How can you keep your survey cost down, while ensuring accurate and trustworthy data? Manageable sample sizes are especially important for surveys conducted by National Research Center, Inc. (NRC) that deal with large populations, like The National Citizen Survey™ (The NCS™) or the Community Assessment Survey for Older Adults™ (CASOA™). In these cases, a “sample” is the number of households asked to complete a survey. Sampling is both a scientifically accurate and cost effective way to understand the population as a whole. NRC Senior Survey Associate Damema Mann explains.
Surveying each and every resident within a city, town or county is just too expensive and time intensive to be viable for most local governments. Put simplistically, larger the sample sizes yield more accurate results, up to a certain point. A sample can get so large that the amount of increased accuracy actually becomes negligible. And for a population survey, the more households you survey, the higher the costs of mailing and analysis. But the good news is you do not need a very large sample size to have a very low, acceptable margin of error. NRC’s recommended sample sizes strike a balance between margin of error and cost so that citizen survey results are statistically valid and representative of the entire community.
The basic sample size for The NCS is 1500 households. Over NRC’s 20+ years of experience, this number has proven to yield accurate representations of communities around the U.S. with various population sizes. For those local governments wanting to send out hundreds more surveys, The NCS pricing page shows the impact of larger sample sizes on cost. Also, NRC survey experts are well-versed in making sample size recommendations to clients for the specific needs of their unique communities.
Modern technology allows us to expand the reach of flagship surveys. In addition to the randomly selected sample size, The NCS basic service includes a web survey that is open to all residents wanting to opt in.
Understanding survey sample sizes can also help you win birthday party guessing games. Say you are challenged to estimate how many marbles in a jar of 1,000 are red. You can’t count out every marble, but you can shake the jar well, take a handful of marbles, count the red ones and get a good idea what percentage of marbles in the jar are red.
The same concept applies to sampling a large pot of stew. You only need a (yummy) spoonful to sense how the entire pot tastes. In the same way, a smaller number of residents can be surveyed to better understand the community as a whole.