Steps for creating a survey
1. Defining the objectives of the survey
"Knowledge is power", Francis Bacon said at the time. Nowadays, we hear more often that "information is power". For collecting and organizing information, an online survey is a perfect tool. However, compiling this requires serious work, especially in terms of preparation and the compilation and development of questions.
Users of the site have already asked for our help in preparing hundreds of surveys. I dare to say quite boldly that where the preparations were not properly completed, the person preparing the survey always ran into serious problems.
Frequent error if the purpose of the research or survey is not defined with complete precision. This determines everything. For example, how big a sample should you work with. Who should be the interviewees. Is the age group, place of residence, or even the ethnicity of the applicants important?
A good film director who knows exactly what the film's message is, and according to which the director constructs everything in it according to this principle, knows what purpose each scene serves. That way, if you end up cutting the 3.5 hour game time, you'll know which scenes are key, whether it's moving the plot forward or developing a character. You won't cut them out. On the other hand, many, even brilliantly successful, parts have to be removed in order for the film to be of acceptable length. This problem often occurs with surveys as well. Several articles will touch on the issue of length. We must know exactly what we are interested in, because if the length of the survey exceeds the tolerance of the people filling it in, then it doesn't matter how well it is structured or how useful it is. It won't hit the target. Although this is not a spectacular part of the job, every minute spent on it will pay off with interest!
2. Defining the main questions
For the sake of clarity, let's take a concrete example here:
A fast food chain, Sültkirály, prepares a survey. For example, the questions should focus on:
- Find out the frequency indicators related to going to the restaurant.
- Get to know the image of Sültkirály in the minds of customers and why the number of visitors to the restaurants has fallen.
- You will know what you need to do to increase the number of guests again.
Either at this point or when setting up hypotheses, we may be faced with the fact that we do not have sufficient knowledge about the subject of our research. If we only take the previous very simple example as a basis, even then, it will hinder the collection of effective answers if the creator of the survey does not know that his restaurant is being boycotted because of the inhumane treatment of the workers. This will not be revealed with closed questions measuring general satisfaction.
It can help a lot if we conduct an in-depth interview with some members of the target group. In this way, we can expand the initial questions even on the fly, adapting to the answers. But similar, results of previous research and surveys should also be known and taken into account.
3. Creating hypotheses
A hypothesis is a preliminary assumption. A conditional statement that we want to prove or disprove during the research. If we have defined the main questions, then we must have some idea of what kind of answers and results we will get.
Staying with our previous example, hypotheses can be, for example:
- Sültkirály's entire image is aimed at young people. Mostly the school age group. Since the target group of all advertising campaigns has been this age group, according to our hypothesis, the advertising activities of the last two years were not effective enough.
- The range of restaurants does not meet the needs of the guests.
- There are not enough or sufficiently attractive promotions for young people.
- They are not trying to reach the target audience on the right channel.
Correctly established hypotheses greatly facilitate and speed up the research process. Thus, if the creator of the survey studies the media consumption habits of the age group during his preliminary research, he will immediately see that it will not be worth dissecting TV viewing habits at length. This age group already uses other channels. However, it is necessary to find out exactly where it is best to reach them. I cannot stress enough that without preliminary research and thorough knowledge of the topic, the survey will only grope blindly.
4. Determination of research subjects
The bigger and better sample we work with, the more certain the data. The big advantage of the online survey is that the system collects the data for us. Based on the hypotheses, it is possible to imagine what kind of results we will get. Do we need socio-demographic data (e.g. age, gender, education level, place of residence, income) and who should we ask? The size of the sample is often given. For example, if you are interested in the satisfaction of the workers of a factory in relation to their working conditions. In this case, it is possible to say exactly how many people we have in our sample.
The situation is much more difficult if, for example, we research how many young women have been sexually harassed. How big is the sample that covers this? If we completed the previous steps correctly, we should already know what we are interested in by staying with this example. Anyone from the entire country, or just one county or city? Just women in one industry, or anyone? Does all form of bullying matter, or only that which escalates into an act? Does childhood sexual abuse also count, or do we only want data on adulthood? Exactly how old are women classified as young adults? If we know exactly what we are interested in, the preliminary research work should reveal how many women on average fall victim to sexual harassment during their young adult years, according to other surveys. This already helps us to determine how big a sample (how many people to ask) will be needed.
Who we will ask is also very important for choosing the language. If only specialists will fill out the survey, it is a logical step to use the specialist jargon they know. On the other hand, if, for example, they are parents of disabled children, then they may include several graduate university instructors and eight with a primary school education. In such cases, it is necessary to take into account that we use a text that is understandable and simple for everyone.
5. Finding a way to the subjects of the research
We have two tasks here. On the one hand, finding a way in a very physical sense. How to get the e-mail address of the people filling in, or how to deliver the survey to them. And the second is more figuratively, to find a way to these people. Simply having an email address is not enough if we do not succeed in persuading the potential filler to commit to it, take it seriously and do it all the way through.
Both are easier for internal research. In the case of external research, it helps if we know a place where we can meet potential fillers. If, for example, a beauty salon or a restaurant or a private doctor's office wants to know the opinion of its guests/patients, how satisfied they are with the service, they have a very easy task.
- We ask for the email addresses of our potential respondents, and we use the list compiled in this way to send out the survey.
- We put the link to the survey on the website of the given service provider and ask the guests to fill it out.
- In places where there is a waiting time, it is worth installing a tablet in the shop. Thus, guests/clients can fill out the survey while waiting.
- It is also possible to provide a one-time-use password (token) printed for the given survey on a paper basis, which also includes the title (shortened link) of the survey, which you can fill out in your browser.
If it is not possible to personally find the potential fillers, it can be a great help to find various Facebook groups, internet forums, social networking sites related to our topic. If we conduct a study on the candy consumption habits of diabetics, we can use this to search for all diabetes groups created on social media. There is a good chance that if the admin receives a letter from the creator of the survey, with the request to share the link of the survey in the group, he will say yes. Many professionals who chose us were able to successfully deliver their surveys to the desired target group in this way. We can really only recommend this method.
If you already have the concrete path to the possible future fillers, then the second task can come. How do we get them to actually fill it out? It is often enough that they are emotionally affected by the survey and feel that filling it out is useful. For example, the given research deals with a problem that also affects them. They often offer to send the results after the results have been compiled and evaluated to whichever breeder requires them. If the result is really interesting or important to the person completing it, this is often enough motivation.
However, a topic of no interest to us, or doing "work for free" for a large company, is less inspiring. A survey that can be filled out for money can be a solution to this, but it only reaches a very narrow group. It is a proven method to advertise a sweepstakes among those who complete it.
We definitely have to keep in mind that people are getting so many similar requests every day that their stimulus threshold has become very high. It is worth thinking about exactly how we draw their attention to the survey and how we motivate them to fill it in.
A very important piece of advice is that an honest and fair opening is essential. If we drag the person filling out the survey in any way, it either means that he doesn't finish the survey, or if he continues angrily, the answers he gives will be greatly distorted. A common example of this is when a survey emphasizes that "filling in only takes 5 minutes", and then contains 120 questions.
6. A complete example
If anything was not clear in the theoretical part above, I hope this step-by-step example will help you put everything in place.
An animal shelter sadly finds that they have to euthanize most of their old animals. Adopters usually insist on getting a puppy or a very young animal.
1. Setting the aim/s of the survey
Find a solution to the problem of adopting old animals.
2. Defining the main questions
- To explore the exact reasons why old animals are rejected.
- Filtering out who are those who would like to do it, but there is an obstacle in their way that can be overcome.
- What motivation is enough for someone to choose an injured or older animal?
3. Creating hypotheses
Before creating the hypotheses, the workers talk to the visitors. On the basis of these discussions, the following assumptions are made:
- Many families with small children adopt. They cling to the young animal so that it can be a part of the family for as long as possible.
- Lonely elderly people often take in a new animal. They also don't want to be alone.
- In the above two cases, it is not worth using violence. However, it can help couples, families, and people who are socially sensitive to learn about the dog's past. For example, that a dear old aunt was his only companion. Or that his former owner regularly beat him and starved him.
- Adoptors often come with very specific goals. It is necessary to reach those who want to help such an animal, but do not know about the possibility.
4. Determination of research subjects
- All interested persons and adopters visiting the shelter.
- Involvement of as large a sample as possible with the participation of people who have never been to an animal shelter.
5. Finding a way to the subjects of the research
Since the animal shelter is not funded by money, there can be no question of prize games or paid fillings. They put a link on their website and everyone who goes there gets a flyer inviting them to participate.
In order to reach those who don't go to the shelter, they make a Facebook post in which everyone who loves animals is asked to fill it out. This can be shared by all workers (of course on a voluntary basis) and by the shelter itself. In addition, they are looking for several groups that deal with dogs or animal rescue. The post will also be shared in these groups.