One of the most common questions we get from our prospective and current clients relates to margin of error when conducting survey research. When looking at a budget for market research, the largest variable cost you will encounter is the fieldwork. When structuring the estimated cost for a study, a market research firm can estimate the amount of time it will take to design a survey, determine the amount of project management time needed, and approximate the amount of time needed for the market research report at the conclusion. However, the majority of cost for typical telephone survey projects is built into the data collection process because it’s often unpredictable. Some market research firms will also bill on a cost per complete basis with or without a set up fee.
The cost of fieldwork for a telephone survey is dependent upon a number of factors including:
- Incidence – the number of respondents who qualify for a survey based upon given parameters.
- Estimated productivity – the estimated number of completes that can be obtained in a typical hour of calling.
- Script length – the length of time it takes an average respondent to complete the survey.
With this is mind, how difficult and how long the survey is will impact your overall cost of the market research. A three-minute quick pulse survey to Central New York residents to inquire about the awareness of restaurants will cost less than a 15-minute survey to renters aged 18 to 24 in four specific Central New York ZIP Codes to inquire about knowledge of the school district budget vote. The first survey has a high incidence, high estimated productivity, and shorter script, whereas the second survey has a low incidence, low estimated productivity, and longer script.
With that in mind, when we meet with our clients to discuss proposals we often hear questions like, “Why did RMS propose 400 survey completes and not 500 or 600?” And rightfully so. By taking a closer look at a margin of error chart and comparing that to cost of fieldwork, we try to provide a little insight into this trade-off.
Based on this chart below, you’ll see that 400 completes is arguably the optimal price point based on the reduced margin of error. Note that the costs listed in this chart are theoretical and are arbitrarily stated to support this post using a rate of $30 per complete and a standard increase of $3,000 per 100 complete increase. As stated above, the cost of fieldwork is highly dependent on a number of factors outside of this scope. Nonetheless, let’s take a closer look below for our research purposes.
You can see the number of telephone survey completes listed on the x-axis and the margin of error listed on the y-axis. The incremental costs listed in the grey boxes represent the relationship between the incremental costs of the additional survey completes versus the percentage point reduction in margin of error. In more basic terms, it answers the question, “What is the additional cost for a more accurate study?” The jump from 300 to 400 survey completes reduces the margin of error by 0.8% (from 5.7% to 4.9%) representing an incremental cost of $3,750. Whereas a jump from 400 to 500 survey completes only reduces the margin of error by 0.5% (from 4.9% to 4.4%) representing an incremental cost of $6,000.
Viewing strictly incremental costs, moving from 300 to 400 completes costs an additional $1,250 for the increased reliability whereas moving from 400 to 500 completes costs an additional $2,250. Basically, it’s double the incremental cost for only a half a percentage point of accuracy. It would appear 400 completes is indeed a sound threshold and optimal trade-off to achieve both low incremental cost and high reliability in a telephone survey. RMS uses a standard of 400 completes for the majority of its telephone survey research because it provides our clients with this favorable trade-off of cost and statistical reliability (approximately +/- 5%). However, if you need to break your data by regions, ages, or any demographic splits – you may need to revisit and think of a larger-scale study beyond just 400 completes.
To read more about why 400 survey completes is considered to be the magic number click here. If you’d like to further discuss how RMS can complement your business through market research contact our Business Development Director Sandy Baker at 315-635-9802 or by emailing her at SandyB@RMSresults.com. Research & Marketing Strategies (RMS) is a market research firm in Upstate, NY serving clients all across the country.
[…] aim for 400 completes which is approximately a +/- 5% margin of error. This threshold seems to be a nice cross between accuracy of results and cost of completes. Margin of error is also highly dependent on the pool or population. If you collect 400 completes […]
[…] participants is representative of the views of the entire population in the target market. The optimal number of completed surveys is 400. 400 completes rests the accuracy of the results within +/- 5% at the 95% confidence interval. […]
[…] at total of 400+ surveys to ensure that the results were statistically significant with a low margin of error. This also allowed for segmented cross-tabs to carry larger sample bases (e.g., location, age, […]