A questionnaire is a data collection technique in which the respondent answers a series of questions. It takes time and effort to develop a questionnaire that collects the data you want, but by taking a step-by-step approach to developing a questionnaire, you can devise an effective means of collecting data that answers your unique research question.
Steps
Part 1 of 3: Design Your Questionnaire
Step 1. Identify the objective of the questionnaire
What kind of information do you want to collect in the questionnaire? What is your main objective? Is a questionnaire the best way to collect this information?
- Come up with a research question. It can be one or more questions, but this should be the focus of your questionnaire.
- Develop one or more hypotheses that you want to test. The questions you include in the questionnaire should be geared toward testing these hypotheses systematically.
Step 2. Choose the question type (s)
You can include several possible types of questions in your questionnaire depending on the information you want to collect, and each has its own unique pros and cons. These are the types of questions that are commonly used in a questionnaire:
- Dichotomous question: This question will usually be a yes or no question, but it can also be a question to agree or disagree. It is the quickest and simplest question to analyze, but it is not a highly sensitive measure.
- Open questions: these questions allow the respondent to answer in their own words. They can be used to gain insight into the opinions of the respondent, but they can also be challenging when it comes to analyzing the data. It is advisable to use open-ended questions to address the "why".
- Multiple Choice Questions: These questions consist of three or more mutually exclusive categories and ask for one or more answers. Multiple-choice questions allow easy data analysis, but may not give the respondent the answer they want
- Sort order scale (or ordinal) questions: Questions of this type ask the respondent to rank items or choose items from a set in a particular order. For example, you could ask respondents to rank five things from least to greatest importance. Questions of this type force discrimination between alternatives, but do not address why the respondent made these discriminations.
- Rating scale questions: These questions allow the respondent to rate a particular issue based on a given dimension. For example, you can provide a scale that offers the same number of positive and negative choices ranging from "strongly agree" to "strongly disagree." While these are very flexible questions, they don't answer "why" either.
Step 3. Develop questions for your quiz
The questions you develop for the quiz should be clear, concise, and direct, which will ensure that you get the best possible responses from your respondents.
- Write succinct and simple questions. You should not write complex statements or use technical jargon. This will only confuse your respondents and cause you to get wrong answers.
- You should only ask one question at a time. In this way, you help avoid confusion.
-
Be careful when asking for private or "sensitive" information. It can be something as simple as your age or weight or as complex as your sexual history.
If you ask questions like these, you will generally need to anonymize or encrypt the demographic data you collect
- Determine if you will include an answer such as "I don't know" or "Does not apply to me." These can give your respondents a way not to answer certain questions, but giving them these options can also cause missing data, and this can be problematic during data analysis.
- Put the most important questions at the beginning of the questionnaire. This can help you gather important data even if you perceive that respondents may be starting to get distracted by the time the questionnaire is over.
Step 4. Restrict the length of the questionnaire
You should keep the questionnaire as short as possible. Most people are likely to respond to a shorter questionnaire, so you should be careful to keep it as concise as possible while still collecting the necessary information. If you can do a questionnaire that only requires 5 questions, do it.
- Only include questions that are directly useful to your research question. A questionnaire is not an opportunity to collect information of all kinds about your respondents.
- Don't ask redundant questions. This will frustrate the people taking your quiz.
Step 5. Identify your target demographic
Do you want to focus on a certain group of people in your questionnaire? If so, it is best to determine this before you start distributing the questionnaire.
- Consider whether you want the questionnaire to collect information from men and women. There are some studies that only survey one sex.
-
Determine if you want the survey to collect information from children and adults. There are many surveys that focus solely on certain age ranges for which the questions apply.
Consider including an age range in your target demographic. For example, you might consider those ages 18-29 as young adults, those ages 30-54 as adults, and those 55 and over as older adults. Providing an age range will help you get more respondents than if you limit yourself to a specific age
- Consider what else would make a person a target for your quiz. Do I need to drive a car? Should you have health insurance? Should you have a child under the age of 3? Be careful to make this very clear before distributing the questionnaire.
Step 6. Make sure you can protect privacy
Before you start writing your survey, make a plan to protect the privacy of your respondents. This constitutes a very important part of many research projects.
- Consider an anonymous questionnaire. You may not want to ask for names on the questionnaire. This is a step you can take to avoid privacy, but it is often possible to determine a respondent's identity using other demographic information (such as age, physical traits, or zip code).
- Consider de-identifying the identity of your respondents. Assign a unique number or word to each questionnaire (and thus each respondent), and refer to them only by that new identifier. Destroy personal information that could be used to determine identity.
- Do not forget that it is not necessary to collect a lot of demographic information to be able to identify someone. People may be reluctant to provide this information and therefore you may get more respondents if you ask less demographic questions (if possible for your questionnaire).
- Be careful to destroy all identifying information after you finish your study.
Part 2 of 3: Write the quiz
Step 1. Introduce yourself
In the presentation, you must explain who you are and what your credentials are. Make it clear if you will be working alone or as part of a team. Include the name of the academic institution or company for which you are collecting data. These are some examples:
- My name is Juan Pérez and I am one of the creators of this questionnaire. I belong to the Department of Psychology at University X, where I focus on the development of knowledge in babies.
- I am Carla Suárez, a third year undergraduate student at Y University. This questionnaire is part of my final exam in Statistics.
- My name is Sergio Jiménez and I am a marketing analyst for La Mejor Empresa. I have been working on developing questionnaires to determine attitudes around drug use in Canada for several years.
Step 2. Explain the purpose of the questionnaire
Many people will not answer a questionnaire if they do not understand what its purpose is. A lengthy explanation is not needed but a few concise sentences will suffice. These are some examples:
- I'm collecting data on attitudes around gun control. This information will be collected for my Anthropology 101 class at Z University.
- This quiz will ask you 15 questions about your eating and exercise habits. We try to establish a correlation between healthy eating, exercise frequency, and cancer incidence in older adults.
- This questionnaire will ask you about your recent experiences with international air travel. There will be three question sections that will ask you to relate your recent trips and your views on them, as well as your travel plans for the future. We seek to understand how a person's opinion about air travel has an impact on their future plans.
Step 3. Reveal what will happen to the data you collect
Do you collect this data for a class project or for a publication? Will this data be used for market research? You may need to pay attention to different requirements before distributing the survey, although this will depend on what you intend to do with the data you collect from the questionnaire.
- Note that if you are collecting information for a university or for publishing, you may need to check with the international review board (IRB) or its equivalent at your institution for permission before you begin.. Most research universities have a dedicated IRB staff, and you can usually find their information on the school's website.
- Do not forget that the best is transparency. It is important that you be honest about what will happen to the data you collect.
- Include an informed consent form if necessary. Please note that while confidentiality cannot be guaranteed, you will make every reasonable attempt to ensure that your information is protected.
Step 4. Estimate how long the questionnaire will take
Before someone sits down to take your questionnaire, it may help to know if it will take 10 minutes or 2 hours. You are more likely to get a greater number of completed questionnaires at the end if you provide this information at the beginning of the questionnaire.
- Take the time to conduct the survey. Then consider that some people will take longer than you and some will take less time.
- Provides a time range instead of a specific time. For example, it is better to say that a survey will take between 15 and 30 minutes than to say that it will take 15 minutes and that some respondents give up halfway.
- Use it as a reason to keep your survey concise. You will feel much better if you ask people to take a 20 minute survey than if you ask them to take a 3 hour survey.
Step 5. Describe the incentives that might be available
An incentive is anything you can offer as a reward at the end of the quiz. Incentives can be of many types: they can be monetary, desired prizes, gift certificates, candies, etc. Offering incentives has both pros and cons.
- Incentives can attract respondents of the wrong type. You shouldn't incorporate responses from people rushing through the quiz just to get the reward at the end. This is a danger of offering an incentive.
- Incentives can encourage people who might not have responded to your survey without a reward to do so. In this situation, incentives can help you reach your target number of respondents.
- Consider the strategy SurveyMonkey uses. Instead of paying respondents directly to take their surveys, they offer 50 cents to the charity of their choice when a respondent fills out a survey. They consider that this reduces the chances that a respondent will fill out a questionnaire out of sheer selfishness.
- Consider entering each respondent in a drawing for a prize if they complete the questionnaire. You could offer a $ 25 gift card to a restaurant or a new iPod or a movie ticket. This makes it less tempting to answer the questionnaire just for the incentive but still offers the possibility of a nice reward.
Step 6. Make sure the questionnaire looks professional
You want people to trust you as a data collector, so your questionnaire needs to look professional.
- Always check. Check for spelling, grammar, and punctuation errors.
- Include a title. This is a good way for respondents to understand the focus of the survey as quickly as possible.
- Thank the respondents. Thank them for taking the time and effort to complete your survey.
Part 3 of 3: Distribute the questionnaire
Step 1. Conduct a pilot study
Ask some people you know to take your questionnaire (you will not include them in any results that come out of this questionnaire) and be prepared to review it if necessary. Plan to include between 5 and 10 people in the pilot test of your questionnaire. Ask them the following questions to get their feedback on your quiz:
- Was the questionnaire easy to understand? Did any questions confuse you?
- Was the questionnaire easy to access? (This is particularly important if your questionnaire is online).
- Do you think the questionnaire was worth your time?
- Were you comfortable with answering the questions that were asked of you?
- Would you make any improvements to the questionnaire?
Step 2. Distribute the questionnaire
You need to determine the best way to distribute the questionnaire. You can distribute questionnaires in several common ways:
- Use an online website like SurveyMonkey.com. On this website, you can write your own questionnaire using its survey builder, plus it gives you additional options, such as buying a target audience and using its analytics to analyze your data.
- Consider using the mail. If you are mailing the survey, you should always be careful to include a self-addressed and stamped envelope so that respondents can easily send their responses back to you. Take care that the questionnaire fits inside a standard business envelope.
- Conduct in-person interviews. This can be a good way to make sure you are reaching your target demographic, and it can reduce missing information from your questionnaires because it is more difficult for a respondent to avoid answering a question if you ask them directly.
- Try to use the phone. This may be a more time-efficient way to collect data, but it can be difficult to get people to respond to questionnaires over the phone.
Step 3. Include a deadline
Ask respondents to complete and return the questionnaire by a certain date at the latest to ensure you have enough time to analyze the results.
- Make it a reasonable time. It should be more than enough to give respondents a maximum of 2 weeks to respond. If it's more than this, you run the risk of respondents forgetting about your questionnaire.
- Consider providing a reminder. A week before the deadline is a good time to provide a polite reminder to have the questionnaire returned to you. Include a replacement questionnaire in case the respondent misses it.