 Good morning, everyone, and then to Oliver. I am a fan. I don't know if you've been told, but as a Ghanaian, I mean, with education in Ghana, it's been Oliver Murray saying there from school and then at work. So thank you for, first of all, giving me this opportunity to talk to you, and I am really excited about this. I mean, for maybe people who might not know you, maybe it's about teaching and all, but this chat has to do with the other side of what you do, and it's mainly as a lead in at the time in the development studies journal. So the main thing I would want to ask is that what kind of advice would you give to incoming researchers, especially in developing countries, who would want to publish? Because you're an editor, so you would kind of get some point on what they can do. I think the big thing for anybody, but especially for people from the global south, is if you're submitting to one of the well-known recognized journals, the best attitude for you to take is that that journal is not interested in your paper, because they get so many submissions that they have to be very selective in which ones get sent for review, and which ones will ultimately be accepted. So you have to convince the editor that you've got an interesting paper, that it's addressing an issue in the literature, that your methods, your data are all high quality, that you've got appropriate methods, appropriate techniques, and you've basically got to do that in the abstract and the first page of the introduction. Because it's helpful to assume that that's all the editor will look at, if they've got five to ten submissions each week to look at, which is typically what I have. You're not going to read each paper. You're going to have a quick look to try and decide, will this fly? Are reviewers likely to be interested? Are they likely to accept the paper if it goes to review? So you've got to think, okay, how do I convince people that my paper is a good paper that they'll be interested in? So you can kind of think of that from page of the first page of the introduction, that first of all has to say clearly what is the issue? What's the gap in the literature? And particularly it's helpful if you can refer to recent established papers that say that we found the following, but there are certain topics that are issues that we really don't have a lot of evidence on, or we don't know what the answers are. And so the first, the start of that introduction frames the contribution. And then the second part of that first page has to say, well, what data do you use? What methods do you use? How do you contribute? That's not easy. So in many respects, you have to write the paper and then go back and write, rewrite, rewrite, and rewrite the introduction. You're happy that the first page is the handle, it's the catch. And it makes people think, oh yeah, that's an interesting topic. We don't know much about it. So nowadays, people are starting to have say papers on COVID where they have data. And most of the serious journals, they're not interested in attitudes or opinions. They don't really want descriptive papers, they want analytical papers. So papers that analyze data and provide an answer. And what has become the big issue that all reviewers would look at is, particularly if it's a quantitative paper, is identification. Now often that's framed in terms of can you draw a causal inference. But more generally, it's really, can you convince the reader, and in particular the referee, that the effect you're claiming is valid? That you've done whatever you can to account for endogeneity, account for all of the problems with the data, so that you can say and demonstrate with confidence that this is a real effect. It may not be a huge effect. You're lucky if it is a huge effect. You may not be able to discount all of the possible confounding influences. You almost never will. Identification is rarely perfect. But you have to be convincing that, look, I've used techniques that are appropriate for the data I've got, that the data are suitable to address the question I've asked. And this is an answer. And this answer is new. Now, new doesn't mean it's earth shatteringly original, because everybody's addressed the topic. In fact, typically, almost all papers are contributing on a topic that where there are questions, there are things that people don't know. And you have to convince that, yeah, well, look, I've got good data, I've found something interesting. It might be specific to the country. That's not a big problem as long as it can speak to a more general issue. And I think broadly speaking, kind of micro papers, papers based on household, experimental, lab in the field, labor market, individual data, survey data that's analyzed, are better able to be convincing than micro papers. So the student, the person who's done Granger causality on exports and growth, or run a growth regression, the big journals are going to be interested in that because it's fairly routine. It's very hard to be distinctive in that. And in fact, I think what's become much more important now, given the large number of researchers throughout the world, lots of people producing papers, data, the data you have and the quality and novelty of that data of what's really important. So for example, in macro type regressions or growth regressions, using annual data and doing fairly routine things that anybody would learn in a master's class, that's not going to cut it. But what you might want, what might be convincing is, well, have you got good, interesting monthly data, higher frequency data, so that you can try to identify an effect? If it's micro data, if you don't have a panel, you're limited. So you've got to think really carefully about, well, how can I identify the effect I want if I don't have a panel. So it's not easy, but that's why the top journals are top journals. They're trying to find papers that have good data, address a good question, and give an answer. And that that answer is something that will be of interest to other researchers. Okay, that's, I mean, that's a, that's the tallest, but then from, I mean, from the letter that I understand that has to do with being concise, trying to fit away at least most of the talent effects of your paper on your first page, try to attack the gap, how you contribute, try to gauge in for the analytical stuff more than the descriptive and opinionated papers, look out for your identification, causal effects, and then your data should be good. Yes, that's mainly a summary of what I understand. But you know, the other thing is that developing economics in general, it's really changing. I mean, when we were learning, there was really, that there was the feel that, okay, well, we knew, we knew what development economics was, but then maybe it wasn't this huge instance where someone had won the Nobel. I mean, when I, when I was doing my program, undergraduate or my master's, no one in development economics had won the Nobel Prize. But now it's huge and it's evolving. So based on papers that have been published in your edited series, what you do, how do you see that field evolving? And how do we, as researchers, as in try to face it head on, because it's not going to change or it's not going to stop? But then how do we tackle this evolving field? So in the first place, do you really see it evolving or it hasn't changed? And secondly, how do we tackle it head on? It has evolved and it has changed. And in one sense, I think there's a, there's a research inequality that needs to be considered. You know, one of the major growth areas over the last 10, 20 years has been the randomized controlled trials. And of course, that's what's generated the most recent Nobel Prize in development. So I would point out that if you go back to the early days, there were a number of Nobel Prizes in development. Arthur Lewis got a Nobel Prize. Martha Sen got a Nobel Prize. They kind of got Nobel Prizes for big ideas, for big thinking. Whereas now it's kind of shifted more. It isn't, to a sense, small ideas, small ideas, addressed in a lot of detail. Now, if you want to do RCTs, which are very popular, there's diminishing market returns now, but also they're expensive. You have to have research funds. So that tends to be dominated by people, particularly North America or European, the Western institutions where they can get the funding. And some researchers in developing countries may benefit by being, by association. They get involved in those projects. But that doesn't help a lot of people who don't have access to those resources. So what they've got to do is think, well, okay, how can I do good analysis with existing survey data? And, you know, there are, there are a lot of survey data around those, you know, the Ghana Living Standards Survey, lots of surveys that have been analyzed, but haven't been completely analyzed. So it's a matter of thinking, well, what new work can I do with that? And in particular, what's become an important option, which requires technical skills, is linking different types of data sets. So, you know, can you identify, can you geocold households and then link it to nightlights data? One are kind of satellite data on crop cover. So different types of ways that that's becoming a lot of people's innovation is combining different data sets. And that provides them with a way of compiling really, you know, time-consuming data sets that, so for example, if you're interested in the effect of an education reform on earnings, you might try and go back and say, well, what was the geographical distribution of this reform? Was it different in different districts of the country? Can you do a lot of time-consuming work to, you know, say that, well, we can have a difference in different strategy, because some districts, the expansion of primary education or secondary education occurred more intensively or earlier than in others, so you've got something to compare. So if you don't have access to the funds to do your own lab in the field or RCT, then it's a matter of, well, okay, how can I bring in additional data and information to use existing data sets? Okay, so then we try to introduce, we should more of try to introduce those other aspects of data sets that haven't been actually worked on to bring in the novelty in whatever research that we do. Okay, there is this question that someone wants to ask, and I think it'll be better to chip in because it has to do with the first question I asked. Andre, he's interested in, he's more interested in how to structure, most importantly, the first two sentences of your introduction. So put in the two sentences that try to put the paper, situate the paper in a larger context with research gaps. How would you catch the eye of the editor, put the two sentences? Early in the introduction, it's, and let me, I haven't perfected this myself, I can't always convince them, but it's a good thing early in the introduction, if you can cite a significant paper that's on the topic. It may not be for the same country, that in itself is not so important, but it should be a significant paper that has kind of influenced how people think about the issue. And you frame your contribution and say, okay, this paper did this with this data, but one of the things they haven't done is the follow-up. Now it might be, they've found this result for Mozambique, but that doesn't necessarily mean it would apply in Ghana. So what I'm going to do is see, does this effect also exist in Ghana, as an example. So you'd start with a significant paper. A significant paper wouldn't necessarily be published in the journal you're submitting to, but you should know that the journal you're submitting to has published papers on that topic, and you should be citing them. You should be citing them. For some editors, they like to see their journal cited, but that's not the main thing. The main thing is so that you know, is this journal actually interested in the topic? Because if the journal has never published a paper on the topic, they're not going to be interested in yours. Are there unlikely to be interested in yours, unless you can convince them that, hey, look, you've been neglecting this important topic. So being aware of, and typically being able to cite some papers in the journal you're submitting to, that address the topic, but that you build on. You don't want to do exactly what they did. Replication is okay for a starting point to kind of say, look, I can get their same result, but there's this problem. Now I'm going to do a bit more and contribute more. Andrew, I guess I'm fine with that. So, I mean, I don't see any question. I can see in the chat, it could be in the first two, it could be the first paragraph would cite that paper, probably, or cite a couple of papers that highlight that this is an important issue. A lot of people have worked on this, and ideally some of those papers are in the journal you're submitting to. And then you say, but there's a number of issues that these papers are not conclusive on. Then your second paragraph would really say why your context, the country, the data, are particularly suitable to address that gap, that question. So in a way, that's kind of the first two paragraphs. And then the rest of the couple of pages in the introduction really set out, this is how I go about it, explains the data in a bit more detail, previews the conclusions, what's the core implication, and kind of make the assumption that most people may not, you want somebody to read past the introduction. So you need to make the introduction interesting enough and informative. So when you're described, when you're briefly describing the data on the first page or the second paragraph, let's say, what years was it? How large is the sample? Is it nationally represented, et cetera, et cetera? You don't want people to be having to look through the papers and say, well, hang on, is this a panel? What years are the data for? Is it, you know, what type of data are it? That should be really clear in the opening paragraph, which ideally would also include a little bit of the country context or the policy context. Is it a big policy issue? Is the government concerned about it? So that you're kind of saying that it's an academic issue. People are writing papers. It's important in this context. And I've got good data to make a contribution. And that will attract people to read a bit further. Of course, they then may get to the econometric or analytical part and say, well, actually, the data aren't nearly as good as you claimed. And you can't really do what you want to do. So then Jacqueline also has a question wondering how to make trade research more inclusive. She suggests that she's asking if it should be through a gender lens. And she wants to know how to use custom straight data for development of monitoring mechanisms. Linking customs trade data to monetary issues is, it's not clear to me that that will be the best way to use it. Or trying to add in a gender dimension. People often think about gender and trade. There's nothing inherently gender about trade. But there may be gender implications of trade. Are the export sectors, do they tend to employ women or not? Are the export sectors that are growing? It might be there might be women involved in smallholder farming or in textiles sector in say countries like Bangladesh, in which case there's an indirect gender effect. But it depends on the composition of labor and maybe wage gaps and issues like that. So you may be using the trade data to give you information about what's happening to particular sectors. But then you need data on the employment or wage composition, gender composition of those sectors, if you want to say, look, here's how trade has gender effects. It's a bit like the literature on the poverty effects of trade. It's not inherently trade that affects poverty. It depends on, well, are poor people, producers or consumers of the goods that are afraid of? What's the effect of the trade on prices? And that's going to be the effect on the poor. So you've got to think through where the effect is happening that you're interested in. There's one other question on by Miquita saying that at times the findings appear to be obvious, but have not been studied in literature. So she's wondering, how would you convince an editor that this should go for a review rather than a dex reject because findings are a bit obvious? If the findings are obvious, then it's the methods that have to be novel. Because in a way, if you're saying the findings are obvious, what you really mean is that you're confirming what is already known. And a paper that's confirming that does nothing more than confirm what's already known and say, look, this was found in Botswana, and look, I found the exact same in Ghana. That in itself is not particularly interesting for one of the leading journals. That might be appropriate for a more local regional journal. But if you're able to say, well, look, I find the same result, but I'm also able to address a technical problem, an econometric issue. I've got a larger sample. I've got more recent data which shows that the effect is persisting. Or it could be, for example, let's say you were trying to do work on people's attitudes to policy advice on what safe, healthy behavior under COVID are attracting people to take vaccines rather than resist vaccines. Now, what you might be able to do in that context is say, well, in this country, if you had a country that had previous experience of HIV-AIDS, and there had been research on HIV-AIDS that show, well, actually, yes, when ART treatments were rolled out, that actually encouraged some people to engage in more risky behavior. So that when people think there's a dangerous health threat and there's nothing they can do about it, then they'll have a strong incentive to behave safely. But once they think there's a treatment there, then it might be a kind of a self-fulfilling damage because they think, all right, now there's a treatment. I don't have to be so safe. So you might be able to look at a different context and draw inferences from that and even use, you know, you might be able to say, well, actually, previous analysis of HIV-AIDS showed that some districts responded less well than other districts. Can you find the same if you've got data on COVID? That's just an example of how you can make something that may appear to be confirming something we know still be novel because you're saying, well, look, it exists in a different context. The other one thing I'd make is the other difficulty that people in the South have is for Northern researchers, they're usually in a lively university environment. They attend a lot of conferences. So their papers get a lot of exposure and comments and reactions from different people before they submit. That can be difficult if you're, you know, at the University of Ghana or the University of Dar es Salaam, there may be very few people that you can present your paper to and get some comments. You should still try. If you've got, if you know somebody, is there somebody in your department say who has a PhD from Europe or North America, ask them to have a look at it. Or if you know somebody, well, outside of your department or if you can get to conferences, present it and get comments before you submit. That's a good point. Yeah. We're just a minute away from the keynote. So I would thank you so much for this. And thanks, Kavina. And best of luck to you. Thanks a lot. And I hope the conference goes well. I'll try and join in some part of it later. Okay. Okay. Thanks to the audience for participation. I think it's kind of useful. Goodbye.