In this second part of the introduction to RADAR questionnaire lectures, we're going to focus a little bit more on what concretely the RADAR questionnaire looks like from the inside. The first part of all of the individual RADAR questionnaires is the information panel. On it you'll see first the questionnaire title all the way at the top. You'll then have the module title. The module title will also give you the initial of this section. That will help you identify where that question is within the questionnaire. For example, in this image, we have the woman's information panel, the initials are WM. If we refer to any question that starts with WM, we know that it'll be within this section of the questionnaire. The first section that you can actually enter information in is more administrative information and the person's ID. What in the street of area there in meaning what district what village, depending on the specifics of your survey. The name and the number of the cluster, the name of the cluster, the household number, and then the woman's line number referring to the woman's line number in the household, this thing done in the household questionnaire. The next big section is the result of consent or assent. If the woman is married and over 15, then she can give consent. If she is between 15 and 17 and unmarried, she can give assent but in order to do so, you must first get parental permission from her guardian, and then you would note the assent. That is a section where you would do that. You then have a place to record the visit interview results. Whether or not it was possible to contact that woman, that's helpful to track callbacks and to make sure that everyone was contacted. Finally, you have the results of whether the questionnaire was completed or not and the reasons why it was not if it was not completed. The household listing is a big part of the household questionnaire. It's one of the first things that do when you enter into the household. What it is is a critical mechanism to identify all of the eligible persons for the individual questionnaires. In the ODK software, the individual questionnaires will automatically be generated from the household does things. ODK will actually recognize, this is a woman between 15 and 49 or this is a child between zero and four and therefore generate the appropriate questionnaire for your interviewer to fill out. The total number of household members is also useful for computing some indicators, for example, those around water sanitation, vector control, and the wealth index. It's actually used as a metric itself. The clear and consistent definition of households is required to identify eligible respondents. For the RADAR questionnaire, there's a very specific definition of who qualifies as a household member and who does not. In the RADAR questionnaire, visitors are included. People who have slept there the night before are included. Making sure that interviewers are clear on this definition ensures that the data is also representative of what it's supposed to be. The household listing should never, ever be removed. I've mentioned that we can add some modules, remove some modules, but the household listing it's not something we can remove because the RADAR questionnaire has the minimum set of questions required in order to do the calculation for many of our indicators. But you could choose to add more depending on your service needs. You just can never remove the household listening. This is what the actual household listing looks like. As we've mentioned, it's going from left to right, the 1st question is the line number which is automatically just linearly one, and so on so forth. The 1st person will always be the head of households, whether or not they're actually the person that you are interviewing at the time. Even if you're interviewing a representative, you would still list the head of household as the 1st person. Then everyone after that would be listed in terms of their relationship to the head of household, so that's HL3. Then there's the gender, date of birth, age, whether or not they live there, whether or not they slept there last night. Then the next three questions are qualifying questions as to whether or not this individual is eligible for either the woman's questionnaire, men's questionnaire, or child's questionnaire. In the case of children, you would then ask whether or not the child's natural mother is still alive, and whether she lives there and if not, then who the primary caretaker of this child would be in order to determine who is the person to interview for a child 0-4. Essentially, the household listing really establishes who's in the household and also who is eligible for additional questionnaires. To review a little bit, also how the questions themselves are structured. All the way on the left you'll start with a question number. We've mentioned this elsewhere, but it's just to really put some more emphasis on this. Changing the question numbers is never a good idea. The question numbers are used in the analysis scripts and changing them will cause a lot of headaches down the road. Even if you are adding new questions, you should create new question numbers and not reuse existing ones. The question numbers are fixed in the RADAR questionnaire and they should not be modified, and they'll be completely on the left-hand side. Then you would have a question. The first line of the question is the question itself. Then depending on how the question is structured, you might have a couple of probes. If the respondent doesn't answer spontaneously, there's a couple of things that you can probe to get more detailed information from the respondent. All the way at the bottom, there are some instructions. Instructions are things that are important for the data collector to see but they would not actually be saying it out loud. It's reminders and for them to help understand what's being said, and a way basically just for them to have more information as they deliver the questionnaire. The response options are in the next box, moving towards the right. The response option will have both a number and typically a label to it, so actual words to know what those numbers mean. In this case, it's a yes-no, which is also a 1, 2 answer options. All the way on the right, you have some skip patterns. If you're using the paper version of the questionnaire, then the interviewers would be able to see what is the next question they should ask based on what the respondent answered to this particular question. In this case, if the person has never given birth, then they can skip the whole fertility model. They can skip to FP1. As I've mentioned previously, the RADAR questionnaire has many different parts of it that can be adapted to specific contexts or specific surveys. In the case of things like facility types, health worker categories, local foods, local drugs, sometimes the name of cell phone providers or mobile banking options, all of those are things that you would really want to make sure that you adapt into a way that the respondents will be able to answer the question in a way that captures the information that you're actually trying to get from them. In the RADAR questionnaire itself, all of the things that you would need to adapt to local contexts are highlighted in red so that it stands out and you know to make those modifications before you actually go to the field. What we've been seeing in this lecture so far is the paper version of the questionnaire or the electronic version of the paper version. But the RADAR questionnaire is also available in electronic data collection format, CAPI. The CAPI version is developed for ODK so it's using the Open Data Kit software and it includes the XLS form that you can just upload and deploy to your ODK app. We strongly recommend using electronic data collection there's many, many reasons for that. Part of it is ease of use, part of it is data quality. Even though we think it's very important that training involve using the paper-based questionnaire and people really understanding how the questionnaire works, which is done more successfully on a paper-based questionnaire than electronically. For the actual data collection, we really recommend the electronic data collection approach. The RADAR questionnaire is available out of the box in both English and French. It's also partially available in Swahili. If the commonly spoken language in your study area is not either French, English, or Swahili, then you will need to translate and back translate the tools. Well, we'd recommend from English to your language and then translate it back from that language into English to make sure that comprehension doesn't get lost because of translation. If the commonly spoken language or languages in your study area are not commonly read, it might be very difficult to find interviewers who can read a translated questionnaire. So in that case, you would need to translate the questionnaire to the commonly read language, and then you train your interviewers in both languages. We go over this in more detail in a number of other lectures. There are a few considerations that you need to keep in mind regarding the questionnaires included in the RADAR questionnaire. One of which is that there are some potentially sensitive questions. We do ask questions about family planning, about contraception. There's also some questions about work and decision-making, including questions on women's empowerment and attitudes towards domestic violence. If you do decide to include these sensitive questions, you should provide appropriate training and skills and suitable introduction for these modules to your interviewers. So make sure that they know how to be delicate about these topics. Then you also need to make sure that they know to ensure privacy before they bring up these sensitive topics, and that they are very conscious of how important keeping confidentiality is. It is generally, but particularly about these more sensitive questions, and you want to make sure that they fully understand how important protection of participant's data is. Then this final point, if you're doing this when you're using a women's questionnaire and you're going to ask these sensitive questions, depending on the setting, it might be appropriate to hire only female interviewers for a number of reasons. In some cases, it might just be more comfortable for women to speak about these topics from one woman to another. Then in other cases, it might be actually culturally very inappropriate for a man and woman who are not married to be having a private conversation in a private area. It just might not be culturally feasible. So for that reason, we recommend prioritizing hiring female interviewers and if possible, hiring exclusively female interviewers, again if you're planning to include these sensitive questions. To end the lecture, we're just going to take a look at some reasons why you should consider using the RADAR questionnaires. There's very little, arguably maybe even no need to adapt the questionnaires. The core RMNCAH and nutrition indicators are already included in the questionnaire, as it is right out of the box. Then the modular frame allows you to adapt the questionnaire to your project needs if you do want to look at some specific topics. It's a fairly straightforward process. Your survey implementation process will be greatly simplified because you have access to the entire toolkit going from the training materials, the manuals, the CAPI, the analysis tools, etc. It all comes ready to use and easily adaptable. It's all part of a coherent package of tools. Again, it's by design compatible with the Lives Saved Tool. So if you then want to use your coverage data to model the impact on mortality, you can do so by using the Lives Saved Tool. Finally, it allows for linking data and for analysis. So you could use care-seeking from household survey and from health facility assessment surveys, and link those data to perform some analyses that way. In some, the RADAR questionnaire is part of the broader RADAR toolkit, which is a comprehensive set of tools that allows you to do a relatively quick, relatively easy, but high-quality household survey. In the next modules, we're going to go into a little more detail about the other tools that are contained within this toolkit.
