Hello. Welcome to the video about surveys. In a previous video, you were introduced to the different research strategies that are relevant in urban research. In this video, we will discuss the survey strategy in more detail. Let's start with a short recap of what we already know about surveys. So surveys are a suitable strategy when your objective is to describe a problem or situation or to test a certain theory. A survey requires a large number of variables and a large number of units. It enables you to generalize your outcome to a broader context, which contributes to the external validity of your research. But, when is it relevant to use this strategy and how do you actually develop this type of research? You will know all about it by the end of this video. Looking at the three dilemmas presented earlier, their survey strategy becomes relevant when you choose a combination of breadth, quantitative data and empirical research. This means that you want to look at the problem on a large scale, you want to be able to generalize and you require external validity. If you have a testing research question, the survey is also applicable even though it might not necessarily have the breadth but goes more in depth. Let's explore an example. In Hannah's first testimonial, she introduced you to the case of bicycle use for commuting in the Dutch city of Rotterdam. Because the bicycle sharing in Rotterdam is lower than in other large cities in the Netherlands, she wanted to explain the encouraging and discouraging factors that determine this. She chose a survey as her strategy because she knew to answer these questions, she needed to conduct empirical research that would result in quantitative data. In this case, Hannah also wanted to see whether there were any other aspects influencing bicycle use in Rotterdam which was not yet in the literature. That's why she complimented her survey research with several interviews with experts. She will explain more about her choices in the second testimonial. In other cases, like when your research question requires qualitative data collection or desk research to be able to answer the research question, the survey is not applicable. Indeed, answer A was correct. The research question aims to describe a phenomenon through a large number of variables which are still largely unknown at this point. For that reason, it is best suited to answer using the survey as a strategy. The other questions are more suitable to be answered through qualitative methods. Well, what is a survey actually? It is a strategy using questionnaires to collect data and requires quantitative analysis methods. These are seven characteristics to a survey. Let's have a closer look. The first characteristic is that you both require and are able to reach a large number of research units. These are often populations of a few hundreds or even thousands of research units. An example are the residents within a specific city or organizations within a region. This results in extensive data collection. The use of efficient methods like questionnaires are necessary to quickly gather data from a large number of research units. It therefore focuses more on breadth than depth as an important objective of the research findings. Some depth may also be possible with a testing question, as said before. The findings are based on survey data collected from a sample which has to be representative of a whole population. Sampling is required because you can not reach the entire population. So you need to use a random sample as you need a representative sample to avoid any bias. Also important is that the larger the sample the more reliable and more precisely analysis is. But you will learn more about sampling in the videos on data collection. The survey strategy enables you to analyze relationships between different variables of the research. Therefore, it is important to identify these variables clearly before the data collection. Remotes, closed data generation enables you to get the data necessary for this analysis. As data is collected in a well-structured way with fixed and closed questions. This ensures a strong reliability. In the end, the survey will produce quantitative data and require quantitative analysis. The data needs to be quantifiable for a systematic and structured analysis. For example by statistical programs such as SPSS. Indeed, the right answer is that the small number of research units does not fit with the survey. The survey needs a large number of units to enable analysis, and an important rule for surveys is that the bigger the sample the higher the reliability and accuracy of the results. Hopefully, this has given you a clearer view on the survey strategy. Now I will give you a short introduction to the different variants of surveys. First, we have the cross-sectional research, which is done through data collection at one moment in time from one and the same group, like when a sample of commuters in Rotterdam are asked about their bicycle use. The second is a panel research, which means that you collect data at various moment but within one group. In the case of Hannah's bicycle research, she chose this cross-sectional research. However, it might also have been interesting to conduct fieldwork at multiple moments. The season might make a difference for many commuters in their choice to ride a bicycle to work and therefore, a panel research would have been interesting. Then there's also a third option. The time series research, which is based on data collection at various moments in time, with varying samples, which means you have a different sample of respondents every time. This can be used to discover trends, like for example, our commuters now more inclined to use the bicycle this year compared to previous years. The choice of variants of the survey strategy should always be based on the research question and objective, but it's crucial to understand the opportunities of a specific strategies compared to its limitations. Advantages of surveys are the scope; as a wide overview can be given and a general results in valid statements. You also have the possibility to determine relationships between variables, like for example, which social-economic factors mostly affected choice of bicycle for commuting. Also, there's much literature available like methodological handbooks that you can use to check your research steps. Crucial limitations however, are that the research strategy is limited in the depth of the knowledge you can realize. The results are limited to the variables and the relationships between them. Also, a thorough preparation is necessary. You need to become a true theoretical expert to design the best possible questionnaire. Lastly, there's less flexibility during the data collection. The moment you start collecting data, the questionnaire cannot be changed anymore. So that is your point of no return. In the next week when we discuss data collection, Dorcas will explain how to make a proper questionnaire. For now, this is all you need to know about the survey as a research strategy. In the next lecture, we'll continue with another strategy namely The Experiment. Thanks for watching.
