For the past few years, I have been fascinated by the evolution of the discussion on “Data Quality”. 

As recently as 2005, this discussion revolved around “Respondent Cooperation” and the need to address a continuing decline in response rates.  Then, in November 2006, things really got interesting.  During a Respondent Cooperation Summit hosted by RFL Communications, Kim Dedeker of P&G shared a case study concerning online data integrity.  Ms. Dedeker reported receiving different results from two separate online research providers on the exact same project.  Overnight, the Respondent Cooperation issue evolved into one of Data Quality. 

Now it’s time for this discussion to come full circle, and return to the basic concept of Respondent Cooperation.  If respondent cooperation levels were anywhere near what they were just twenty years ago we wouldn’t be discussing data quality.  In fact, the dwindling pool of available respondents is the spark that has ignited the data quality flames. 

Even our term “respondent” speaks to the commodity status we place on our most critical benefactor.  And a “professional respondent” is even worse; this is associated with those crazy outliers who actually enjoy taking surveys.   Only our industry would attack the people keeping it aloft.

Unfortunately, the relative importance of the professional respondent has been magnified by the continuing decline in cooperation among the public at large. Let me take pains in separating out “Professional Respondents” from “Fraudulent Respondents”- it’s the fraudulent group (e.g. Speedsters, Straightliners, Satisfiers) for which we have set up our traps. The Professional Respondents are more benign, they actually do attempt to answer correctly and honestly. Still that doesn’t take away the fact that they belong to multiple panels. Years ago, these folks would have been “box-top” people (e.g. your wacky Uncle Frankie who clipped every possible sweepstake entry).  In the 21st century, they have traded their scissors and envelopes for PCs and notebooks. They now feed their sweepstakes obsessions via online market research surveys.

What happened to our seemingly endless source of fresh respondents?  I think the answer is best exemplified by the comic strip Pogo: “We have met the enemy and he is us.”

It wasn’t easy… we had to keep hammering away at it for close to half a century, but the public finally got tired of our “just a few moments more” pretext and relatively lame incentives, and turned their backs on us.   After all, it’s not like the respondent is some petulant child who simply refuses to cooperate. 

They lost their willingness to cooperate because we had broken the tenant that holds any relationship together, trust.

We looked at the respondent as a disposable good, one that we use once and then cast aside.   Their sustainable value becomes the next researcher’s problem.   You can almost imagine a huge landfill of former respondents, folks that used to take surveys but have now soured on the experience.  This Tragedy of the Commons is playing out before our eyes.

Can we reverse the tide of respondent disenchantment?   Well, we would have to actually care what the respondent thinks after they finish our survey (or qualitative experience).  This is normally the moment we cast them aside, since we’ve already gotten what we need from them.   

Patrick Glazer of MRA/CMOR has emerged as an industry spokesperson on declining cooperation rates.  Patrick revisited the definition of respondent cooperation and properly put the emphasis on “the relationship between the public and the researcher.”  This is a far more proactive definition than the previous “pertains to the degree to which the public is willing to participate in research.”

The long term sustainability of our industry is incumbent upon our willingness to strike up an honest and transparent relationship with respondents.   In fact, I prefer the word “participants” over “respondents” because I feel “Participants” is a more respectful term that better describes folks who are willing to enter into this relationship with us.

So let’s focus our attention on ways to maximize “Participant Engagement”.   Most of this is achievable:

• Keep surveys to reasonable lengths,

• Incorporate interactive features

• Add some multimedia entertainment value

• Properly tie the incentive to the survey experience

• Share survey highlights

• Acknowledge participant input and contributions

• All in all, honor our portion of the relationship agreement.

In addition, I recommend employing the IMRO Survey Satisfaction Survey five-point scale as a way of gauging the participants’ experiences

• Time and effort required to complete the survey was reasonable

• This survey was a good use my of time

• This survey was well written

• I am being adequately compensated for my input

• This survey was well organized

• The survey’s subject matter was important to me personally

• Overall, I am satisfied with the survey experience

Do you have the courage and conviction to ask these questions?  Are you willing to ask them with every survey experience?  If the answer is yes, that means you actually care what the respondent felt about the experience.  Simply asking the questions, regardless of the answer, is a sign of respect to our industry’s lifeblood.

The time has come to attack the root of the Data Quality problem - Participant Engagement.  We can set all the traps we want, but that’s only masking the overriding problem, which is the fact we have become over reliant on the “lunatic fringe”* to answer our surveys.   Minimizing the effect of the Professional & Fraudulent Respondent, let alone reducing non-response bias, is completely contingent on reestablishing a relationship with the public.   Any other attempt to directly combat Data Quality Issues is merely window dressing.  You want to kill the problem; you have to attack the cause.  That is why we need to enter a relationship with engaged “participants”.

Of course, this all speaks to a set of standards that we as an industry would need to follow.  It requires a commitment to participant sustainability, perhaps even an MRA Certification, for this to have teeth.  Most importantly, it would require end users to resist the siren song of “fast and cheap” research that has weakened our industry and broken our trust with the participant.   Are you going to remain part of the problem or become part of the solution?

Respecting the fact that the participant is a limited resource; it is our responsibility to sustain this asset for the next generation of market researchers.   It’s time to GO GREEN my friends!

NOTE: I would refer the reader to RFL Communication’s “Platforms for Data Quality Progress” for a free download  http://www.rflonline.com/

*    “The lunatic fringe” coined by Dennis Murphy of Marketing Stories LLC