How it was? DataTalks # 3: Online Surveys

    On July 23, DataTalks # 3: online polls took place in Minsk. This time I was lucky to be among the organizers. And now I would like to share my impressions with you. Of these, the most important thing is that the topic of online polls is really in demand in the community. Both professional researchers and newcomers to the polls came to DataTalks # 3. Everyone was able to learn something useful for themselves. The meeting turned out to be very eventful: 3 reports from practicing speakers and a lot of informal communication among the participants.
    Under the cut you will find recordings of performances, as well as my comments on them.
    Happy viewing!





    How not to lose one’s head while reading user’s thoughts: secrets of conducting online surveys
    Svetlana Polovinkina, head of audience research at the Game Stream JLLC, Minsk Wargaming Development Center
    I spoke directly about the features of conducting and using the results of online surveys. They do not represent such accurate data as internal statistics, but they make it possible to understand why the user is behaving in one way or another, and also to identify his needs.
    It is believed that online polls are not representative. In my speech, I give an example of how the results of a user survey on a forum and a targeted study can differ (when an invitation to participate in a survey is sent to a sample of users with certain criteria). Surveys most often involve the most involved and loyal users to the product. They are the audience of specialized forums. Therefore, to make the survey representative, you need to study your audience and find a way to determine who is responsible. Combining behavioral data with answers, we get the opportunity to adjust the sample and understand the differences between users with different experiences. I’m convinced that the user’s behaviors regarding the product are the same socio-demographic characteristics,
    You can find my presentation here . It will be interesting to those who conduct online surveys and would like to increase their effectiveness. If you are just about to conduct your first survey, check out the article “8 Facts About Online Surveys That Every Researcher Must Remember”



    Needs and hidden objections: how to investigate them and when polls lie?
    Tamara Kulinkovich, co-founder of Sorokin and Kulinkovich Studio,
    Tamara talked about seven misconceptions that most often prevent researchers from conducting online surveys (and in my experience, customers are also subject to the same misconceptions). She demonstrated the main mistakes by inviting the listeners to become respondents themselves. We must pay tribute to Tamara for her humor. Filling out the questionnaire, the listeners were a little dumbfounded at the beginning, but then gradually came to understand how the respondents feel and what answers they give.
    Tamara also tried to predict the answers of the participants to the questionnaire. In this she was helped by knowledge in the field of psychology. Aggravation, social desirability, projection, the error of causal attribution, and other psychological effects often lead us astray. Despite the abundance of terms, Tamara in her speech explained clearly what the essence of these phenomena is. It turns out that people tend to exaggerate their deplorable state, often choose socially desirable answers, do not always talk about their real behavior, and also seek to justify themselves and judge others.
    The main conclusion is that you cannot blindly trust the respondents and be sure that we are asking the right questions.
    Since you did not fill out Tamara’s questionnaire, you may not be able to “fit into the shoes” of those present at DataTalks # 3 when watching the video, but definitely reconsider your approach to polls. Presentation of Tamara and several life hacking, how to prepare the questionnaire, you will find here .



    The main trends of online research: audience, mobile devices, gamification
    Aigul Mavletova, Deputy Dean for Science, Higher School of Economics
    Hague Aigul prepared a detailed overview of the main trends in online research in the world, explained why it is important to pay attention to mobile users, and shared her experience in optimizing profiles and engaging respondents in online surveys.
    All respondents to online surveys can be immediately divided into at least 2 groups, depending on which device they answer from - mobile or desktop. The features of the device’s interface with which the user answers questions affect the length of the questionnaire, the likelihood of completing the survey to the end, etc. For the results of the study to be representative, the representatives of each group should be included in the sample. Aigul told how to cope with this in world practice.
    Gamification is now a popular trend in all areas. Often it is used in online surveys. It can be expressed in the adjustment of the wording of the question towards the fan, the use of unusual tasks (you can read more about this in the Jon Puleston blog) or in a complete change in the approach to the formation of the questionnaire and the process of the survey: the respondent is informed about the rules and objectives of the study, tasks are described, feedback and incentives are given, passing levels are introduced, etc. Aigul is an adherent of the second option of gamification and talks about him in detail in his speech.
    In my opinion, gamification is just a way to engage a respondent in a survey. When conducting a survey, the most important thing is a qualitative result, since errors in the study lead to incorrect business decisions. If the use of gamification reduces the ability of the questionnaire to solve a research problem, then it is worth using the traditional approach. Aigul adheres to the same principle.
    You can find the presentation of Aigul here.. The information will be useful to those who are already conducting online surveys or are about to start. Aigul raises professional research questions, but the material itself is accessible and contains good examples.



    I hope the recordings of speeches and other materials from the last meeting were useful to you. I will be glad to answer questions and share my experience.
    Meanwhile, the next DataTalks is just around the corner. Follow the announcements in our group on Facebook !

    Let's research!

    Also popular now: