Quality data = quality insights.
Data quality is the most important aspect of all research projects.
Unfortunately, there are so many areas where this can go wrong. The sample plan is impractical, the screener questions are worded poorly, the survey questions are inconsistent so accurate comparisons cannot be made…just to name a few. With relentless focus on the planning and design details, we are able to guarantee our clients receive valuable consumer insights. We take great pride in our commitment to data quality and will never risk the integrity of the data to save money. We are dedicated to each of the four key parts noted below, as they are all equally critical to ensuring quality data.
Research Design.
It all starts with a good research design. Knowing the right methodological mix, research market(s), and sample plan are vital to ensuring the data is relevant and meaningful.
Many times, our clients already know what they want, and that’s fine, but we are still always providing guidance. We have a solid understanding of the different quantitative and qualitative methodologies and techniques and know how to design the ideal mix in order to yield the highest quality results. We will tell you if your sample plan is realistic or if it is unreasonable based on the research market, and we will tell you if we think a certain methodology will result in more relevant insights. We have conducted over 200 projects across the US, so we are fully versed in the different market characteristics and how the research topic and methodology affects recruiting and data collection.
In the world of automotive, we know vehicle sales and how they differ by market, and even how the research location within an individual market can affect recruiting. We are able to design a sample plan that results in the desired respondent mix outcome, but also one that results in quality respondents, not poor quality respondents that result from having to fill impossible to meet recruiting quotas.
Recruiting.
At the heart of any successful project is a quality recruit.
And, at the heart of any quality recruit is a well designed screener. Whether the screener is designed for online, or over the phone, understanding the research audience is key. Having designed hundreds of screeners over the years, we are well aware of what works and what doesn’t. We understand the difference between questions designed for surveys that respondents will read themselves versus questions that are read aloud over the phone and how answers can differ. We also know that responses can legitimately change, so all respondents are re-screened before attending to assure qualification.
At Sift, we have an extensive background working at and with recruiting agencies so he has a keen understanding of the behind-the-scenes processes. In short, we are very aware of potential recruiter cheating and the professional respondent community. All respondents are recruited via our own internal database, client supplied or purchased lists, or trusted partner recruiting agencies. Our research studies are never posted in online classifieds or social media, which typically attracts the “professionals.” We have complied a list over the last 10+ years of names and aliases used by “professional respondents” and other frequent attendees so our clients can be certain we are obtaining fresh, quality responses from the right audience.
Questionnaire Design.
The questionnaire design process can be broken down into four facets: content, wording, scales, and flow. Whether our clients deliver an already well structured questionnaire, or we design it from scratch, we will collaborate in all areas to ensure the content desired is included.
Having designed hundreds of questionnaires, we have a firm grasp on the proper wording and scales to use. We understand when a 9-point scale may be more appropriate than a 5-point scale, and know that scale anchors play an equally critical role in response quality. Selecting thoughtful, appropriate wording and being consistent throughout allows for accurate comparisons of questions on the back-end analysis.
An often overlooked aspect of good questionnaire design is the flow. Completing questionnaires that have a logical flow to the questioning reduces respondent fatigue and keeps them engaged through the entire survey. At Sift, we take the time to develop cohesive and easily understandable questionnaires that have an organic flow.
Data Collection.
Whether it is quantitative or qualitative data, we go to great lengths to ensure proper techniques and processes are followed.
All of our quantitative surveys are programmed using the Decipher survey platform. We take great strides to ensure error-free data by utilizing a stringent review process for survey programming - multiple team members will check survey programming to make sure there are no mistakes and that all data is being collected correctly.
We take various steps to verify the quantitative respondent data we are including is valid and thoughtful, including: reviewing survey completion time (any survey that is completed unusually fast will be thrown out), searching for straight-liners, reviewing all open ended responses, and incorporating red herring questions to identify poor respondents.
Qualitatively, our moderators are some of the best in the industry. All have decades of experience and understand the various techniques used to extract accurate feedback from respondents. They are each highly specialized in their respective fields so they are able to provide their own valuable insights.
When working with clients from other countries who provide their own questionnaires and stimuli, we will always “Americanize” (including fixing grammar) so that there is no respondent confusion when completing surveys.