Determining Sample Size: How Much is Enough?

Many associations who are about to implement a large quantitative survey where the size of the population is unknown (e.g., surveying the total number of prospects within a market) often ask us the question, "How large of a sample do we need to get good results?" The answer to this is that there is no finite number that determines whether a sample is "good" or "bad". Instead, there are other factors that need to be taken into consideration. Below are some tips that can help your association determine the appropriate sample size to meet its specific needs when dealing with a population size that is unknown.

• Decide how much error you want associated with the results: Since it's nearly impossible to survey every single member of larger populations, statisticians use a formula to determine the error associated with results. This statistic is known as the margin of error or the assumed error that corresponds to a specific sample size. The chart below summarizes how the margin of error decreases as more responses are obtained.

Margin of Error	Number of Responses Needed
+/-9.8%	100
+/-6.2%	250
+/-4.4%	500
+/-3.6%	750
+/-3.1%	1,000
+/-2.8%	1,250
+/-2.5%	1,500

All of these percentages are calculated at a 95% confidence level, meaning that if we were to conduct the same survey 100 times, we would get similar results, plus or minus the margin of error, 95 out of 100 times.

• Decide what your budget will be: Not only is it nearly impossible to survey everyone in a large population, but it can also be quite costly. The beauty of statistics is that you don't need to survey everyone to get results that will provide you with direction. Prior to conducting your survey, determine how much money you have in your budget to perform the research. This will give you a better understanding of the sample size you can afford to get.

• Decide what your timeline will be: Obtaining large sample sizes can take a significant amount of time. Depending on what your needs are, allow enough time for data collection so you are able to get enough responses that will give you a margin of error that satisfies your needs for the research.

Questions? Comments?

Contact us:

Erik Schonher, MBA

Phone: 703-706-0358

Email: Erik@MarketingGeneral.com

Dr. Adina Wasserman, PhD

Phone: 703-706-0373

Email: AWasserman@MarketingGeneral.com

Jeff Tranguch, MA

Phone: 703-706-0364

Email: JTranguch@MarketingGeneral.com