortable or willing to share their true feelings/perceptions about the topic at hand, simply because of the gender divide. Choose the “wrong” moderator and you’ve wasted your money.

4. Craft questions carefully – Words and images matter, particularly when you are conducting research with people from other cultures. If you were interviewing recently immigrated Latinos to a metropolitan area and you wanted to find out about their perceptions on public transportation, you’d have to decide which word to use for bus depending on where the person was from: autobús, guagua, colectivo, micro, buseta, and the list goes on. Use “guagua” with someone from Chile and they’ll think your talking about a baby. Or say you are interested in introducing a new soft-drink and you want to understand how people might prefer to consume it – either with a straw or directly out of a bottle. Say the word “pitillo” to a Colombian and they’ll understand that you are talking about a straw. A Spaniard, on the other hand, will be wondering why you are referring to cigarettes. Others would be expecting to hear the word “paja.” But be careful, the word “paja” to some means masturbation.

Consider using back translation or a collaborative translation approach when creating moderator guides and developing messaging that you want tested.

5. Select images carefully - In some cultures dogs are looked at as dirty creatures and not considered “mans best friend” as is often the case in the U.S. In other cultures the cow is sacred, whereas for others it's looked at as food. Like words, images take on different meanings across cultures and subcultures.

Work with cultural experts and translators who understand the nuances of the cultures. When using images to convey ideas or draw out insights, be mindful of how cultures may view the same image from a different cultural lens. Otherwise, you may get some interesting information that is completely off subject.

6. Consider the audience’s geopolitical situation – Several years ago I spoke to a marketer with an expertise in the U.S. Iranian community. He noted that most Iranians in the U.S. would not finesse their taxes or claim all IRS exemptions. Why? Because Iranians are afraid of being questioned or hassled by governments, they’ve had bad experiences with them. This piece of information would be important for anyone polling Iranian Americans about government related subjects. You’d need to weigh your understanding of their feeling toward governments when reviewing their responses to questions. Understanding the geopolitical situation of a target audience allows you to get to valuable insights about why an audience may perceive a product/service in a particular way or elicit a specific type of behavior.

7. 2 + 2 ≠ 4 - Research shows that respondents from different countries do not use positions on a scale in the same way. For example, respondents from Asian Pacific countries tend to use extreme ends of the scale when responding to questions with multi-response scales. According to an abstract of a July 2006 Journal of International Business Research "Americans are thought to exhibit a 'midpoint bias' when confronted with a statement with which one must indicate some measure of agreement or disagreement on five-point or seven-point scale... By contrast, respondents from cultures in Asia Pacific countries such as China are thought by some to exhibit an extreme response style, picking numbers toward either extreme of the scale." This means you need to understand how to “calibrate” responses. You need to read between the lines understanding cultural cues be it body language and ways that people describe something.

8. Lean in to listen - The Greek philosopher Epictetus said, “We have two ears and one mouth so that we can listen twice as much as we speak.” No truer more relevant words could have been spoken about researching diverse audiences. As marketers, PR professionals, designers and researchers, we need to actively “listen” to the audiences that we’re studying. That means not just listening to the words spoken, but observing and taking in the whole picture. Several years ago, when leading research for a highly targeted immigrant community (they were not Hispanic), we held focus groups in four markets across the U.S. over several days. The findings during the focus groups were incredibly important and helped us shape effective creative and messaging points. But one of the biggest things we noticed was what happened outside of the focus groups. By the time we’d finished the first set of groups in the first market, communities in the other three markets were already in the know that we were coming. Word spreads like wildfire. The insight: if we took a wrong step in any market, we’d be dead in the water throughout the U.S. Had we not been actively listening, we would have missed this key bit of info.

As with all that we do in marketing, we’re constantly learning. By taking into account the 8 items noted above you’ll be ahead of the game when trying to target diverse audiences.

*Why identify 8 components? Well, we’re talking about multicultural audiences and for the Chinese, 8 is a lucky number. I figure we all need all of the help we can get, so 8 it is! Good luck.