How do you adjust your sample size for different survey modes, such as online, phone, or mail?

1 Response rate

The response rate is the percentage of people who complete your survey out of the total number of people who are invited or contacted. It is an indicator of how well your survey reaches and engages your target population. Different survey modes have different average response rates, depending on factors such as convenience, accessibility, trust, and incentives. For example, online surveys tend to have lower response rates than phone or mail surveys, because they are easier to ignore or delete. However, online surveys may also have lower costs and faster turnaround times than other modes. To adjust your sample size for different survey modes, you need to estimate the expected response rate for each mode, and multiply your desired sample size by the inverse of the response rate. For example, if you want a sample size of 500 and you expect a 20% response rate for an online survey, you need to invite 500 / 0.2 = 2500 people to take your survey.

2 Sampling frame

The sampling frame is the list of people or units that you can potentially select for your survey. It should match as closely as possible your target population, which is the group of people or units that you want to generalize your findings to. Different survey modes may have different sampling frames, depending on the availability and quality of contact information, such as email addresses, phone numbers, or postal addresses. For example, phone surveys may have a larger and more representative sampling frame than online surveys, because more people have phones than internet access. However, phone surveys may also have more problems with non-response bias, because some people may not answer or refuse to participate. To adjust your sample size for different survey modes, you need to assess the coverage and accuracy of your sampling frame for each mode, and adjust your sample size accordingly. For example, if you have a sampling frame of 10,000 people for a phone survey, but only 80% of them have valid phone numbers, you need to increase your sample size by 20% to account for the missing data.

3 Margin of error

The margin of error is the range of values that contains the true value of a population parameter with a certain level of confidence. It is affected by the sample size, the sampling design, and the variability of the data. Different survey modes may have different margins of error, depending on factors such as the sampling method, the question format, and the measurement error. For example, mail surveys may have higher margins of error than online or phone surveys, because they are more prone to non-response bias, low data quality, and delayed feedback. However, mail surveys may also have higher response rates and lower costs than other modes. To adjust your sample size for different survey modes, you need to determine the acceptable margin of error for your research question and objective, and calculate the required sample size for each mode. For example, if you want a margin of error of 5% with a 95% confidence level for a mail survey, you need a sample size of about 385, assuming a simple random sampling and a 50% response distribution.

4 Choosing a mode

Choosing a survey mode is not a straightforward task, as it involves balancing cost, quality, and feasibility. Different survey modes may be better suited for different research purposes, contexts, and populations. To make the best choice for your survey, you need to consider your research goal and question, your target population and sampling frame, your budget and timeline, your data collection and analysis methods, as well as any ethical and legal obligations. After assessing these factors, you can select a survey mode that meets your needs and expectations, with an appropriate sample size.

5 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?