In This Issue

Volume I March 2005

Forward This to a Colleague

Seeing Red, Feeling Blue
By Andy Arthur, VP Client Services, MRI 

In the world of market research, the Presidential election cycle is a long-anticipated opportunity to put theory into practice and, in some cases, to steal the show. In that spirit, the 1996 election famously ingrained "soccer moms" into the collective consciousness and the 2000 and 2004 elections hammered home the idea that there are two types of voters: Red and Blue. One group is rural, plain-speaking and right-leaning, preferring the Republicans; the other is urban, intellectual and left-leaning, favoring Democrats. Using respondent-level data from the 2004 Fall MRI study, we set out to explore if the Red/Blue split - particularly the much touted "value" divide - is as real as the media and exit polls suggested. 

Strictly speaking, the designation of Red or Blue is done by state. Everyone who lives in a state that went for Bush is Red (even those people in a Bush state who voted for Kerry), and everyone in a state that went for Kerry is Blue. By scoring each of its nearly 26,000 respondents as Red or Blue based on their state of residency, MRI was able to tabulate "Red Staters" and "Blue Staters" against its vast single-source repository of demographic, media, psychographic and product usage data. 

Scratching the Surface
A demographic comparison of the Red and Blue populations in the MRI study confirms what we already know: Blue Staters are concentrated heavily in the largest urban areas. Nearly 55% of all Blue Staters live in the top ten markets, as opposed to only 31.5% of the general population.

Several other demographic differences seem to emanate from that urban skew. Generally, the larger an urban area, the higher the salaries and cost of living. Indeed, Blue Staters have a median household income of $58,254, 28% higher than the Red state median HHI of $45,502. Because income and education are considered to be related, it makes sense that Blue Staters also have higher education levels: whereas Red Staters are 13% more likely than the general population to be high school drop-outs, Blue Staters are 15% more likely than the general population to earn a 4-year degree, and 19% more likely to earn a post-graduate degree.

Other demographic characteristics provide less contrast. Red Staters have a median age of 44.8, only 1.5 years older, on average, than Blue Staters. Blue-state households are only 2% more likely to include children. Such ambiguous results reflect the inherent limitations of the Red State/Blue State split.

Higher Resolution Redness (and Blueness)
Since we know that many Kerry voters reside in Red States, and many Bush voters live in Blue, we can expect the differences observed by comparing these states to be less distinct than the differences we would see if we were able to directly compare Bush voters and Kerry voters. To establish stronger correlations between voting choice and other attributes, we investigated a new way to segment voters.

Theoretically, if the likelihood of an individual to vote for Bush ("Redness," as opposed to "Blueness") is positively correlated with a given characteristic, then that characteristic should be more prevalent in populations where more people voted for Bush. So we matched the addresses of MRI respondents to county election results, scoring each respondent with a percentage based on the voting split in his or her county. In other words, a respondent living in Bergen County, NJ, where the vote went 52% for Kerry, was assigned a Blueness value of 52, and a Redness value of 48. 

Having marked each respondent with his or her county voting percentage, we were able to isolate specific values and ranges on the Red/Blue Spectrum. The resulting distribution follows a bell curve, with more than half of the population (55.6%) living in a county that voted within ten points of an even split between the two candidates. As one would expect, the curve is shifted toward the Red end of the Spectrum. However, the extent of the shift is more pronounced than the election results suggested - in the election, 51.4% of voters chose Bush, whereas 54.3% of adults live in counties that did so. The difference is that the Red counties were significantly "Redder" than the Blue counties were Blue; 20.3% of adults lived in counties that were more than 60% Blue, whereas 24.1% lived in counties that were more than 60% Red. Only relatively higher voter turnout in Blue counties, it appears, kept the race interesting.

32.1% of the adult population is reclassified when the Red/Blue assignments are done by county instead of state, meaning that they are either occupants of Red counties in Blue States, or vice versa. A shift of that magnitude, naturally, has a profound impact on the properties of the two groups. Overall, the gap in the median household income between the Red and Blue populations shrinks from 28% to a mere 6.1%. Looking at median HHI on the Red/Blue Spectrum, we can see that the positive correlation of household income and Blueness is really fairly weak. The "Reddest" extreme of the population may be relatively downscale, but so is the "Bluest" extreme. To put this in perspective, the Reddest county in the U.S. (which went 93% for Bush) was the very rural Madison County, ID (for the record, the 12th-largest county in Idaho); the Bluest was the District of Columbia (91% for Kerry), home to a population that, while very urban, is not affluent. In between the two extremes, the HHI numbers are relatively flat.

The Great Values Divide 
A consensus emerged following the election that demography alone was an inadequate predictor of voting choices, and the best way to understand what happened behind the curtain was to look at a person’s values. But the exit polls which brought the personal values differences to light were widely discredited for inaccurate projections early on Election Day, and the question remains: Were personal values really correlated with voting choices? MRI’s battery of LifeMatrix personal values statements allows us to test this assumption and we did, indeed, find a value divide on the Red/Blue Spectrum.

For starters, there is a tendency for "Bluer" respondents to rate "Wealth: Having Material Possessions, A Lot of Money" and/or "Status: Achieving a Higher Social Status" either a 6 or 7 on a 7-point scale of importance. "Being in Tune With Nature: Fitting in With Nature" was another Blue value, as was "Social Responsibility: Working for the Welfare of Society." If you’re wondering why the Kerry camp devoted so much time to discussing the stormy clouds on the economic horizon, or the consequences of Arctic drilling, look no further.

Meanwhile, the Bush team was touring Red America, arranging for gay-marriage referenda on state ballots and chartering buses to transport people from churches to polling places on Election Day. They seem to have been aware that, to the Red public, "Faith: Holding to Religious Faith and Belief" and "Traditional Gender Roles: Following Traditional Gender Roles for Men and Women" were among the key issues. To appeal to another Red value, "Duty: Fulfilling Obligations to Family, Friends and Country," the war in Iraq was characterized in terms of "getting the job done."

New Targets?
Red and Blue voters have proven powerful, which raises a key question for marketers: Do Red and Blue consumers exist? Isolating the Reddest third of the population and tabulating them against all MRI standard product categories by a "likelihood to use" index, it’s clear that we are onto something.

The numbers show that the Reddest third of the population is 30% more likely than the general population to have used "Pectins" (gelling and thickening agents for homemade preserves, jellies and jams) in the last six months, 23% more likely to have used "Chili," and 21% more likely to have used "Packaged Moist Dog Food." By contrast, the bluest third of the population was 29% more likely than the general population to have used "Cognac," 20% more likely to have used "Vegetarian Frozen Burgers," and 17% more likely to have used "Soy Milk."

Given the income/education difference between the Red and Blue populations, one might expect the Blue population both to afford and to seek more informational input through the media than the Red population, leading to higher overall media consumption. In fact, those differences do exist, but, given the geographic dispersion of the Red population, they are surprisingly slight.

On a per-capita basis, in counties that selected Kerry over Bush, adults read 9.8 magazines in a typical 30-day period; in Bush counties, the per-capita average is 8.9. Blue county residents listen to 16.6 hours of radio in a typical week, whereas their Red county counterparts listen to 15.7. When it comes to TV, the Red population averages 28.4 hours per week of TV, compared to 27.1 for the Blue population. 

There is some irony to the observation that the segments of the population least likely to change their minds - those at either end of the political spectrum - are the most news-hungry. Residents of counties that voted 80% or more for either candidate watch, on average, 11% more cable news programming than the population as a whole. The differences between the Blue and Red extremes, though, are telling. The bluest tenth of the population watches more than twice as many hours of BBC America as the average adult; their Red counterparts watch 30% more Fox news.

back to top

Driving Your Brand Forward … By Narrowing Your Focus
by Chris Perry, National Manager, Hyundai Motor American

Chris Perry, National Manager, Hyundai Motor American, discussed the custom segmentation system that helped re-build the Hyundai brand at a recent NOP World-hosted conference. Large, multi-national companies have come to rely on research providers such as MRI and NOP World to help them focus their marketing efforts and develop real world solutions. Excerpts of Mr. Perry’s presentation speak to the power of research as a means to address strategic brand positioning issues. Mr. Perry has nearly 20 years experience in the market research, strategic planning and marketing fields and has worked in supplier, agency and client environments. 

How do you build your brand? Companies continually ask themselves this critical question. To many, the answer is elusive. But some do find the solution and succeed at redefining and growing their brands. The task seems daunting at first, but once it is achieved, there is a sense of complete satisfaction. 

Time for a Radical Change
Hyundai entered the U.S. market in the late 1980s. Sales quickly climbed, and we sold more than 260,000 automobiles in the first three years.

Shortly afterwards, however, we experienced quality problems, and sales spiraled down to 90,000 cars in 1998. Our decline reached the point where Hyundai seriously considered dropping out of the U.S. market. 

Before making a final pullout decision, however, we made changes previously unheard of in our industry. We introduced a 10-year, 100,000-mile warranty on all Hyundai products. Then we initiated several quality improvements and repackaged and re-priced all of our products. Finally, we made a commitment to dealers that we would stand behind these changes and, in return, we asked each of them to renew their commitment to the Hyundai franchise. Within just one year, automobiles sales increased 82 percent, jumping to 164,000. Hyundai’s quality initiative is one of the key reasons behind our success. Last May, JD Power’s annual IQS Survey showed that, in terms of quality, Hyundai and Honda are tied for second place, and are only one point behind Toyota.

Sales, profits and quality have increased significantly over the last five years, but we still face challenges, many of which can be summed up by a woman who, when introduced to one of our new models said, "Wow, I just love that vehicle, but I’ve worked too hard to drive a Hyundai." 

Hyundai recognized the need to expand beyond our core segment if we wanted to double sales. And, we needed to modify our brand image to re-align customers’ perceptions with the reality of the new Hyundai. 

To further develop our brand and heighten awareness, we conducted a custom study [among consumers planning to purchase automobiles] incorporating LifeMatrix, NOP World’s segmentation tool that details the consumer landscape using values, lifestyle, life stages and demographics. The study addressed three general areas of questioning. 

1. A product perspective - asking detailed information about the styling, the feel, the features and the characteristics people want in an automobile. 

2. A consumer perspective - gathering intelligence on target consumers’ values, lifestyles and lifestages (LifeMatrix battery). 

3. A brand perception standpoint - gaining insights into the perceptions of our brand and competitive brands. 

Of 11,000 surveys mailed, we received 5,700 responses - an impressive 52% response rate to our 28-page questionnaire. 

NOP World then created 10 custom consumer segments derived via the LifeMatrix attributes plus our specific automotive-related questions. Interestingly, the demographic differences among the 10 segments were miniscule; however, when we looked at product needs and consumer characteristics, significant differences among the segments were discovered. 

Understanding Consumer Needs and Desires
The segmentation study provided a complete understanding of the product needs and characteristics of our consumers, enabling us to identify the right target segments for our brand - those who would help differentiate us from the competition, who held values and beliefs consistent with our brand and who would be receptive to our product pipeline. The study identified our prime targets: people who strive for an authentic lifestyle, including being a good parent or spouse … being true to oneself … and being involved in social causes. 

Based on this study, we also developed a battery of questions and an algorithm that allowed us to drop those questions into additional studies. Now, when we conduct continuous tracking and copy testing studies, we not only get results on how the general market responds to our product development initiatives and promotional messages but also on how specific targets react. 

Importantly, the study also allowed us to build a bridge model with other NOP World databases, including Mediamark Research. Using LifeMatrix questions from our survey, we scored the MRI database based on our segmentation criteria. When we analyze media runs, we now can look beyond general demographic segments to our own specific targets, ensuring we make the most effective media buys. In addition, we now have the capability to find out how our target consumer is interacting with and consuming media. From an activity standpoint - using the MRI database - we’re able to understand the activities our targets participate in. For us, this is one of the key takeaways.

back to top

DemandMatch and MRI Data Identify Your Best Prospects 

For many companies, direct marketing is a suspect mix of science and luck - with hit rates one rarely needs more than a few fingers to count. Now DemandMatch, a direct marketing database integration tool that combines syndicated data with proprietary customer information, can increase a marketing campaign’s ROI by identifying those prospects and customers most likely to take advantage of an offering. 

DemandMatch was developed by Mediamark’s sister-company NOP World. At its core it is a rigorously tested and proven bridge-modeling formula that enables marketers to pinpoint their target in direct marketing campaigns and to ascribe data to customer lists. The application can be extended to niche markets to help, for instance, digital cable providers deliver addressable messaging to individually targeted households. DemandMatch increases the likelihood that each consumer receiving a direct marketing message is a likely candidate to respond positively to that message.

The MRI Connection
The merging of MRI data within the DemandMatch tool provides clients detailed, actionable information about their marketplace. MRI’s representative sample, high household count and overall robust data produces an accurate snapshot of the U.S. public while DemandMatch identifies - by name, address or phone number - individual households that fit MRI profiles. "Once a target population is identified, MRI can be used to form a cohesive marketing strategy that can be projected to a greater number of prospective customers," says SVP, Database Marketing/DemandMatch Mike Dolan. For example, MRI data can find a specific consumer population, predict their purchasing behaviors based on previous purchase/usage and intent-to-purchase data and project this information to others who share similar consumer behaviors, values and lifestyles. "A direct dialogue can then be established with these consumers," adds Dolan. 

The capability to accurately measure ROI - either of campaign, word-of-mouth, customer loyalty or advertising - enables DemandMatch clients to gauge the effectiveness of current efforts and plan future campaigns that deliver the right message to the right consumer. Direct marketing inherently lends itself to ROI measurement and MRI’s reliable, robust data further increases ROI by initially identifying a client’s best prospects. 

For more information about how DemandMatch can help your direct marketing programs, contact your MRI account representative.

back to top

• SPORTS ENTHUSIAST - baseball, basketball, football, Frisbee, soccer, softball, tennis and volleyball. 
• PUZZLES & GAMES - backgammon, board games, crossword puzzles, electronic games (not TV), bingo, cards, word games, and trivia games. 
• HUNTERS & FISHERS - archery, fishing (fresh water), fishing (salt water), hunting with a bow and arrow, hunting with a handgun, hunting with a rifle, hunting with a shotgun, and target shooting. 
• OUTDOOR SPEEDSTER - auto racing, power boating, jet skiing, motocross, motorcycling, roller blading/inline skating, snowboarding, snowmobiling, and water skiing. 
• OUTDOOR ADVENTURER - backpacking/hiking, canoeing/kayaking, horseback riding, jet skiing, rock climbing, sailing, scuba diving, skiing (cross country), skiing (downhill), snorkeling/skin diving, surfing/windsurfing, and whitewater rafting. 
• CULTURED NESTER - attend other music performances, attend dance performances, concerts on radio, dining out, cooking for fun, entertain friends/relatives at home, go to live theatre, go to museums, and reading books. 
• COLLECTOR - model making, coin collecting, figurine collecting, stamp collecting, sports trading card collecting, electric trains, tropical fish. 
• PASSIVES - These are individuals that tend to participate in the activities (leisure and sports) included in the analysis at lower levels than the general population or not at all. 

back to top

The Source
Mediamark Research Inc.
A NOP World Company
75 Ninth Ave 5th Floor, NY, NY 10011
www.mediamark.com           www.mriplus.com

New York (212) 884-9200 • (800) 310-3305 • Chicago (312) 329-0901 • Los Angeles (323) 882-6325

Comments? Want to be removed from The Source mailing list? E-mail and let us know.

^©2005 MRI All Rights Reserved