WEBVTT 00:03:20.570 --> 00:03:24.000 I'm going to go ahead and get started. 00:03:24.000 --> 00:03:25.307 Can people hear me? 00:03:27.359 --> 00:03:29.000 Yes. 00:03:29.000 --> 00:03:36.000 Well, Hello, everybody. I'm Dr. Maria Lisa Christie at the Center of International Research, Education and Development at Virginia Tech. 00:03:36.000 --> 00:03:46.000 Welcome to our Global Development Discussion Series. Thanks to the Continuing and Professional Education for co-sponsorship this year. 00:03:46.000 --> 00:03:52.261 We're so happy you could join for today's discussion featuring Dr. Cheryl Doss as guest speaker. 00:03:52.261 --> 00:04:00.298 The title of her talk is "Does Women's empowerment mediate household food insecurity in the event of shocks: Evidence from Ethiopia." 00:04:00.298 --> 00:04:10.992 I am very excited to have her here, as I have been following her work for 20 years and never met her until our e-meeting just a few moments ago. 00:04:10.992 --> 00:04:17.219 So, and I also want to thank everyone for attending the webinar today and anyone who helped us get the word out about it. 00:04:17.219 --> 00:04:27.399 As of this morning, we had over 80 people registered for today's event. We'll see how many end up joining as we get rolling here. 00:04:27.399 --> 00:04:37.737 But that was from institutions in 12 countries in Bangladesh, Benin, Canada, China, Ecuador, Ethiopia, Germany, India, Kenya, Liberia, Switzerland, and Uganda. 00:04:39.532 --> 00:04:54.215 And besides students, faculty, and staff from Virginia Tech and Virginia Tech's lifelong Learning Institute, we have registrants from other US Universities, Emory and Henry, UC Davis, University of Florida, University of Illinois, and Cornell. 00:04:55.856 --> 00:05:15.045 We also have, no surprise, many CGIAR folks, so including the International Food Policy Research Institute, better known as IFPRI, the International Livestock Research Institute, ILRI, the International Rice Research Institute, IRRI, and the International Potato Center. All those are CGIAR. 00:05:15.045 --> 00:05:32.000 But we also have folks from other development organizations, both governmental and non-governmental, including Cultural Practice, Technoserve, Catholic Relief Services, Chorus International, Mercy Corps, Salvation Army World Support Office, 00:05:32.000 --> 00:05:37.000 the Society for International Development, or SID, Jessie Global, 00:05:37.000 --> 00:05:52.000 African Economic Research Consortium. Belgian Agency for International Cooperation, the Alliance for Biodiversity and CEAT in Canada, the German Development Agency for International Cooperation, 00:05:52.000 --> 00:05:57.000 Actions transforming lives in Liberia and former USAID people. 00:05:57.000 --> 00:06:03.000 I'm sorry I can't say, and also USAID former retirees. 00:06:04.180 --> 00:06:19.000 So some brief housekeeping notes. As you saw, if you saw the welcome slides, this event lasts approximately 45 min. Our speaker's presentation will be about 2025, followed by about 20 minutes of discussion and Q&A with the audience. 00:06:19.000 --> 00:06:30.618 All participants will be muted to enable the panelists to present without interruption. During any time during this talk, you can post questions in the Zoom Q&A, and we'll read those out after the presentation, or 00:06:30.618 --> 00:06:43.133 after the speaker's presentation, people can raise their hands through the virtual buttons, and we'll ask you to unmute yourself so you can speak directly to the speaker or to others if we have a discussion going on. 00:06:43.133 --> 00:06:47.305 But please wait your turn to speak so we avoid interruptions and maintain etiquette. 00:06:47.305 --> 00:06:56.353 If you'd like to enable captions at the bottom of your Zoom screen, you can select more, then show captions, and select English. 00:06:56.917 --> 00:07:05.271 And our event will be recorded, and that was in the welcome thing, and it'll be available on our website and on the Virginia Tech Library VT works. 00:07:05.271 --> 00:07:22.000 So a little more on our speaker. Cheryl Doss is a professor of economics at Tufts University. Her research focuses on rural transformation and gender issues in agriculture, household and intra household decision making and women's ownership and control over assets, including land. 00:07:22.000 --> 00:07:32.000 Her research on methodological innovations and survey design and analysis is designed to allow richer understanding of multiple and competing voices within households. 00:07:32.000 --> 00:07:36.000 Increasingly, her work is interdisciplinary and incorporates qualitative data. 00:07:36.000 --> 00:07:53.000 Her publications span the fields of economics, agricultural economics, and development studies. She's worked with international organizations, including IFPRI, the World Bank, UNDP, FAO, African Development Bank, and the Bill and Melinda Gates Foundation. 00:07:53.000 --> 00:08:04.511 Today, the discussion will provide an overview on the role of women's empowerment in mitigating the impacts of shock on household food insecurity in rural Oromia, Ethiopia. Please welcome our speaker. 00:08:12.461 --> 00:08:16.615 Great. Thank you very much. I'm really delighted to be here today. 00:08:18.769 --> 00:08:34.000 And I'm really looking forward to comments and questions we are in the very, very final stages. I promised my co-authors that I will get them the final draft of the paper in the next few days, so glad to have anything, any suggestions or things that aren't clear. 00:08:34.000 --> 00:08:41.000 What we're going to look at today is this question about whether women's empowerment can 00:08:41.000 --> 00:08:54.000 mediate the impact of shocks on household food insecurity, and I also want to recognize my two co-authors, Maria Hillesland and Martina Careta. 00:08:54.000 --> 00:09:17.000 See if I can make this change slides. There we go. Good. So as I'm sure most of you know, there's a long history of thinking about women's importance to food security. People have been writing on this for decades now. We know that women are producers of food as well as being involved in the processing, preparing, and allocating of food within the household. 00:09:17.000 --> 00:09:27.000 Right? So, long history of work on this. More recent work has generally found a positive relationship between women's empowerment and food security. 00:09:27.000 --> 00:09:44.000 So, the literature on that is pretty broad as well. It's a little bit all over the place, in part because people use really different measures of women's empowerment and very, very different measures of food security. But generally, most of the work finds a positive relationship. 00:09:44.000 --> 00:09:59.641 between those two. One of the contributions of our work today is that we're going to consider women's empowerment not as being a standalone thing, but we're going to think about it in the context of her household, also considering the empowerment levels of her husband. 00:10:02.000 --> 00:10:09.000 And the other thing we do is we're going to think about the relationship between women's empowerment and food security when there's shocks. 00:10:09.000 --> 00:10:15.000 So, we know that shocks generally increase food insecurity. 00:10:15.000 --> 00:10:30.566 I was only able to come up with good pictures for a couple of kinds of shocks. Price shocks are harder to think about what that looks like, but there's a huge range of kinds of shocks that households face, and pretty clearly they often increase food insecurity. 00:10:30.566 --> 00:10:38.487 But the question then today is there's women's empowerment that we know can be important for food security. Is it also important in these contexts with with shocks? 00:10:40.692 --> 00:10:58.000 So the data that I'm going to use is data from 528 couple households in Oromia, Ethiopia. We have two rounds of data collection. So the first round was done in December of 2016 and January of 2017. 00:10:58.000 --> 00:11:04.000 And then went back two years later, in February and March of 2019. 00:11:04.000 --> 00:11:13.565 The data was actually collected as part of an impact evaluation of the UN joint program on rural women's economic empowerment. 00:11:14.591 --> 00:11:22.298 But we're not going to use any of the of the impact evaluation kinds of things. We're just going to be using the data. 00:11:22.657 --> 00:11:37.977 We do everything with the the full sample, controlling for program participation, but then we also run the regressions and do that with just the control sample as well, which gives us basically the same results. 00:11:37.977 --> 00:11:51.000 So we're not going to talk about the impact evaluation piece of it. What we're able to do, I'm just going to give you the quick overview now to give you a sense of what we're doing. In the initial round, we asked people questions about empowerment. 00:11:51.000 --> 00:12:07.000 Questions to both the husband and the wife in the household using the project-level Women's Empowerment in Agriculture Index. So we have initial levels of empowerment of the husband and wife. We also administer the FIS, the Household Food Insecurity Experience Scale. 00:12:07.000 --> 00:12:24.000 In the initial round. When we go back two years later, we asked lots of other things, too, of course, but we asked if the household experienced any shocks, and we readminister the fears. So what we're able to do is look at whether households the change in household food insecurity. 00:12:24.000 --> 00:12:31.974 How that's related to them having experienced shocks. And to the empowerment levels of the husband and wife in the household. 00:12:35.154 --> 00:12:51.000 So, just a little bit on the context of where this is, these are rural households that rely on agriculture for their livelihoods. The region was experiencing a drought at the time of the first survey round. So, you know, we're not. 00:12:51.000 --> 00:12:58.000 In some sense, we are looking at specific new shocks that happened over this two-year period, but things were not particularly good at the initial point. 00:12:58.000 --> 00:13:17.000 It's poor, fairly marginalized communities, more than a third of the men, and about 70% of the women are illiterate. About 10% of the households have tap water, either in their dwelling or on their compound, and 24% have electricity. Almost all of them report living in homes with. 00:13:17.000 --> 00:13:25.000 Earth, dirt, or sand floors. So that just gives you a sense of the context in which we're working. 00:13:25.000 --> 00:13:36.000 So what I need to do now is tell you about the 3 kinds of measures, 3 kinds of data that we're going to be using. So the 1st one is the women's empowerment data. 00:13:36.000 --> 00:13:52.000 And as I said, we're using the ProWEAI. The ProWEAI has thinks about three kinds of agency. So the ProWEAI is part of the family of women's empowerment and agriculture indices. This one was designed specifically. 00:13:52.000 --> 00:14:04.000 for impact evaluation. So thinking about it at that kind of a level of a set of communities, and trying to think about changes over time in empowerment. 00:14:04.000 --> 00:14:13.000 Three types of agency, intrinsic agency, which we often think about as the power within kind of whether you have the self-confidence. 00:14:13.000 --> 00:14:29.000 to be able to make changes in your life, collective agency, which is the power with, working with others to accomplish things, and then instrumental agency, which is the one that most development programs actually typically work on. 00:14:29.000 --> 00:14:34.000 And think about, um, so your ability to do things. 00:14:34.000 --> 00:14:41.000 And here's the basic levels of empowerment that we find at the beginning in our baseline survey. 00:14:41.000 --> 00:14:51.000 So the ones to notice first are the ones in blue. Those are the ones where men's empowerment is higher than women's, on average. 00:14:51.000 --> 00:15:02.000 and particularly notice that that's in terms of work balance. 80% of men, but only 38% of the women are empowered in this particular dimension. 00:15:02.000 --> 00:15:16.000 And also, men have a much higher… much higher share of men are empowered than women. So 70% of women of men, but only 55% of women are empowered in this sample. 00:15:16.000 --> 00:15:29.000 The ones highlighted in green are the ones where women are more empowered than men. So an input in productive decisions, but also in control over use of income. 00:15:29.000 --> 00:15:39.000 And so these are these real differences, not just in the levels of empowerment between men and women, but in which dimensions of empowerment. 00:15:39.000 --> 00:15:48.000 they're empowered in is gonna matter for our trying to understand what we see in our results. 00:15:48.000 --> 00:16:00.533 What we do with this data is we're going to create four household types. We're going to look at households where both are empowered, where neither is empowered, where only the wife, or only the husband is empowered. 00:16:00.533 --> 00:16:08.693 So to think about what the wife's empowerment, women's empowerment, not just alone, but also how does that relate to whether or not her husband is empowered? 00:16:08.693 --> 00:16:30.000 Of our households, 228 of them. Both are empowered, but we have 61, which was a little surprising to us, in which the wife is empowered and the husband is not. We also have 97, where neither are much more common for the husband to be empowered and the wife not to be empowered in the other way around. 00:16:34.000 --> 00:16:50.000 So what kinds of shocks did these households experience? So I'm showing you this data by these four different household categories, neither only husband, only wife, or both empowered. Just looking at the graph before I tell you much about the shocks. 00:16:50.000 --> 00:17:01.000 You ought to be able to see that the big story is across the different types of shocks, rather than differences across households, the different household types. There's some differences across household types. 00:17:01.000 --> 00:17:07.000 We first look at whether households experienced any type of shock. 00:17:07.000 --> 00:17:17.000 So whatever kinds of shocks, if they reported any kinds of shocks, and about 30% of our households reported experiencing a shock during those two years between the first round and the second round. 00:17:17.000 --> 00:17:22.000 We then categorize most of the shocks into one of three types. 00:17:22.000 --> 00:17:32.000 The first one is agricultural shocks, which could be shocks to either crop production or livestock production. Lots of animals, loss of crops. 00:17:32.000 --> 00:17:49.000 The second one is the increases in prices, and that could be either increases in input prices for the things that they're purchasing for their agricultural production, or it could be consumer prices, particularly food prices, and food prices were going up at this time. 00:17:49.000 --> 00:18:00.000 And the third category of shocks is health shocks, which are either a serious illness of someone in the household or the death of a household member. 00:18:00.000 --> 00:18:14.000 So those three are almost all of the things that people reported as any shock. There were a few people who reported theft and a few people who reported loss of jobs, but basically those three capture most of them. 00:18:14.000 --> 00:18:31.000 We also were interested in sort of in thinking about all the literature on dealing with shocks and whether communities are able to ensure each other and how people kind of their own risk networks. We thought it would be useful to look at compare idiosyncratic shocks. 00:18:31.000 --> 00:18:37.000 Those were just that household had a shock and not everybody else with covariate shocks. 00:18:37.000 --> 00:18:45.000 And to categorize these, we counted all health shocks as idiosyncratic shocks. 00:18:45.000 --> 00:18:52.000 That wouldn't be true during the pandemic, but it was true during this particular time period in Ethiopia. 00:18:52.000 --> 00:19:12.000 There was not kind of systematic health issues across the communities. So those were idiosyncratic. All of the increases in prices we treated as covariate shocks, everybody experienced those. That wasn't all of the communities talked about that when people talked about. 00:19:12.000 --> 00:19:17.000 Price shocks. And then for agricultural shocks, we treated them as covariate shocks. 00:19:17.000 --> 00:19:31.000 If 48% or more of the people in the community reported having agricultural shocks. So those agricultural shocks were widespread. We treated them as idiosyncratic shocks if. 00:19:31.000 --> 00:19:43.000 20% or fewer experienced agricultural shocks. We didn't have anybody in that middle category, so it was really clear to how to separate those two out. So we have those two groups that we're going to look at. 00:19:43.000 --> 00:20:05.000 As well. Then we need to think about our food insecurity measure. What we're going to use is the FEIS, the Food Insecurity Experience Scale. And basically, this is a scale asking about people's experiences with regard to food and the lack of food. 00:20:05.000 --> 00:20:21.000 households that get reported as being moderately food insecure, that means that at times they may have needed to reduce the quantity as well as the quality of food that they would normally eat. And households experiencing severe food insecurity would have gone for extended periods of time. 00:20:21.000 --> 00:20:32.000 without eating. In our survey, in the initial round, 48% of the households experienced either moderate or severe food insecurity. 00:20:32.000 --> 00:20:40.000 In the second round, it was a little bit lower at 45%, so things were a little bit better for some people. 00:20:40.000 --> 00:20:56.000 although households that did not experience a shock were on average 10% less likely to experience moderate to severe food insecurity in the second round than those who did experience the shock between those. So we do see, as we would expect, that having a shock. 00:20:56.000 --> 00:21:02.000 makes you more likely to be more food insecure in the second round. 00:21:02.000 --> 00:21:12.000 We're gonna actually use the full continuous variable of the food insecurity scale, not just whether they're moderate or severely food insecure. 00:21:12.000 --> 00:21:17.025 And then we're gonna look at changes in their index of food insecurity. 00:21:20.718 --> 00:21:36.000 So for those of you who like to see an equation in a slot in a presentation. Here's the one that you're going to get. This is our empirical strategy, pretty much what I just told you earlier. We're going to look at the changes in food insecurity from over the time period. 00:21:36.000 --> 00:21:49.000 Whether the household is empowered. So one of those four, are they both empowered, neither empowered, wife only or husband only, then shocks. And we'll use different specifications using different. 00:21:49.000 --> 00:22:08.000 The any shock or the various different types of shock. And then an interaction term between the empowerment and the shock variables. We also have a vector of household characteristics in the initial survey, pretty standard kinds of things, as well as thinking about the changes in household size. 00:22:08.000 --> 00:22:19.000 And the number of children under age 5 over the two-year period. We control for fixed effects in the warita, and we also control 00:22:19.000 --> 00:22:23.359 For whether or not the households were members of the of the program. 00:22:25.000 --> 00:22:27.302 So let's look at the results. 00:22:30.226 --> 00:22:51.000 This one is going to show us the impact of experiencing any kind of shock on food insecurity. And what we see is that on the left two sides for neither and only husband, we see that when there's a shock, it increases the probability of food insecurity. 00:22:51.000 --> 00:23:02.102 But when either only the wife or both, so when the wife is empowered, we see that having the shack has no change in the probability of food insecurity. 00:23:04.615 --> 00:23:12.051 Suggesting that, at least at this level, right, women's empowerment is mediating the impact of shocks on food insecurity. 00:23:14.308 --> 00:23:25.000 We can then look at this in terms of the other, specifically thinking about other types of shocks or categorizing them as 3 types. 00:23:25.000 --> 00:23:40.000 And what we see again on the left, for agricultural shocks and health shocks, we actually see that those don't have any impact on food insecurity. So experiencing an agricultural shock or a health shock does not. 00:23:40.000 --> 00:23:47.000 change does not make you more likely to be more food insecure in the second round. 00:23:47.000 --> 00:24:01.000 But with the price shocks, we do see an increase in food insecurity, and we see an increase in food insecurity, an increase in the probability of food insecurity, to be technically correct. 00:24:01.000 --> 00:24:06.000 For all of the types of households except those in which only the wife. 00:24:06.000 --> 00:24:15.409 is empowered. So if only the wife is empowered, she is able to mediate the impact of price shocks on household food insecurity. 00:24:20.000 --> 00:24:28.000 And then, again, we can look at divided up this other way thinking about covariate and idiosyncratic shocks. 00:24:28.000 --> 00:24:45.000 Here, with idiosyncratic shocks, on the right, we see no change in food insecurity in the probability of food insecurity because of an idiosyncratic shock, which says that people probably are, in some sense. 00:24:45.000 --> 00:24:53.000 ensuring each other and taking care of each other in the community to some extent when there's a covariate shock. 00:24:53.000 --> 00:25:09.000 When the woman is empowered, either only the wife or both, then we again see that she's able to mediate the impact, and there's no impact of the shock on food insecurity. Whereas if she's not empowered. 00:25:09.000 --> 00:25:15.230 And it's either neither or only the husband, then food insecurity is worse after the shock. 00:25:18.000 --> 00:25:31.000 So there's lots of reasons why this might be the case. So we thought we would also look at the coping strategies that they used to try to understand a bit about what's going on. 00:25:31.000 --> 00:25:45.000 And we do see some pretty significant differences in the coping strategies that were used. People were asked to list up to three coping strategies that they used to deal with the shock if they said that they'd experienced one. 00:25:45.000 --> 00:26:00.000 the main strategy that all the households use. Or yeah, pretty much is to sell livestock. So again, it's, I think, useful to remember that these people were already experienced had been experiencing a drought. 00:26:00.000 --> 00:26:17.000 Before we initially talked to them and then had an additional shock after the drought. And so selling livestock is a coping strategy. It's also one that can make them worse off in the long run. But we see that all the households did that. 00:26:17.000 --> 00:26:36.000 But we see that the households in which only the wife was empowered was much more likely to use savings, which is interesting. That suggests that those households have some savings, but also more likely than the other kinds of households to receive help from friends or family. 00:26:36.000 --> 00:26:40.282 So, somewhat different patterns depending on who's empowered within the household. 00:26:43.000 --> 00:27:00.000 We also see a little bit of difference just in the number of coping strategies that were used. So we asked people to list up to 3. Only a few people listed 3. So we think that that was fine to only ask about 3. We're not missing something important. 00:27:00.000 --> 00:27:13.384 That they didn't tell us in their fourth, fifth, or sixth ones. But what we do notice is that when only the wife was empowered, they're more likely to use two coping strategies that any of the other types of households. 00:27:18.000 --> 00:27:32.000 So let's think a little bit about what we think this means. What are the implications? So what I've just shown you is that women who are empowered are able to mediate the impact of shocks on household food insecurity. 00:27:32.000 --> 00:27:44.000 And that when households have experienced any shock or covariate shocks, those worsen food insecurity in those households where the wife is not empowered, right? 00:27:44.000 --> 00:27:54.000 And again, there's no impact of price shocks when only the woman is empowered. So what do we think is going on? Let me say that. 00:27:54.000 --> 00:28:10.000 We were a little bit surprised by the price shocks one. We had expected that having women empowered in the household might mediate the shocks, but we thought that having both people in the household empowered would be better. 00:28:10.000 --> 00:28:22.000 than having only the woman empowered. That kind of household dynamics when they're both empowered would be better and that they would be better able to deal with the shocks. We do see that's. 00:28:22.000 --> 00:28:35.000 not necessarily better, but that when both of them are powered, that mitigates the shocks as well for some of these types of shocks, but not for price shocks. So what do we think is going on with price shocks? 00:28:35.000 --> 00:28:43.000 Well, one of it may be that women are more likely to be empowered in the dimension of control over income. 00:28:43.000 --> 00:29:00.000 Right? And so it seems to be, or at least the story would be consistent with that, when women are more empowered, and they have control over income, and particularly when they have control over income, and their husband says that he does not have control over income. 00:29:00.000 --> 00:29:07.214 They're able to use that control over income to protect food security. 00:29:10.000 --> 00:29:23.782 We also see a bit that the coping strategies differ across households, and we don't have a way with the data that we've got to be able to really show that that's what's causing it, but certainly the patterns that we see are suggestive that different coping strategies. 00:29:25.269 --> 00:29:45.000 The different households are used different types of households are using different strategies, and so that may be part of the story. And so again, this bigger part of the story is that we're looking at men's and women's empowerment and asking whether the men and women, the husband and wife in the households are empowered. 00:29:45.000 --> 00:29:57.230 But they're empowered in different dimensions, and so it may be that the dimensions that women are empowered in are the ones that they're really being able to take advantage of and use to protect their food security. 00:30:00.000 --> 00:30:15.000 I want to just put in 2 caveats. This first one, I think, is really important, which is that we're only measuring one kind of welfare outcome food security. Food security is a really important welfare outcome, but certainly not the only one. 00:30:15.000 --> 00:30:22.000 So, although we can say that having the women be empowered really matters, and that in this one case. 00:30:22.000 --> 00:30:38.000 We see that not having the men empowered helps to protect against price shocks. We actually don't know about other outcomes, and so we don't know. We don't have the data to look at whether both of them being empowered may be associated with other kinds of positive outcomes. 00:30:39.744 --> 00:30:45.000 So, it may be also good for the household in other kinds of ways. 00:30:45.000 --> 00:31:03.000 And then just finally, I would note that this is a very particular context with very high levels of food insecurity in the initial round, as well as in the later round. And so I would be really interested to look at some of these kinds of issues in other contexts as well. 00:31:03.000 --> 00:31:17.897 But I think part of the takeaway is that we really gain something more when we don't just look at women's empowerment on its own, but put it in the context of thinking about whether the woman's empowered, but also whether the man is empowered in the household. 00:31:20.000 --> 00:31:21.641 And I will stop there. 00:31:24.154 --> 00:31:32.000 Wonderful. Thank you so much, Cheryl. And now the discussion starts, and I'm I'm going to read the 1st question from. 00:31:32.000 --> 00:31:47.000 A student at UC Davis, who is in a quiet study room. So Anna Cepeda asks, are loans granted only granted to men? I noticed that when none or both are empowered, loans are a coping strategy. 00:31:47.000 --> 00:32:03.000 No, so that one's a little bit complicated because part of this project that was being implemented in the area was providing loans. There are loans. Women were able to take out loans. The project didn't actually work all that well in this area. 00:32:03.000 --> 00:32:12.254 and some people who initially had credit lost credit, but certainly it wasn't the case that only men were accessing loans. 00:32:25.795 --> 00:32:32.000 Well, well, somebody gets brave and, asked to unmute themselves. I have a question. 00:32:32.000 --> 00:32:33.000 Okay? 00:32:33.000 --> 00:32:42.000 I'm wondering how it's really a broader question. It's kind of more of a discussion. But how has your work incorporated qualitative data. 00:32:42.000 --> 00:32:53.000 What are some of the benefits? What have been some of the pushbacks? I mean, you're an economist, right? I do qualitative research working with mixed methods and, you know, working with ag economists, but. 00:32:53.000 --> 00:32:59.000 over and over we we get that's just stories. So where's the data? 00:32:59.000 --> 00:33:07.000 Right? So I just wonder if you can say what you know how you've managed over your career. Has this changed and. 00:33:07.000 --> 00:33:09.000 That's the topic. I'll stop. 00:33:09.000 --> 00:33:29.000 That is a big topic. Yes, I've thought lots about qualitative data. One of my co-authors, Dee Rubin, who's, I think, on the call or listening in, she and I have a paper where we've looked at how qualitative and quantitative data. 00:33:29.000 --> 00:33:41.000 used together has really changed how we think about women's empowerment. Looking back over, I think we talk about 15 years and a number of different projects where, 00:33:41.000 --> 00:33:56.000 I think one of the things that we argue in that paper is that having long-term relationships between people doing qualitative and quantitative work really makes a difference so that you have these iterative back and forth conversations. 00:33:56.000 --> 00:34:14.000 Where you try out something, the quantitative data tells you one thing, and then the qualitative data maybe tells you something different, and so you have to go back to the quantitative data and try to figure out what's the story. And we have a number of examples in that paper of where we really learned some new things. 00:34:14.000 --> 00:34:25.000 from putting those two kinds of data together. So we don't have very much of it. We don't have much qualitative data in this particular paper, but most of my work has. 00:34:25.000 --> 00:34:41.000 I've incorporated more, but I mean, I certainly, as an economist, I talk to other economists who say, oh yeah, we had, we commissioned a qualitative report and they did it, and they sent it to us. And so we've got some qualitative work, right? I'm sure you've experienced that. 00:34:41.000 --> 00:34:54.000 But I think it's really those deep conversations over and over, back and forth, trying things out that really kind of it shapes the new questions that come out. 00:34:54.000 --> 00:35:03.000 Okay, I'm gonna I'm gonna ask to see we have a bunch of questions, but they're coming in in 2 different places. So they're kind of hard. But Bernice, do you want to ask your question, or do you want me to read it? 00:35:03.000 --> 00:35:05.615 Unmute yourself if you want to ask it directly. 00:35:10.949 --> 00:35:12.256 If I don't hear from you. I'm going to go ahead and read it. 00:35:30.667 --> 00:35:44.000 Yeah, can you hear me? I wasn't muted. I was muted before, so… Yeah, great presentation. Thank you so much, and I really got excited, because this is something I'm particularly interested into when it comes to empowerment. So I was. Well, the question is. 00:35:44.000 --> 00:35:54.000 purely economical, given that the coping strategies you mentioned, gives different outcomes. I was wondering if. 00:35:54.000 --> 00:36:10.000 At any point, you are accounted for the possibility of an endogeneity between their empowerment as an agency and then the food security, and what a possible mediating factor could be outside of the variables you considered. 00:36:13.667 --> 00:36:27.590 So we haven't thought about other mediating factors. I think with the endogeneity, what we're looking at. So we're looking at changes in food security or food insecurity, but we're using initial levels of empowerment. 00:36:29.101 --> 00:36:44.000 So we're thinking about how people's empowerment in that initial stage affects their food security two years later. So there could be some back and forth, right? It could be that food security could change people's empowerment. 00:36:44.000 --> 00:36:52.283 We do not have the empowerment data in the second round, so we don't look at changes, we don't look at changes in empowerment. 00:36:53.024 --> 00:37:09.000 that would be interesting as well, but that wasn't collected. But we think we can argue that there's some causality between the empowerment and the changes in food security, because we're just looking at the initial empowerment levels. We're not looking at changes in empowerment. 00:37:09.000 --> 00:37:12.000 changes. Okay. Thank you so much. 00:37:12.000 --> 00:37:13.358 You're welcome. 00:37:17.000 --> 00:37:22.820 Okay, I just sent a message. Let's see. Are you going to speak, or shall I read it? 00:37:26.667 --> 00:37:39.179 Let me go ahead and ask that question. Assuming a baseline and follow-up survey. The FIES scores at baseline and follow up were were the scales calibrated to ensure that a common scale for the items in the 2 surveys? 00:37:45.000 --> 00:37:53.000 I'm going to read. Yeah, it's I think it's got some typos. But so I'm not sure how to correct it, because I don't understand the question. Assuming a baseline and follow up survey. 00:38:00.000 --> 00:38:06.974 that a common scale for the items, probably that there was a common scale for the items in the 2 surveys. 00:38:08.640 --> 00:38:33.000 So, I guess I'm not understanding the question quite. It's the survey in the two rounds, right? It's the same questions. And there's two ways that people use the the one is where they just add up the number of things that they 00:38:33.000 --> 00:38:40.000 That the kind of add up, kind of just add up the indices, and then there's a more complicated way. 00:38:40.000 --> 00:39:01.692 that you do more econometric modeling of it to look at the probability scores. I don't want to try to explain that without, slides, and I don't have a slide that does that, but we've used that. I'm happy to send the details on it, but we're using basically the same approach for measuring food insecurity in the baseline and the second round. 00:39:03.590 --> 00:39:05.000 So can you hear me? I'm sorry. Can you hear me? 00:39:05.000 --> 00:39:07.000 Yes, yes. 00:39:07.000 --> 00:39:15.000 Okay, thank you. So, yeah, I if you are using the probabilistic approach or. 00:39:15.000 --> 00:39:18.000 Right. Yes. 00:39:18.000 --> 00:39:36.000 Okay. Okay, so my question is the two Scores should have some common support so that they are comparable. This is from, I used to work for USAID project. 00:39:36.000 --> 00:39:45.000 before the demise. So we used to have, like, FIES scores a lot, you know, to compare. 00:39:45.000 --> 00:39:51.564 over time. So what we did was we kind of did talk to Carlo Cafeiro, who is at. 00:39:51.564 --> 00:40:09.564 You're an FAO, who I think was involved in developing the affairs in R. So, what he told us was that we need to have a common support so that both are comparable. 00:40:16.205 --> 00:40:29.000 Because the two surveys were collected at different times in points. So my assumption is you use the rush model. 00:40:32.000 --> 00:40:35.000 I did not calculate this, my co-author calculated it, so I don't know all the details of it. 00:40:35.000 --> 00:40:46.000 Okay, okay, okay, that's fine, because, yeah, that's fine then, I'll stop, because it's basically, you have to use a Rash model to calculate the FIES scores. 00:40:52.000 --> 00:41:02.000 There are so many, you know, factors that could change because this is an attitude. I mean, it's just so well that's from my, you know. 00:41:02.000 --> 00:41:19.000 experience in the past while I was working for USAID project, which was, like, we had FIES scores at different times in points, so we need to have some kind of new this course should be comparable. 00:41:19.000 --> 00:41:20.000 Yeah, yeah. 00:41:20.000 --> 00:41:27.000 Okay, I'll go have a conversation with my co-authors and see what we've done, and if there's more that we need need to do on it. 00:41:27.000 --> 00:41:29.000 Okay, thank you so much, thank you. 00:41:29.000 --> 00:41:38.640 Thank you. So the next question I want to. I may just go ahead and read the question if you don't answer right away, because I want to have the time for all these questions. So, Chavi, do you want to speak your question? 00:41:42.948 --> 00:41:52.282 maybe just comment. Okay, so I'm going to read the question from Chavi Tiwari is, did you also look at the food insecurity at the baseline? And if how, and if and how that changed? 00:41:55.718 --> 00:42:01.000 Yes, what we're looking at is the change in food insecurity from the baseline. 00:42:01.000 --> 00:42:09.000 to the second survey. So how much did food insecurity change over that time period? 00:42:09.000 --> 00:42:24.000 That's the measure that we're using. Right, so we're not just saying, are household's food insecure in the second round after the shock, but we're looking at for that household, how much did food insecurity change? 00:42:24.000 --> 00:42:32.000 Okay, thank you, thank you so much, and I may have missed that in the beginning, because I joined a bit late. But yes. 00:42:32.000 --> 00:42:47.000 I'm really excited about this research, Dr. Doss, and we have a similar kind of research and your results validates what we are finding. So if when it gets published, if it's possible to share, that will be. 00:42:47.000 --> 00:43:02.000 awesome, because we are also looking at women empowerment in Ethiopia and its effect on food safety, food insecurity. 00:43:03.000 --> 00:43:15.000 During the War and like what happened. So, we are also finding the importance of livestock and women empowerment as a protective factor against food security. So this was really good. 00:43:16.000 --> 00:43:20.000 I would love to see what you're doing, so when you've got something, please send it to me. I'd be really, really, really interested. 00:43:20.000 --> 00:43:25.000 Yeah, yeah. We just finished that draft of first draft of the paper. So yeah, okay. 00:43:25.000 --> 00:43:27.000 Great. 00:43:27.000 --> 00:43:29.000 So, would you like to ask your question? 00:43:29.000 --> 00:43:44.000 Yeah, sure. So I was just struck by the language that you were using, talking about people as empowered or not empowered. And I just wanted to sort of ask you. 00:43:44.000 --> 00:43:54.000 to tell me or think about the idea of empowerment as a kind of on-off duality, rather than. 00:43:54.000 --> 00:44:09.000 empowerment as a continuity, and why you chose that approach in your in your analysis. I mean, it seems that that it 00:44:09.000 --> 00:44:30.000 covers over a lot of things. The differential in what we might mean by men's empowerment versus women's empowerment, for example, or the ways that we might not have a linear relationship, that there might be some level of empowerment that. 00:44:30.000 --> 00:44:47.000 that helps with food insecurity, and then above that, it may actually have a, you know, a negative impact. I mean, there's just a whole bunch of questions that get raised by treating empowerment as an on-off, rather than a continuum. 00:44:47.000 --> 00:44:52.000 across multiple measures. So I just wanted to hear your thinking about that. 00:44:52.000 --> 00:44:59.000 So, certainly, I completely agree with you that empowerment's much more complicated than we've 00:44:59.000 --> 00:45:08.000 done it here. I think the reason we did it this way is because we wanted to be able to think about. 00:45:08.000 --> 00:45:19.000 One way to do it would be to think about just the difference in men's and women's empowerment within the household, and see if that matters. But we think it's not just the. 00:45:19.000 --> 00:45:34.000 difference, but also the levels which is a little bit what you're saying. And so, by being able to categorize households as these four types. 00:45:34.000 --> 00:45:37.000 Then we can actually look at it, then empirically. 00:45:37.000 --> 00:45:53.000 It's fairly straightforward to look at the differences. So there's a lot of nuance that we're not able to pick up, but we think, part of what we're really saying in the paper, one of our strong points is that we need to think not just about women's empowerment. 00:45:53.000 --> 00:46:00.000 But also to think about men's empowerment within the household, and didn't want to just put them in as. 00:46:00.000 --> 00:46:09.000 We could have put them each in as separate continuous variables with some kind of right with their empowerment score from the pro WEAI. 00:46:09.000 --> 00:46:27.000 but that doesn't capture the relationship between them right that we think that there's something different in households where they're both empowered versus both not empowered. And sure, there's a lot of right. It's those choices of where you say empowered versus disempowered. 00:46:27.000 --> 00:46:44.000 are arbitrary, and I will certainly grant you that. And lots of other ways to look at it, and love to see other people looking at it in other kinds of ways. But I think we thought that these 4 categories really helped us to think about those relationships between within households. 00:46:44.000 --> 00:46:49.000 In a way that there wasn't another approach that let us do that. 00:46:49.000 --> 00:46:54.000 So I have a question. But I realize that your study perhaps. 00:46:54.000 --> 00:47:13.000 wouldn't have addressed this. But in your powerpoint. And actually, I actually did research with IPM Innovation Lab in Oromia also that involved women and livestock cattle in particular. And so I see that, you know, the coping strategy of selling livestock is one of women's and also joint men and women's. 00:47:13.000 --> 00:47:27.000 strategies. But it concerns me that, you know, then you actually mentioned it with, then then they can, you know, how does that strategy rebound if they've sold all their livestock? It's kind of scary if that's in that, you know, if that's like one of the important. 00:47:27.000 --> 00:47:31.000 Strategies for women. Any thoughts on that? 00:47:31.000 --> 00:47:37.000 Well, that was an important strategy for everybody, and I think that is a problem, right? There are coping strategies. 00:47:37.000 --> 00:47:54.000 that really leave you worse off, and there's some that may not leave you worse off. I don't think we're in a position in this paper to try to think about the long-term implications of this, but we do worry, right? You do worry about a place where you see that many people selling off livestock. 00:47:57.000 --> 00:48:10.000 I mean, I really appreciate that you're looking at, of course, men and women's, but it's just in our work, and then looking at even some of the just older surveys in in Ethiopia. 00:48:10.000 --> 00:48:22.000 you know, cows and dairy are like one of women's few coping strategy, few assets, let's say, so not studies on it being used as a coping strategy. So that's why it's just it's scary. 00:48:23.000 --> 00:48:30.333 We still have 15 minutes, maybe 10 for questions. Maria has. Oh, good. Okay. Maria, you want to unmute yourself? 00:48:35.052 --> 00:48:44.564 No, can't unmute. Okay, I'm going to read it. So I was interested in the higher use of savings for women in power households. Are women more likely to be involved in savings groups in Oromia? 00:49:09.000 --> 00:49:18.000 No. So there are savings groups and there are savings groups for women in these areas. 00:49:18.000 --> 00:49:25.000 But I don't know. I could have to go back and see if we can figure that out in the data, whether women are more likely to be. 00:49:25.000 --> 00:49:32.000 In a savings group than men. So that's a great question. 00:49:32.000 --> 00:49:50.615 Great. Thank you. Yeah, I so I work in similar projects where we do a lot of women empowerment and savings groups connection. So I was just I was curious to see if that was part of, you know, maybe why women in empowered households had a higher relative use of savings when the shocks came. 00:49:54.051 --> 00:49:58.102 Good, I'll go and try to play with that a little bit and see if I can come up with a better answer. 00:50:10.000 --> 00:50:11.512 Any other questions? 00:50:19.000 --> 00:50:26.820 Okay, so. Deborah Rubin, I don't know if you're still in a car. If you can unmute yourself, go ahead. Otherwise, I'm going to read this. 00:50:29.744 --> 00:50:45.000 I'm not in the car. And thank you, Cheryl. This was a great presentation and very clearly presented, even for me, the qualitative researcher. I'm really curious about whether. 00:50:45.000 --> 00:50:54.000 there was access to livestock risk insurance in this area, and if so, whether both men and women used it, and if you. 00:50:54.000 --> 00:51:06.000 had any information about its effects on food security. I realize that's not the major part of this particular paper, but part of the context. 00:51:06.000 --> 00:51:17.000 I don't think so. I haven't heard any conversations about livestock risk insurance in these particular communities. 00:51:17.000 --> 00:51:25.000 But I will check that out too, see if we I don't think we have any data on it, but I can ask colleagues who are working in there. 00:51:25.000 --> 00:51:25.998 Thanks. 00:51:38.000 --> 00:51:52.332 Well, this is called a discussion series for a reason. So if anybody would like to say something and not just grill Cheryl Doss, you are welcome to unmute yourself and have any comments. 00:52:02.641 --> 00:52:13.000 I'm happy to share that. So I'm a practitioner, not a researcher, and this is one of the, you know, I actually really appreciate this study and how. 00:52:13.000 --> 00:52:33.153 Practical in terms of those of us who are practitioners can read it and understand it, or at least the presentation a lot more than, you know, sometimes some of the other ones. And, you know, we use a lot of, like, the FIES, and we use a lot of these tools and measurements, so kind of being able to see some of these correlations is really helpful. 00:52:36.846 --> 00:52:39.000 Good, I'm really glad. 00:52:39.000 --> 00:52:44.334 Amanda Crump, UC Davis, unmute yourself, please. 00:52:44.334 --> 00:52:51.000 Hi. All right. I think I did. Okay, great. Good to see you, Cheryl and everybody will kind of seeing you. 00:52:51.000 --> 00:53:02.000 Um, so I just had a clarifying question about the women, especially the households in which the women are the empowered folks. 00:53:02.000 --> 00:53:10.000 And I'm not really that familiar with Ethiopia, but, 00:53:10.000 --> 00:53:26.000 Are the household structures all the same, or are we talking about women that may be widows? I ask because in some of the research from Chesney McCumber on the community concept drawings where they were. 00:53:26.000 --> 00:53:40.000 Defining empowerment. They were finding that in some cultural contexts, more empowered women were actually ones who had lost their husband and like the community offers some of that. 00:53:40.000 --> 00:53:54.538 empowerment support. And I just I might have missed it, but I just wondered about household structure, or is there, like, more than one wife in some of these situations, and does that impact the empowerment of those women. 00:53:56.923 --> 00:54:12.000 Good. These are all couple households. So there's not widows in the households, or they're not widows being interviewed. They're all households where we've interviewed both the man and the woman so that we have information on them. 00:54:12.000 --> 00:54:24.000 man's empowerment, as well as the woman's empowerment. We did do a little bit trying to look at those households where the women were empowered and trying to figure out if there was something. 00:54:24.000 --> 00:54:37.000 really substantially different about them, and we there's a little bit of stuff, but it wasn't. There wasn't something where we could say, oh, wow, these are all households in which this particular characteristic 00:54:37.000 --> 00:54:43.000 was happening. But they are all couple households. 00:54:43.000 --> 00:55:01.000 Cool. Yeah, that was my main question is just drawing on that work from Chesney was just really interesting when like the community defines empowerment, it really shifts the conversation about what empowerment means. And so kind of away from. 00:55:01.000 --> 00:55:06.000 the way we define it through the WEAI. So yeah. 00:55:06.000 --> 00:55:08.000 Very, very fascinating. Thanks. 00:55:08.000 --> 00:55:13.819 I did some work years ago where we're looking at asset ownership, in Uganda. 00:55:14.255 --> 00:55:25.255 And was really surprised by the finding that widows were more likely to own land than other women. 00:55:25.511 --> 00:55:34.230 That was the category of women most likely to own land, which didn't make sense of a lot of the qualitative work and the stories that we were hearing out of these areas where we were hearing about women. 00:55:35.000 --> 00:55:51.000 Who became widows being dispossessed, and various other things. And what we finally decided was that part of what was going on is that the only widows who show up in a survey, are the ones who have been able to stay on their land. 00:55:51.000 --> 00:55:52.000 And so it's the fact that they can stay on their land. 00:55:55.000 --> 00:56:10.000 means that they show up in the survey, and the other widows, the ones who were dispossessed of their land, may have left the communities they may be living in somebody else's household, right? Moving in with their husband's brother, traditionally, and may or may not show up in our data. 00:56:10.000 --> 00:56:19.000 At all, and so the women who show up are the ones who own land, and the other ones just kind of disappear. 00:56:19.000 --> 00:56:24.000 Yeah, that makes sense. And that could be part of what's going on here, too. Yeah. 00:56:24.000 --> 00:56:25.000 Cool. Thanks. 00:56:25.000 --> 00:56:32.000 Okay, so Jean Sumner, so apparently you people can't unmute themselves very easily. I'm going to ask her question. 00:56:33.000 --> 00:56:42.000 Okay. Did you hear me? Okay, did you control for the educational level of the women and, for that matter, the men? 00:56:42.000 --> 00:56:48.000 We did. We controlled for education levels. The education levels were all pretty low, but we did. We certainly did control for them. 00:57:07.000 --> 00:57:13.000 So I'm going to start wrapping up unless somebody wants to ask another question. Raise your hand or put it in the. 00:57:13.000 --> 00:57:14.024 Q&A or the chat. 00:57:23.000 --> 00:57:37.000 I want to thank everybody for joining us. If somebody puts a question, I will stop, but I want to thank everybody for joining us. We have just 5 min to go and for Dr. Doss for presenting. 00:57:37.000 --> 00:57:47.000 Thank you for a wonderful presentation. And we're sharing a short survey. If one of my people is not putting it in the chat, then. 00:57:47.000 --> 00:57:55.000 Thank you, Pratiraksha. My lovely grad student. Please respond to it to help us. 00:57:55.000 --> 00:58:11.000 raise funds for future discussion series, and you'll also get it, I think, through through Zoom. But your opinion is very important to us. And I also had put in earlier, and I'll try to put it in again if I stop talking how you can just. 00:58:11.000 --> 00:58:21.000 as yourself or to our listserv directly. And our listserv, I always say I don't bomb people with anything other than information about the discussion series. 00:58:21.000 --> 00:58:25.000 and you can check our website for more information. 00:58:25.000 --> 00:58:31.000 to get updates on future events. Our future event would be in the fall semester on the academic year. 00:58:31.000 --> 00:58:34.000 and lots of thank yous on there. 00:58:34.000 --> 00:58:40.820 Great, and thank you all for your comments and questions. I'm going to go figure out a few more of these. So that's that's great to do. 00:58:44.205 --> 00:58:46.306 So here is a link.