Surveying effectively

Define the problem

The first step is to figure out exactly what you're trying to accomplish. Ask yourself the following questions. The answers will help you figure out what you want to know, what you will do with the information when you get it, and the tradeoff between knowing and not knowing.

• What's my objective?
• What do I want to know?
• Why do I want to know it?
• What will I do with the information when I get it?
• How much is it worth to know, and what am I prepared to spend?
• How soon do I need to know?

Once you have determined these things, you're ready to go to the next step.

Figure out the best way to get your information

Now you're ready to figure out the best way to find the answers. If you still aren’t sure you want to do a survey, you might be better off finding the information some other way. These questions will help you figure out the best way to collect your information.

• Do I already know what I want to know?
• What is the best source of data?
• Who knows what I want to know?
• What is the best way to ask?
• How much have I budgeted for information gathering?
• What is the most effective use of the money?

Select the survey procedures

If you decide that you should conduct a survey, ask yourself the questions below. The answers will help you figure out who should be surveyed and whether you should use the telephone, mail, Internet, or some other method or combination of methods (multimode) for your survey. You also will decide whether to ask all of the people in your target population or just a portion.

• Who do I want to ask?
• What is the best way to reach those people?
• Should I use a census or a sample?
• If I use a sample, what kind should I use?
• How big should the sample be? 

Put together a plan

The answers to all of the questions up to this point will also help you develop a plan for the survey. The plan does not have to be extremely detailed, as long as it addresses these basic questions.

• What are you doing?
• How are you doing it?
• How will you promote it?
• How will you know if you were successful?

Create the survey instrument

By now you have decided that a survey is the best way to collect your data, you know who you want to participate in your survey, and you know how you will collect the data. The next step is to develop the survey instrument. 

• Create a list of all of the data that you want to collect.
• Eliminate all the information that you can obtain from other sources (unless you are using it for validation or are planning to use it to study correlation).
• Develop a list of questions.
• Compare the list to the list of data to collect and verify that the list of questions is complete and that there are no extraneous questions.
• Select the question type for each question and write the questions.
• Organize the questions and construct the survey instrument.
• Test the survey by administering it to a small group for feedback.

Now think about how you're going to get people to take your survey.

Develop a marketing and promotion plan

Marketing and promotion may seem like an unnecessary expense. However, promotion will lead to higher response rates so that you will be able to send out fewer surveys and still be able to have a statistically valid sample size. In addition, to the extent that marketing and promotion activities encourage uniform participation, you will be able to reduce bias and uncertainty caused by low response rates.

• Think about timing.
• Always use a cover letter.
• Use announcements or invitations.
• Create a sense of urgency. 
• Send reminders.
•  Offer incentives for participating.

Conduct the survey, analyze the results, and prepare a report

When the planning is completed, the survey instrument is ready, and the promotion underway, the next step is to conduct the survey. If you are using an Internet-based survey, send out your e-mail notices, place the appropriate links on your Web site, or do both. If you are using the mail, send out the surveys. If you are using the phone, start making calls.

Avoid allowing too much time for people to respond. A longer response time encourages procrastination and leads to lower response rates. Requiring a shorter response time creates a sense of importance and urgency.

You can begin analyzing the results as soon as responses begin coming in. However, it is wise to avoid drawing any conclusions until at least a large percentage of your expected responses have been entered into the dataset.

Once the survey has closed, the responses are entered into the database, and the analysis is complete, it's helpful to prepare a written report. This is useful even if the survey and the knowledge gained are only going to be used internally.

For answers to your questions about the survey process, contact Nancy Cotton at 800.580.8272, extension 6235, or .