Businesses, educators, government employees, and the common people have a vested interest in gathering information. That's a survey: a way to gather information and learn from respondents. While surveys may look easy at first glance, they are actually much more difficult. See Step 1 for more information on how to create the best and most useful surveys to make life easier.
Step
Part 1 of 3: Designing a Survey
Step 1. Set the goal of your survey
In short, what do you want to gain from a survey? The question you ask needs to point to this important idea.
- For example, let's say you are a boss and you want to find out if your employees are satisfied. The questions in your survey, whether direct or indirect, should be aimed at the satisfaction of your employees. You can ask directly, "On a scale of 1 to 10, how satisfied are you with your job?" or you could form an indirect question, such as "True or False: I wake up every day feeling that my work has a purpose."
- After you have designed all of the survey questions, it may be a good idea to go over each question and ask yourself how it can help you achieve the survey's objectives. Any questions that do not provide you with important information about the purpose of the survey should be omitted.
Step 2. Take the time to think about how to ensure the most honest answer
If your goal is to find out if your employees are satisfied, you want honest answers. In fact, with surveys, you always want honest answers. But getting honest answers can be difficult if your employees feel they could lose something (respect, position, etc.) for being honest. Think about whether you need to find another way to get honest results. With employee satisfaction surveys, for example, you can give respondents the option to fill in anonymously.
Step 3. Decide on the best method to gather valuable information with surveys
Some options include telephone surveys, face-to-face interviews, mail surveys, and Internet questionnaires. Each survey method has advantages and disadvantages, which must be weighed against funds, available personnel and other considerations.
- In general, face-to-face interviews, while expensive and time-consuming, give the most representative results and the most detailed responses. On the other hand, online questionnaires sometimes result in significant bias, but are the cheapest and easiest type of survey to conduct.
- If you will rely on only one form of survey, such as an online questionnaire, consider surveying more people to address bias. For the cleanest results, you may have to carry out several different types of surveys.
Step 4. Think about how to ensure accuracy in your survey
Surveys involving one or two respondents will tell you something about each respondent, but do not provide enough accurate information about trends. To find out how many people to survey, you need two kinds of information:
- Population size. What population do you want to understand? If you want to understand satisfaction in your company, your population is the size of the company. If you want to learn about condom use in Uganda, your population is the size of Uganda, or about 35 million.
- The assurance that your results are accurate. Regarding survey accuracy, we talk about two ideas: the margin of error and the confidence interval. The margin of error is the degree of uncertainty in the survey results. Confidence interval is the degree of certainty that the survey accurately sampled the population.
Step 5. Based on the target population and desired level of accuracy, select your sample size
Once you have answered the question above: what is the population I am targeting? and how accurate the survey results do I need? You can start thinking about how many people you have to survey to get the results you want. In the table below, select your target population on the left, then select a margin of error to estimate how many surveys you will need. As a general rule, the more surveys you give, the lower your margin of error.
| Population | Margin of Error | Confidence Interval | ||||
|---|---|---|---|---|---|---|
| 10% | 5% | 1% | 90% | 95% | 99% | |
| 100 | 50 | 80 | 99 | 74 | 80 | 88 |
| 500 | 81 | 218 | 476 | 176 | 218 | 286 |
| 1.000 | 88 | 278 | 906 | 215 | 278 | 400 |
| 10.000 | 96 | 370 | 4.900 | 264 | 370 | 623 |
| 100.000 | 96 | 383 | 8.763 | 270 | 383 | 660 |
| 1.000.000+ | 97 | 384 | 9.513 | 271 | 384 | 664 |
Part 2 of 3: Asking the Right Questions
Step 1. Decide whether to use structured or unstructured questions, or a combination of both
How well do you know your respondents? Do you aim to gather information about ideas you already know or explore new ones? If you're gathering data about a known idea, you may want to rely on structured questions. If you're gathering really new ideas, you might want to lean on unstructured questions.
-
Structured questions ask a question and provide answer choices below it. Examples of structured questions are:
(1) "What is your favorite online activity?"
(a) Chat/IM
(b) Social Networks
(c) Knowledge Sharing/Forum
(d) Shopping/E-commerce
-
Unstructured questions removes the answer specified before the equation. Instead of leading respondents in a certain direction by giving them a choice of answers, unstructured questions encourage respondents to provide highly personalized answers. An example of an unstructured question is
(2) "Tell me about your first time signing in to the Apple Store."
Answer:
Step 2. Select some structured questions to get a bit more detail but still analyze
The drawback of structured questions is that they are usually not very specific. The disadvantage of unstructured questions is that responses are difficult to analyze and/or load in a spreadsheet. Include some structured questions. The existence of some structured questions will cover the shortcomings of each:
(3) "How would you describe your attitude towards paying for music? Choose all that are appropriate." (_) I never pay for music(_) By law, I pay for the music I listen to(_) I often download music illegally(_) I rarely download music illegally (_) I could be interested in paying for music if i get more (_) no one can entice me to pay for music(_) i feel sorry for musicians trying to get a decent paycheck(_) i have no concern for musicians trying to get decent pay
Step 3. Ask "Assessment" questions
This is part of a structured question. The goal is to answer how respondents would rate their experience on a scale. Your scale can be numeric or have more complex rubrics:
(4) "Ragunan Zoo is fun for both kids and adults." (a) Strongly Disagree(b) Disagree(c) Agree(d) Strongly Agree
Step 4. Ask a "Racing" question to get a list of sequential choices
Rating Questions are better than Rating questions for finding out what people think about a given topic. Examples of Ranking questions are as follows:
(5) "In the dots below, rate the brand you trust the most, '1' meaning most trusted and '5' least trusted." (a) _ McDonald(b) _ Google (c) _ Walmart(d) _ Costco(e) _ Apple
Step 5. When developing structured questions, include additional phrases at the end of each answer
It is helpful to include options such as "Other," "Not All," etc. in each of your answers. This option usually makes the response more accurate. Without these phrases, respondents who do not find an answer that suits them are forced to choose an inaccurate answer to complete the question.
Part 3 of 3: Distributing Surveys
Step 1. Find a way to distribute the survey
Once you've decided on the type of survey you're going to use, you'll need to think about how to give all of these questions to respondents.
- The Internet makes online questionnaires very easy to compose and send. Services like Google Forms, SurveyMonkey, and others offer free, easy-to-create surveys.
- If you're going to distribute a survey over the phone or want to carry out a face-to-face survey, be prepared to spend some money. The data you collect is usually more representative, but it comes at a price. You can hire a professional contractor to carry out the survey for you.
Step 2. Make the return of information as easy as possible
Free shipping for surveys by mail will increase the chances of the survey being returned. Distributing surveys at inappropriate times will hinder participation. Groups of respondents who are asked to participate after working hours or at the end of a busy day may give skewed information because they are tired and annoyed.
Step 3. Analyze the survey results
If the data isn't in a bundle at one location, now might be the time to collect it. Excel is a great tool for this. Use excel to create formulas, graphs, and analyze data. In short, find out what respondents are saying.
Step 4. Develop your observations and apply them
Now ask yourself why. Why are your employees dissatisfied, for example? The answer may already be in some of your answers. Otherwise, you might be able to create a new survey to help you answer this important question. Then, once you know why: My employees are dissatisfied because they don't get enough benefits, you can implement and repeat a new strategy.
