[SOUND] Welcome to Introduction to User Experience Design. Today, we will discuss the survey technique in detail. We will discuss its goal, the type of data we can collect during this user study session, and how we can use the data. We will also cover some advantages and disadvantages associated with it. A survey is quite simply a set of questions. For the purpose of this course, we will use the term questionnaire and survey interchangeably. The goal of surveys is to get the user's opinions. Most often, we get them to provide a self report about their behavior, perceptions, attitudes and feelings. We can also get their opinion about someone else. Surveys can take place in the field or the lab. We have all had the experience of going about our business, only to have a person walk up to us and ask if they can ask us a couple of questions. They might even come to our home or we might get someone calling us on our cellphone. Surveys require little direct interaction, the survey may be physical where someone hands you a sheet of paper, but now it's more common to have them be completely digital. Either you step up to a kiosk, or one opens up automatically when you browse a website. At this point, I want to remind you that we already had a lesson on how to come up with a plan for interacting with users, and the goals for the material that you want to administer. The quality of the data will be directly related to how well you prepare. Using surveys, we can collect both quantitative and qualitative data. Survey questions can provide us with quantitative data that we are able to aggregate across users. These are closed ended questions. In the dichotomous response, we ask users to choose from two options, yes or no. We can also ask them to provide a Likert rating. We can inquire about how much they like their current smartphone where one is not at all, and five is very much. We can also give users a list of items and have them rank them from the most to the least preferred. Unordered responses simply present the user with questions that don't have a logical order. For example, demographic questionnaires such as sex or education level, or lists where the user just checks off all that apply. For example, we may want to know what different brands of sneakers they've bought in the past five years. Survey questions can also provide us with qualitative data. This is usually in the form of open ended questions. Our goal is to get brief answers from the user about their opinions, preferences, attitudes in a couple of sentences. All of the techniques we will discuss have advantages and shortcomings. Among the advantages of a survey is that we can collect data from a number of individuals simultaneously. Also we can analyze the data pretty quickly. All the techniques we discussed also have advantages and shortcomings. I just mentioned that the major advantages of surveys are that they're efficient, they provide us an efficient way of collecting data, and that they're easy to analyze. However, this also leads to perhaps its biggest shortcoming. That is, that the information we collect is superficial. It doesn't really provide us with insight into the problem area or what we want to design. Other shortcomings include recall bias, social desirability, and sample bias. First, when we ask users to tell us about things that happened in the past, there is a chance that their recollections are incorrect or inaccurate. Second, when we ask users about their opinions, they may want to respond in a way that pleases the researcher. Or that is in accordance with their cultural norms, but that may not reflect their actual attitudes. Finally we only have data from the users in our study that chose to participate. And these people may not represent the users we couldn't reach or that didn't want to participate. Survey data can provide impetus for future user studies. The analysis can highlight interesting results that we may want to delve into more deeply. As you know, a survey is part of the iterative design process. As such its findings can lead us to use other techniques. For examples, let's say that the survey indicated that there are these three prominent ways the user completes a task, and that one of these is favored by males, while another one is favored by females. We can then organize a focus group to try to understand what is driving these preferences. In the previous slide, I provided you with an example of going from the least to the most direct interaction with users. We could just as easily decide that in fact, following the survey we would like to do a naturalistic observation where we observe males and we observe females engaging in the task I described earlier. As a good user experience designer, we need to use a technique that provides us with the information we actually need. So you get to choose what you want to do. In this lesson, we reviewed surveys. I look forward to seeing you in our next lesson, where we will review focus groups. [MUSIC]
