Will the Survey Sample Reflect the Whole Community?

by NRC on April 7, 2017

-NRC Q&A-

The truth is, it is extremely expensive to interview each and every resident within a jurisdiction. Luckily, you can save yourself a significant amount of time and money with a survey sample of community instead. But how can you know if your survey sample is scientifically representative of the entire target population? National Research Center, Inc. (NRC) takes a number of steps to ensure that survey samples closely reflect each of our client communities. Senior Research Associate Laurie Urban further explains our process of survey sampling and why our cities can trust in their survey results.

Watch on YouTube

Randomly Selecting the Survey Sample

With our national benchmarking surveys, we randomly choose about 1500 households from a pool of all the households within the jurisdiction to survey. This helps avoid any bias in the sample, and ensures equal dispersion throughout the community.

Boosting the Response Rate

To foster a strong response rate, we send out communications about the survey before we begin collecting responses. First, we typically send a postcard notifying the household that they have been selected to participate. Second comes a letter from city leaders explaining the importance of the survey and appreciation for resident participation. Lastly, we send the actual survey to those selected households for people to complete. Understanding that mailings are often lost or discarded, we keep in-line with industry best practices and send the survey out more than once to give residents a better opportunity to respond. We also strongly encourage local governments to publicize their survey through all of their existing channels of communication. These tactics combined allow residents multiple chances to engage and get excited about the survey.

Comparing the Demographics

Following data collection, we look closely at the demographics of the respondents and compare that to the city’s Census and other known population data. To ensure that each of those diverse demographics are represented, we statistically weight the data so the final results delivered are fully and accurately reflective of the entire community.

Related Articles