1 00:00:00,470 --> 00:00:09,110 Well, first, we welcome everyone to the second of these exciting SLC supported young lives methods workshops, 2 00:00:09,110 --> 00:00:18,380 and I'm just very briefly and we're all really familiar with young lives, but it really reflecting on your lives as a research project. 3 00:00:18,380 --> 00:00:22,430 It has been groundbreaking, I think, in four really significant ways. 4 00:00:22,430 --> 00:00:26,600 The first is in its intellectual leadership and its its research excellence, 5 00:00:26,600 --> 00:00:34,340 and that can be seen by the 400 papers published astonishing over the last 50 years. 6 00:00:34,340 --> 00:00:42,410 The second is then in its in its real commitment to combining methods to quantitative and qualitative methods, 7 00:00:42,410 --> 00:00:48,500 and those have had the opportunity to inform each other over a long period of time. 8 00:00:48,500 --> 00:00:53,240 The third has been in its commitments to policy engagement and impact, 9 00:00:53,240 --> 00:01:00,320 and to sustaining that in the four countries in which young lives as works, but also regionally. 10 00:01:00,320 --> 00:01:04,430 And perhaps the loss is perhaps what makes your lives most unusual. 11 00:01:04,430 --> 00:01:07,910 And that is its generosity as a research project. 12 00:01:07,910 --> 00:01:18,960 It is very rare to find a large project that releases all of its data open access, and that really does it rather than force, doesn't it? 13 00:01:18,960 --> 00:01:24,500 And this this really is what I think makes them stand out in this way. 14 00:01:24,500 --> 00:01:32,930 It's truly an altruistic research project, but they're also showing great generosity in sharing not only the data, 15 00:01:32,930 --> 00:01:41,240 but also some of the challenges and solutions of doing this kind of research over long periods of time in low middle income countries. 16 00:01:41,240 --> 00:01:50,450 And this workshop is an example of that generosity, and we really look forward to finding out more about the exciting set case study for it. 17 00:01:50,450 --> 00:01:54,050 We've got two speakers. We've got to marry Penny, 18 00:01:54,050 --> 00:02:03,680 who is a British physician originally trained at Gut's and college in a in a in a university which we shall not speak the name of. 19 00:02:03,680 --> 00:02:15,980 Not so far from here, a friend who she worked as a as a doctor in there in the NHS before being awarded a Wellcome Trust scholarship to study in Peru. 20 00:02:15,980 --> 00:02:18,380 And when I took to Madrid, 21 00:02:18,380 --> 00:02:28,160 we identified that Mary is it is a young lives teenager because she has been involved in the project for 19 and possibly 20 years now, 22 00:02:28,160 --> 00:02:38,200 and has a real in-depth understanding of of of the engagement and and young lives within Peru and leadership within that. 23 00:02:38,200 --> 00:02:50,660 Alan is the Young Lives child has been involved for a mere eight years in the project as young lives through and as a senior researcher at Grade. 24 00:02:50,660 --> 00:02:55,690 And he's an economist by Background in Oxford, 25 00:02:55,690 --> 00:03:06,100 the university to which we shall name and interview and really interested in looking at social programmes and skills formation. 26 00:03:06,100 --> 00:03:09,730 So we're really excited to hear from both of them. 27 00:03:09,730 --> 00:03:18,610 What we'll do is we'll hold your questions until the end, but then we're wanting an excited and engaged discussion. 28 00:03:18,610 --> 00:03:29,600 The deputy flight over to Marion Alan. Well, thank you for for for for me here. 29 00:03:29,600 --> 00:03:37,490 I'm very glad to have this opportunity to discuss with the experience that we have been accumulating over the over the years, 30 00:03:37,490 --> 00:03:41,860 very onerous delays with many, you know, she's she's actually one of the founders of the study. 31 00:03:41,860 --> 00:03:53,480 As Lucy was saying, I joined the army in 2012, a long time ago, but that, you know, I, I realise that it has been, 32 00:03:53,480 --> 00:03:58,700 you know, a big opportunity for me also to be part of a team that already knew how to do this process. 33 00:03:58,700 --> 00:04:04,910 And, you know, I have been trying to help since then as well. 34 00:04:04,910 --> 00:04:08,490 So, yeah, I mean, I know that some of you already know a lot about the United States. 35 00:04:08,490 --> 00:04:17,210 I just went to a and I just wanted to put some ideas and thoughts that we have about our experience with the study. 36 00:04:17,210 --> 00:04:24,080 First, to give you a brief description of what we do as you know that you like to study. 37 00:04:24,080 --> 00:04:33,530 It has been tracking two bird cohorts over 15 years in four developing countries Ethiopia, India, Vietnam. 38 00:04:33,530 --> 00:04:46,100 Overall, the distress call a has because lot of understanding the causes and consequences of child poverty. 39 00:04:46,100 --> 00:04:53,120 So these are status. It's been done in four countries. But I mean, today in particular, we're going to be talking about what we do in Peru. 40 00:04:53,120 --> 00:05:00,170 However, I should mention that most of the design is shared with other countries, although in some cases it's specific to Peru. 41 00:05:00,170 --> 00:05:04,190 I we try to highlight when that is the case in the case of Peru. 42 00:05:04,190 --> 00:05:09,140 We are tracking two cohorts the the younger cohort, 43 00:05:09,140 --> 00:05:17,240 which is the one that we tracked since they were eight one approximately, and we started tracking that cohort in 2002. 44 00:05:17,240 --> 00:05:24,920 This is about 2000 children, and the last time we visited them, they were 15. 45 00:05:24,920 --> 00:05:28,160 And then we have the older cohort, which is smaller, 46 00:05:28,160 --> 00:05:35,330 is composed by an original sample of 700 children that we started tracking at the age of eight years, 47 00:05:35,330 --> 00:05:41,840 and they had 22 years the last time we visited them and both cohorts. 48 00:05:41,840 --> 00:05:50,480 I mean, the design is such that actually both cohorts are random samples from the same communities, 49 00:05:50,480 --> 00:06:02,870 and we chose 20 clusters at random using the map of poverty from from Peru for 2001. 50 00:06:02,870 --> 00:06:05,360 It's not actually nationally representative. 51 00:06:05,360 --> 00:06:14,510 I mean, the idea was to for the sample to be proposed so that the the the richest five percent was excluded from the universe. 52 00:06:14,510 --> 00:06:21,560 And then from the remainder, we randomly sampled 20 clusters and then within each clusters, 53 00:06:21,560 --> 00:06:27,080 we randomly sample a point to start to search for households. 54 00:06:27,080 --> 00:06:30,680 So therefore, what happened in the wrong one? 55 00:06:30,680 --> 00:06:47,270 And since then, we have visited these two cohorts e on in fact, in four locations in 2006, 2009, 2013 and 2016 with to continue. 56 00:06:47,270 --> 00:06:53,120 And, you know, over the night, it's been 15 years now, so obviously we have the last part of the sample. 57 00:06:53,120 --> 00:07:01,540 I'm giving you some numbers there and I'll get back to the point of of attrition, which has been relatively low for this type of study. 58 00:07:01,540 --> 00:07:11,360 And this diagram shows how much better the design of the studies of the day in the first row would you consider the older cohort, 59 00:07:11,360 --> 00:07:16,250 the ages they're starting from age eight all the way to 22, 60 00:07:16,250 --> 00:07:21,650 and then the second row is the younger cohort starting from each one all the way to age 15. 61 00:07:21,650 --> 00:07:27,410 And something interesting that is not random is that you can see that they are comparable. 62 00:07:27,410 --> 00:07:39,260 So you can compare the younger cohort at age eight with the older cohort the same age, but with data that was collected seven years before. 63 00:07:39,260 --> 00:07:45,020 So this has been actually very useful for us to engage in conversation with policymakers 64 00:07:45,020 --> 00:07:49,770 in Peru because we can actually say something about what's improving or not. 65 00:07:49,770 --> 00:08:01,010 For four children in Peru. But also, I mean, I'm talking about the quantitative side of the study, but in parallel we have. 66 00:08:01,010 --> 00:08:11,720 We started in 2007, a qualitative study that is composed of honest at some point within each cohort, a small sample. 67 00:08:11,720 --> 00:08:15,140 And for this sample of children, we have a starting quality. 68 00:08:15,140 --> 00:08:25,300 We collected information through interviews and focus groups with it with a small group of young children with these statistics in 2007 and. 69 00:08:25,300 --> 00:08:38,200 The last round was 2014, and you know, is the good thing is that we have an interdisciplinary team, so I'm an economist, 70 00:08:38,200 --> 00:08:44,780 but we also have a question of psychologists are part of the team of anthropologists in Sicily working with us. 71 00:08:44,780 --> 00:08:51,130 So we've been dividing the work but trying to think carefully that you know this, this study as a whole is just one. 72 00:08:51,130 --> 00:08:57,340 But it has these two components a quantitative and qualitative side. 73 00:08:57,340 --> 00:09:02,290 This is just an example of how the original sample was distributed. 74 00:09:02,290 --> 00:09:11,530 I said the original sample because of migration is no longer where the sample is located, but you can see 20 red dots in the map. 75 00:09:11,530 --> 00:09:20,110 Those are the 20 districts that were randomly sample in wrong one in particular because see that there are three red dots in Lima. 76 00:09:20,110 --> 00:09:31,540 And because Lima concentrates one-third of the population, they had a higher chance to be selected by a by a random method. 77 00:09:31,540 --> 00:09:37,420 So originally we were, we had to go to these 20 clusters located, 78 00:09:37,420 --> 00:09:44,260 which basically is the centre has to say we have to go to 20 districts in 12 regions in the country. 79 00:09:44,260 --> 00:09:47,470 But in the last visit, although I don't have a map to show you, 80 00:09:47,470 --> 00:09:53,110 but I can tell you that we have had to visit more than 200 districts in it in the 24 regions. 81 00:09:53,110 --> 00:09:59,260 So that in the the rate of internal migration is very high in imperial. 82 00:09:59,260 --> 00:10:05,530 And you know, this is what this has been one of the challenges for us to keep up with that. 83 00:10:05,530 --> 00:10:12,860 I'll tell you more about that in a moment. I also wanted to mention something about the instruments that we use. 84 00:10:12,860 --> 00:10:19,540 In this case. I want to focus on the quantitative survey because I am in charge specifically of that component with with many. 85 00:10:19,540 --> 00:10:28,780 So we have a household questionnaire, which we've done since the beginning and is the caregiver of the child that typically answers the questions. 86 00:10:28,780 --> 00:10:33,910 We also collected the anthropometric for the for the for the children, 87 00:10:33,910 --> 00:10:41,860 for basically what we call the index, the index children, the child that was selected within the household. 88 00:10:41,860 --> 00:10:44,470 We since 8:00, since the age of five. 89 00:10:44,470 --> 00:10:55,060 We also collect cognitive achievement information through a that in that first visit, we used to be with a picture of capillary test. 90 00:10:55,060 --> 00:11:01,690 And then after that, we started introducing other tests to measure in comprehension of mathematics. 91 00:11:01,690 --> 00:11:07,900 Since the age of eight, there is also an individual question there used to be a child specific, 92 00:11:07,900 --> 00:11:15,650 but now you know there are so we can no longer call them children and and use that instrument to collect information about many aspects. 93 00:11:15,650 --> 00:11:27,820 I mean, it was it was mainly about the school and the school life in the beginning and aspects of especially such emotional dimensions. 94 00:11:27,820 --> 00:11:39,670 Then in some educational history, so we can try to get education history about each year, what they were doing and they went to the. 95 00:11:39,670 --> 00:11:44,500 Even since the beginning, we have also been collecting information about child work and then eventually, why just work? 96 00:11:44,500 --> 00:11:49,900 Because there now there are no others, especially the other cohort. 97 00:11:49,900 --> 00:11:57,670 We also have the self-administered questionnaire, which he saw, which was introduced first for the older cohort because of the patient. 98 00:11:57,670 --> 00:12:02,560 We use it for the first time in the last visit to a younger cohort, 99 00:12:02,560 --> 00:12:14,920 and we use that instrument to measure aspects related to risky behaviours and sexual behaviours, a knowledge of sexual information. 100 00:12:14,920 --> 00:12:20,140 There are some questions about a bit of violence as well. 101 00:12:20,140 --> 00:12:27,460 And we also have a community questionnaire, which is basically something that we collect to learn more about the districts. 102 00:12:27,460 --> 00:12:31,330 So that's those are the components of the quantitative survey. 103 00:12:31,330 --> 00:12:40,270 One thing I want to I wanted to highlight has to be related to the attrition rates of the study, so it's relatively low. 104 00:12:40,270 --> 00:12:48,730 What I am doing in this graph is to compare the cohorts with other longitudinal studies that 105 00:12:48,730 --> 00:12:56,230 are similar in that they have been following our cohorts since birth for a very long time. 106 00:12:56,230 --> 00:13:02,800 There might be other stuff, but these are those that are closest in design to to our study. 107 00:13:02,800 --> 00:13:08,380 And what you can see in the in the red bars, one of them, one of those is a little sample. 108 00:13:08,380 --> 00:13:13,120 So we have to have a higher graduation rates than some of the other countries. 109 00:13:13,120 --> 00:13:19,000 But overall, it's much smaller compared to two other studies for that for the younger cohort. 110 00:13:19,000 --> 00:13:23,890 The red bars and the green bars show you'll see better information for the older cohort. 111 00:13:23,890 --> 00:13:29,800 So this is this from my licence abortion rights. So it means like, for instance, every year we lose one percent of a sample. 112 00:13:29,800 --> 00:13:34,960 So if by 15 years lost that 15 15 percent approximately. 113 00:13:34,960 --> 00:13:40,300 But but that's lower than than what we actually may have expected. 114 00:13:40,300 --> 00:13:48,400 And what we want to talk a little bit about on this occasion is, you know what, what we have been doing to keep attrition low. 115 00:13:48,400 --> 00:13:52,930 So there have been a number of strategies. 116 00:13:52,930 --> 00:14:01,390 So first, we have what I call the main strategies, what we do with respect that is there by design to reduce attrition while. 117 00:14:01,390 --> 00:14:10,600 One thing we do is we do a tracking of the whole sample one year before each round. 118 00:14:10,600 --> 00:14:17,270 We I mean, basically, it's a physical visit that we look to some this to basically make sure that they are still there. 119 00:14:17,270 --> 00:14:26,360 But if they moved, we combine that with with the contact information we have from from from these families. 120 00:14:26,360 --> 00:14:33,710 Well, I'll see you in some bye from their neighbours to contact them, so we won't know where they are, where they have moved and we will. 121 00:14:33,710 --> 00:14:37,640 I mean, our goal is to look for them. 122 00:14:37,640 --> 00:14:44,720 Across the country, no matter where, no matter where they are within the national borders for free for international migration, 123 00:14:44,720 --> 00:14:52,790 there's there's nothing we can do about it other than to verify that they're actually went abroad. 124 00:14:52,790 --> 00:15:02,780 And so that's one key part of the strategy for me has been to do this tracking exercise and to collect this contact information. 125 00:15:02,780 --> 00:15:11,840 But we also do a number of other things that we think have contributed to keep the data crescendo. 126 00:15:11,840 --> 00:15:20,830 But I want to give the body now to many to explain to us what these orders that they have been. 127 00:15:20,830 --> 00:15:27,220 We've had a series of different activities that are designed to make sure that we know where people 128 00:15:27,220 --> 00:15:33,370 are and you contact with them and that they feel they belong and they do actually feel they belong. 129 00:15:33,370 --> 00:15:37,530 When you go to visit them, they say for me as a Romanian. 130 00:15:37,530 --> 00:15:45,310 As a result, we have managed to create a feeling that you belong to something that that's what we on the on the list, actually. 131 00:15:45,310 --> 00:15:49,790 But we we have them in Sicily. 132 00:15:49,790 --> 00:15:51,680 They the institute. 133 00:15:51,680 --> 00:16:01,600 We have a very active ethics committee, which is approved by the government through this quite well developed state ethics committees. 134 00:16:01,600 --> 00:16:07,210 And it has to go through that as a government agency, which which approves them. 135 00:16:07,210 --> 00:16:12,470 And the ethics committee takes their life very seriously and takes that work very seriously. 136 00:16:12,470 --> 00:16:20,370 And so we have to have a form which is registered with this ethics committee and also get some reports. 137 00:16:20,370 --> 00:16:28,960 And we've also been very lucky in that we've had we've been able to have some of the same experienced field workers, 138 00:16:28,960 --> 00:16:35,410 but in other projects visiting houses and might not necessarily like our lives. 139 00:16:35,410 --> 00:16:41,170 And and despite the fact that it's like Typekit work every four years or so, 140 00:16:41,170 --> 00:16:48,650 we can able to always have some leads that even least half the people have been involved before, and that's very helpful. 141 00:16:48,650 --> 00:16:56,830 So when you go back to the community, you sort of be known. I think that these is very important because there's a lot of crime through. 142 00:16:56,830 --> 00:17:06,730 And so people are sort of a little bit more confident that this is somebody that they've at least some of the group they've known before. 143 00:17:06,730 --> 00:17:16,300 And so that's been that's been very helpful having same field work, some key staff that in particular, for instance, it's really important. 144 00:17:16,300 --> 00:17:21,280 There's a number of ethical cases, there's a number of problematic cases, 145 00:17:21,280 --> 00:17:28,060 and it's important that the field worker knocks on the door they can ask for, you know, the father is the father left. 146 00:17:28,060 --> 00:17:40,450 So there's a certain amount of being able to be quite careful about about the unfortunate. 147 00:17:40,450 --> 00:17:44,710 When we get back to the house, we've collected all this information. 148 00:17:44,710 --> 00:17:51,280 So people assume that, you know, you can remember what they told you the time before which, but you can't 60000. 149 00:17:51,280 --> 00:17:56,770 But at least you know that that you know, this particular household subtropical environment. 150 00:17:56,770 --> 00:17:58,990 So it's it's possible. That's very helpful. 151 00:17:58,990 --> 00:18:07,360 Two things are helpful for retention of the field workers and prior knowledge of of of these ethical cases, 152 00:18:07,360 --> 00:18:14,200 which allow us to display that at some point. We also have to work. 153 00:18:14,200 --> 00:18:21,340 We have. You can't just march into a community where you normally have to have some kind 154 00:18:21,340 --> 00:18:26,950 of entrance through the local so you don't need to know what you're up to. 155 00:18:26,950 --> 00:18:30,700 And that's what of course, having having. 156 00:18:30,700 --> 00:18:37,600 Being registered and having had the ethical committee help somebody come up to you and ask you, what are you doing here, 157 00:18:37,600 --> 00:18:46,510 particularly political parties, but that's not what, particularly in the period in which we have national problems with Senator Nelson. 158 00:18:46,510 --> 00:18:50,620 There's so so we also present to district authorities, 159 00:18:50,620 --> 00:18:53,890 know what we're doing and generally get that help and also get some information 160 00:18:53,890 --> 00:19:01,990 from them because they also participate in in the survey of the community that we. 161 00:19:01,990 --> 00:19:06,490 It's also quite useful because there are other Indians in this pair, 162 00:19:06,490 --> 00:19:13,060 and this means that for lots of different people working in communities, it's also we don't even know. 163 00:19:13,060 --> 00:19:19,630 It's useful that they also know who we are. So we're not competing with them because they impose their capital. 164 00:19:19,630 --> 00:19:30,130 And so it's sort of if we maintain good relations with the other NGO, we're not we're not doing the same thing as they are competing with them. 165 00:19:30,130 --> 00:19:34,250 We have then of course, you know, farmers do expect something back. 166 00:19:34,250 --> 00:19:38,380 They've given all this information. It takes quite a long time to be up to date. 167 00:19:38,380 --> 00:19:40,360 Maybe related to ours, 168 00:19:40,360 --> 00:19:48,760 but it's more often four hours and you're sitting there in the household and everybody else is looking through that sort of thing. 169 00:19:48,760 --> 00:19:53,200 So we do give we have to give different information. 170 00:19:53,200 --> 00:20:00,370 We give some nutritional advice. We try not to. I mean, we we give sort of general, 171 00:20:00,370 --> 00:20:04,870 we don't give particularly specific advice because we're not trying to find the balance 172 00:20:04,870 --> 00:20:10,000 between being ethical and obviously not letting somebody stall and on the other hand, 173 00:20:10,000 --> 00:20:14,590 changing the lives of these children. So we sometimes seek a balance between these. 174 00:20:14,590 --> 00:20:18,730 And obviously, if we do find things that are absolutely urgent, we can. 175 00:20:18,730 --> 00:20:25,210 Yes, we do give advice to the doctor or whatever. 176 00:20:25,210 --> 00:20:32,140 So we now know so we have so we give them the results, for instance, of the heights of weight. 177 00:20:32,140 --> 00:20:42,130 And so the results of the anthropometric and and we've now got a sort of consultation guide, 178 00:20:42,130 --> 00:20:43,970 which is just this business that we haven't done that before. 179 00:20:43,970 --> 00:20:55,340 And this is this round is that is that the thing that we see from two to three nights one this time is I forgot the others, 180 00:20:55,340 --> 00:20:59,050 but this tells you where you can get help from, from within. 181 00:20:59,050 --> 00:21:06,490 You have such a nice, wonderful experience of like a booklet because you have to improve it. 182 00:21:06,490 --> 00:21:13,270 So we give it, you know, so we give them advice where they could get help for different things. And then this is something that we give back. 183 00:21:13,270 --> 00:21:23,000 And the other thing which has been very popular is that we take a photo of the child, child of the mother of a couple of photos. 184 00:21:23,000 --> 00:21:30,190 Then we give the secretary back, say to the family, at least we select one and get back to the family. 185 00:21:30,190 --> 00:21:35,500 And if you go back into the houses, you sometimes they have relatives young photos that's been. 186 00:21:35,500 --> 00:21:38,840 I think that was a very good move because of some of the things I did. 187 00:21:38,840 --> 00:21:43,270 It was a good moment because then that has been very popular because not everybody 188 00:21:43,270 --> 00:21:48,040 can afford to get around to having certain they that children of different ages. 189 00:21:48,040 --> 00:21:54,610 And so I think that was that was a very good idea. We don't pay. 190 00:21:54,610 --> 00:21:59,230 We might get a small gift, but are the children that have been buried and drowned, 191 00:21:59,230 --> 00:22:09,340 but we don't actually pay them for this because we think that would be one of the ethics committee, as you would believe, would allow us to do that. 192 00:22:09,340 --> 00:22:17,990 Not in this context. They do allow in some context. So I think those are the the main strategies, and I have to say, 193 00:22:17,990 --> 00:22:22,520 I think this should probably this is terribly satisfactory because you do because on the whole, 194 00:22:22,520 --> 00:22:28,330 100 percent of the whole people are quite pleased to see that again despite the pandemic. 195 00:22:28,330 --> 00:22:36,280 I think one of the things perhaps we have mentioned have mentioned strategy is that I think people feel kind of like important. 196 00:22:36,280 --> 00:22:36,550 I mean, 197 00:22:36,550 --> 00:22:46,450 they do feel that they're contributing there so that we know that what we are told us is being taken into account has been absolutely unbelievable 198 00:22:46,450 --> 00:22:55,150 in terms of the production of materials and papers and come out and said they're looking at probably people now we get up and implement, 199 00:22:55,150 --> 00:23:02,560 I'm not sure they do. And so I think that, you know, they feel they belong to that. 200 00:23:02,560 --> 00:23:07,890 So I think that the millennials sort of force that feeling of belonging. 201 00:23:07,890 --> 00:23:17,870 And I think it's also quite helpful to me and I know some people I've got to ask and don't agree with that. 202 00:23:17,870 --> 00:23:23,870 But I think the efficiencies being driven by really considering that we do ask quite a lot of. 203 00:23:23,870 --> 00:23:30,830 I also think that have to say, I think in Peru, families are very tolerant and that are traditionally rather not. 204 00:23:30,830 --> 00:23:34,450 So that that helps us. 205 00:23:34,450 --> 00:23:41,560 When I said that, we've established a relationship with the family, the people going through this without forgetting that it was not. 206 00:23:41,560 --> 00:23:45,110 Yeah. So in a way that makes it feel important, we feel. 207 00:23:45,110 --> 00:23:48,520 But brothers also, I have to say, very flexible. I mean, if I have to go with something, 208 00:23:48,520 --> 00:24:00,240 if I have to go and have a night or they have to go at 6:00 in the morning and do this and that they have to be there accompanied by someone. 209 00:24:00,240 --> 00:24:05,110 But it's there's a lot of them. There's a lot of emphasis on whatever it takes. 210 00:24:05,110 --> 00:24:09,580 We will we will try and get this interview done at the time. 211 00:24:09,580 --> 00:24:14,590 That's convenient for the family. The other thing is that, yes, I also agree other things. 212 00:24:14,590 --> 00:24:18,740 I think that we do ask about things which families consider important. 213 00:24:18,740 --> 00:24:24,830 So this is not just survey about something that's utterly unbearable, different to are. 214 00:24:24,830 --> 00:24:30,530 These are asking questions. They also consider important in the lives that we have. 215 00:24:30,530 --> 00:24:39,790 And as mentioned, the self-administered questionnaire, which, you know, the child goes off home building without anybody else looking. 216 00:24:39,790 --> 00:24:45,320 And that's and I think that's that's that's that's actually very interesting. 217 00:24:45,320 --> 00:24:52,480 And increasingly so as children, the older, we've never worked on the project, 218 00:24:52,480 --> 00:24:58,970 which has produced so many publications and presentations nationally and internationally. 219 00:24:58,970 --> 00:25:09,070 I think I think Joe and started with another school had that feeling changed to Oxford, and I think all people be involved, 220 00:25:09,070 --> 00:25:17,260 all the rounds of cuts that have been incredible in terms of getting information out there. 221 00:25:17,260 --> 00:25:25,870 And I think if you go back to, I think that that really fundamentally not all of them by all means. 222 00:25:25,870 --> 00:25:30,580 I mean, some people are a bit reluctant to have to come out at different times and they do, 223 00:25:30,580 --> 00:25:33,610 but they usually have very many people who actually refuse it. 224 00:25:33,610 --> 00:25:37,680 More people, I think we lost because we can't find that we've migrated things like that. 225 00:25:37,680 --> 00:25:44,900 I think they do feel that they're kind of contributing somebody special because their young lives. 226 00:25:44,900 --> 00:25:48,280 So I think that's what we've learnt from. 227 00:25:48,280 --> 00:26:03,610 Well, certainly what I've learnt from from from doing this longitudinal study that's been that this kind of evolved into a life that's. 228 00:26:03,610 --> 00:26:11,710 Really, do this really is now open to questions and thoughts and comments or discussion. 229 00:26:11,710 --> 00:26:19,360 I have a first one about the photographs and will bring a brilliant idea and a lovely idea you can see for the families. 230 00:26:19,360 --> 00:26:24,340 I wonder whether they might also have a use in record in working out that you've got the right person, 231 00:26:24,340 --> 00:26:28,960 the right child, to use them for identification purposes as well. 232 00:26:28,960 --> 00:26:38,950 When I join, who are they using this strategy? But we use it, especially in the in the first couple of rounds and the second round because, 233 00:26:38,950 --> 00:26:42,400 you know, a bit all has improved a lot in the last few years. 234 00:26:42,400 --> 00:26:48,370 But in many cases it's not that the House has a number of sites. 235 00:26:48,370 --> 00:26:53,590 So we know the street. Then with that one, we are in the community. 236 00:26:53,590 --> 00:26:55,180 We know the name of the street with. 237 00:26:55,180 --> 00:27:03,460 The picture is very useful to find the house in order to make sure that they were looking for the same for the same child. 238 00:27:03,460 --> 00:27:08,710 But I will say that over time, that has become less important, 239 00:27:08,710 --> 00:27:14,620 and what has become really important is the fact that the children are actually expecting the picture. 240 00:27:14,620 --> 00:27:22,180 I mean, in fact, the Sun-Times has contacted us over time, saying, Look, I lost the photo that you took in. 241 00:27:22,180 --> 00:27:30,820 Round one is the only photocopy with my with my grandmother. So can you send me a copy of questions from them? 242 00:27:30,820 --> 00:27:38,750 I think it's really important and interesting to also reflect on the changes that you experienced over the course. 243 00:27:38,750 --> 00:27:49,900 I mean, I don't know if the changes have been so dramatic and improves they have in other countries, but the rise of internet use mobile phones, 244 00:27:49,900 --> 00:27:57,250 and I know you've talked in the past about the use of Facebook to be in touch with the children and so on. 245 00:27:57,250 --> 00:28:04,720 I think we really could reflect on that, but also what the introduction of handheld computers has meant, 246 00:28:04,720 --> 00:28:11,650 because when I started, it was all done with paper questionnaires. 247 00:28:11,650 --> 00:28:22,000 So indeed, we introduced a tablet officially for the full sample in in our fourth visit in 2014. 248 00:28:22,000 --> 00:28:26,610 But in 2009 around, we are ready to introduce them. 249 00:28:26,610 --> 00:28:30,490 But you know, I mean, this was before I joined, 250 00:28:30,490 --> 00:28:39,190 but the main point was that we were not sure about the implications of introducing this technology and in particular was 251 00:28:39,190 --> 00:28:48,070 going to be the impact on the on the on the answers when you move from a paper based questionnaire to a tablet base, 252 00:28:48,070 --> 00:28:56,530 you know. But basically, what we always try to do is to keep the methodology a constant over time so that we don't lose comparability. 253 00:28:56,530 --> 00:29:04,150 So taking this culture into consideration is the protein we know with with her 254 00:29:04,150 --> 00:29:09,160 former colleague from Oxford came up with a design for the paint chart back then. 255 00:29:09,160 --> 00:29:16,120 How we call a work on the idea of randomising so that half of the interviews were 256 00:29:16,120 --> 00:29:21,430 done with the tablet and the other half with with a paper based questionnaire. 257 00:29:21,430 --> 00:29:31,210 Naturally, there is a Polish paper with the results. Some of them are that, you know, the overall quality has not been affected if the information. 258 00:29:31,210 --> 00:29:36,570 But is faster is faster to have to look to run an interview with a tablet. 259 00:29:36,570 --> 00:29:51,470 There might be a little bit of a loss in terms of information. E when, when, when the when the questions are cut, many items, but actually does. 260 00:29:51,470 --> 00:29:53,030 That works better with a computer, 261 00:29:53,030 --> 00:30:00,770 because it computers you can block very quickly start tonight when you know that the batteries are essentially no no. 262 00:30:00,770 --> 00:30:06,050 It's easier to to to plug things in with the tablet. 263 00:30:06,050 --> 00:30:12,170 So overall, I think there have been an improvement in the quality of information that we collect. 264 00:30:12,170 --> 00:30:21,860 And then you round four as a consequence of this. We introduced the use of tablets for the full sample and we had two challenges there, I think. 265 00:30:21,860 --> 00:30:29,990 One was that some of our enumerators, which are very good, they didn't know how to handle properly a tablet, 266 00:30:29,990 --> 00:30:35,150 especially that touchscreen tablet, because you need a touchscreen tablet to do this quickly in the field. 267 00:30:35,150 --> 00:30:38,510 So we spent some time utterly. I mean, mentally, they go there. 268 00:30:38,510 --> 00:30:44,450 I mean, I have to be it. By 2014, I myself had not used a touchscreen at that very moment. 269 00:30:44,450 --> 00:30:55,070 And the same with our narrators. We spent some additional weeks of training just to make sure that we we were all at the same level. 270 00:30:55,070 --> 00:31:03,650 Maybe after the book probably may have lost a couple of enumerators just for that reason because they never got used to the tablet, to be honest. 271 00:31:03,650 --> 00:31:11,420 And another concern also had to do with security. Once we were in the field, you know that this, especially in round four. 272 00:31:11,420 --> 00:31:15,620 I mean, nowadays, laptops, tablets are very cheap. 273 00:31:15,620 --> 00:31:23,120 They were not that cheap in 2014, especially given that we had to run a software there. 274 00:31:23,120 --> 00:31:29,270 So actually, we went to a field in Peru with tablets that list that cost a hundred dollars back then. 275 00:31:29,270 --> 00:31:36,570 That's a lot of money for a tablet. We were concerned about security. We have only lost one tablet. 276 00:31:36,570 --> 00:31:47,750 In the first week of training of the of the of the of the fieldwork, which made us very concerned, but luckily was the only tally we lost, 277 00:31:47,750 --> 00:31:52,560 we got the stolen, to be honest, and it was because actually there had been a breach in the protocol. 278 00:31:52,560 --> 00:32:00,690 I mean, in part because of this, because of that, we made the protocol very tight so that, you know, 279 00:32:00,690 --> 00:32:07,470 if you are if you have an interview at nine a.m., you need to stay alert when you're done. 280 00:32:07,470 --> 00:32:13,080 So we sent a taxi or our driver to to pick you up. 281 00:32:13,080 --> 00:32:19,810 You know, you shouldn't you shouldn't go by yourself with the tablet, but you know these are. 282 00:32:19,810 --> 00:32:28,960 Protocols that we already had, but they became important not only now for the for the for our enumerators security, but also for for the Quitman. 283 00:32:28,960 --> 00:32:31,150 There was also another protocol. 284 00:32:31,150 --> 00:32:40,060 I mean, obviously the important thing was the numerator, like if someone comes with a gun and tries to take the it's obviously you give the tablet. 285 00:32:40,060 --> 00:32:49,690 But we are also concerned about information. So we have this protocol that once you complete an interview in Housefull before you leave the household, 286 00:32:49,690 --> 00:32:54,220 you safe everything in your USB and you put it in your pocket. 287 00:32:54,220 --> 00:33:01,480 So that in case for some reason you lose either because you get a stolen or simply your father's and you, you lose the tablet. 288 00:33:01,480 --> 00:33:07,720 We still keep the information something very interesting the happening around 4:00 that we 289 00:33:07,720 --> 00:33:12,070 were only allowed to leave because we introduce a tablet that is that for the first time, 290 00:33:12,070 --> 00:33:18,220 we introduce a tablet based test to measure of protective function. 291 00:33:18,220 --> 00:33:25,720 So this was a very interesting experience because, um, for many reasons, 292 00:33:25,720 --> 00:33:30,700 one of the reasons is actually some of the children have not used that outlet before. 293 00:33:30,700 --> 00:33:38,490 So it was so we had to. We had a big pilot to make sure he was on at work. 294 00:33:38,490 --> 00:33:45,040 You know, I'm talking about a few years now. Nowadays, things have improved in Peru. 295 00:33:45,040 --> 00:33:51,240 There there's been a lot of migration to urban areas, so maybe the portion of his sample, the you settled it. 296 00:33:51,240 --> 00:33:54,720 Obviously, they have youth. We ask them to use them. 297 00:33:54,720 --> 00:34:02,160 But, you know, are in their general life. I think now they have much more access to computers, to the internet, to smartphones. 298 00:34:02,160 --> 00:34:08,200 We have had to adapt our our position that is sometimes adding that question about use of social networks. 299 00:34:08,200 --> 00:34:17,430 But yeah, I mean, that's one of the I think part of the challenge is also to keep the question updated to the changes in the technology. 300 00:34:17,430 --> 00:34:25,760 Another challenge to keep it. As short as possible, I think that there's always a lot of other things you'd like to ask. 301 00:34:25,760 --> 00:34:31,930 There's already long who it's, it's long as we we see this part that we need to repeat, we want to repeat down. 302 00:34:31,930 --> 00:34:36,890 And between that and this of new information, it is a long it's a long interview. 303 00:34:36,890 --> 00:34:41,160 But but we have to be careful not to make it even longer. 304 00:34:41,160 --> 00:34:45,850 No, this is just a challenge. That is right. We'll take you back. 305 00:34:45,850 --> 00:34:59,570 I just had something follow up on that. Switching from paper to computer surveys in terms of the paper, do you suppose the earlier hardcopy surveys? 306 00:34:59,570 --> 00:35:08,710 Well, we started for a few years, but after a given time, we have to not shred them right? 307 00:35:08,710 --> 00:35:12,910 All right. Yes. Because in terms of what might get lost when you make that change, 308 00:35:12,910 --> 00:35:19,580 I know some studies in the UK positive studies have looked at marginalia in surveys. 309 00:35:19,580 --> 00:35:26,440 So sort of the power of data, the kind of writing in the margins that some workers might do about a particular 310 00:35:26,440 --> 00:35:32,470 householder or child that doesn't really get factored into the data that's entered. 311 00:35:32,470 --> 00:35:37,890 But that's another sort of main context data that that may or may not. 312 00:35:37,890 --> 00:35:39,980 Well, another advantage actually to and that's what I remember. 313 00:35:39,980 --> 00:35:47,860 I was rung up by somebody from way off somewhere and they were they were in some companies. 314 00:35:47,860 --> 00:35:53,410 It's a problem of litigation or there was something that was really important to know the child. 315 00:35:53,410 --> 00:35:59,500 They were now adolescent when they were born and what their name was and what their father's name was. 316 00:35:59,500 --> 00:36:03,910 And and we so we went back. We just don't have the paperwork. 317 00:36:03,910 --> 00:36:11,400 And so I went back and found there the. Agreement form. 318 00:36:11,400 --> 00:36:17,240 You know, when they agreed to participate in the project and was able to scan that and said we offered to them first. 319 00:36:17,240 --> 00:36:24,410 So I think it worked. So sometimes it's an advantage and unexpected advantage. 320 00:36:24,410 --> 00:36:27,170 Yeah, they were going to do something really important. 321 00:36:27,170 --> 00:36:34,390 And in fact, just to this point, even though it's true that we we have to destroy the physical copies of the question after a few years. 322 00:36:34,390 --> 00:36:46,140 But we have a digital copies, not of everything, but of the key things that are protected in that in the server that to a project. 323 00:36:46,140 --> 00:36:50,620 Yeah. OK. So first of all, thank you so much for this interesting presentation. 324 00:36:50,620 --> 00:37:02,560 Just wonder if you have any information or feedback from the families in this study, how their involvement in the study had affect their lives. 325 00:37:02,560 --> 00:37:05,950 And you have very you mentioned very interesting boy, 326 00:37:05,950 --> 00:37:15,790 that many of the families or maybe their parents said that they feel that they feel special, they feel that they contribute to the study. 327 00:37:15,790 --> 00:37:24,850 And as a social worker, I see this as an act of empowerment for these were never and never been for me personally. 328 00:37:24,850 --> 00:37:30,250 So I would like to listen. You have any reflections? 329 00:37:30,250 --> 00:37:33,910 Well, I think that's I think that's a very good point, actually. 330 00:37:33,910 --> 00:37:41,560 I mean, it is difficult to know whether we do have an impact on the family, and I don't know that we really sort of empower them greatly. 331 00:37:41,560 --> 00:37:47,630 But I mean, we haven't been able to measure that, obviously, because we don't have that control. 332 00:37:47,630 --> 00:37:55,330 But it's a good point, I think, because it's only once every four years and there's a lot of other surveys going on. 333 00:37:55,330 --> 00:38:00,040 I mean, you know, through a number of national surveys and things, it's not. 334 00:38:00,040 --> 00:38:08,020 It's not. This is a bit more in depth and a rather slightly so it's a rather different survey, but it's still happening and it's a good point. 335 00:38:08,020 --> 00:38:18,490 And it's it's difficult to measure, but I think we probably don't, but it's difficult to say because we don't have a control. 336 00:38:18,490 --> 00:38:22,420 We did put it in the source of the applications to do that, to create what was it, 337 00:38:22,420 --> 00:38:29,180 a match sample in Peru and her cohort and to look at their outcomes compared to those who've been in the study. 338 00:38:29,180 --> 00:38:36,270 Unfortunately, it wasn't funded this from. But it's that question is key and it comes up again and again. 339 00:38:36,270 --> 00:38:40,600 But it's also important to highlight that we do have other concerns. 340 00:38:40,600 --> 00:38:50,980 The exact opposite, actually, which is because we're only selecting one child in each household and we have actually sort of the qualitative research. 341 00:38:50,980 --> 00:38:57,940 People have indicated that they think that that child was selected for some particular reason. 342 00:38:57,940 --> 00:39:03,190 We actually worry about the possibility that that child may be favoured in some way within the household. 343 00:39:03,190 --> 00:39:06,610 There may be more investment in that child as compared to the others. 344 00:39:06,610 --> 00:39:12,310 There's no way of knowing, but we've just got little anecdotes with comments or questions that have made us think, 345 00:39:12,310 --> 00:39:18,340 Gosh, is that an indication that we've actually created some kind of inequality within the household? 346 00:39:18,340 --> 00:39:26,770 I hope to goodness it's not true, but we don't know. I think the learning, though, is that research is always an intervention. 347 00:39:26,770 --> 00:39:38,260 And I remember a young lives was was evaluated with I funding some years ago by a group of statisticians, 348 00:39:38,260 --> 00:39:45,700 and they actually talked about the contamination of survey data by using qualitative data, 349 00:39:45,700 --> 00:39:51,820 qualitative research because you develop more intense relationships with quantitative research. 350 00:39:51,820 --> 00:39:58,100 And they were really worried that the research would actually have a negative impact on the survey data. 351 00:39:58,100 --> 00:40:03,970 So there are lots of ways in which we would love to be able to understand those those dynamics better. 352 00:40:03,970 --> 00:40:12,550 And the funding would be needed, though, to do it. As you say now, we absolutely would need to control and let. 353 00:40:12,550 --> 00:40:20,470 Those statisticians could argue the opposite direction is welcome, but all that it audit it, improve the survey results. 354 00:40:20,470 --> 00:40:24,820 You could, I mean, maybe just to just to at least not. 355 00:40:24,820 --> 00:40:28,070 I mean, I I mean, we might be disagreeing with many of these. 356 00:40:28,070 --> 00:40:33,220 I I do think we have we must have some type of impact. 357 00:40:33,220 --> 00:40:39,670 I mean, I mean, as Bill was pointing out, there is no such thing as a purely observational design. 358 00:40:39,670 --> 00:40:44,080 I mean, you are there, you're obviously something new or unique. 359 00:40:44,080 --> 00:40:50,530 And you know, they know that we're coming again every three years. They might feel particularly special because they are. 360 00:40:50,530 --> 00:40:55,810 Many of the millennials are the only full time Spanish in particular. 361 00:40:55,810 --> 00:41:02,760 You know, we have this consultation date. Which provides access to information that is publicly available, 362 00:41:02,760 --> 00:41:11,700 but that some of the families might not have the resources to get this information because of the inefficiency of the government. 363 00:41:11,700 --> 00:41:15,600 So we have provided information that may be useful for them that I love myself. 364 00:41:15,600 --> 00:41:21,270 But then there is a question about what we ask about in Britain. 365 00:41:21,270 --> 00:41:24,330 We have a question about educational aspirations. 366 00:41:24,330 --> 00:41:32,130 So we've been asking for for many years to this, to this group of participants know if you don't have any constraint in life, 367 00:41:32,130 --> 00:41:35,400 what is the highest level of education you would like to achieve? 368 00:41:35,400 --> 00:41:44,640 You know, probably no one has asked this question before to them, so that surely must be reflected into, you know, maybe I should or more. 369 00:41:44,640 --> 00:41:53,170 I mean, so we indeed we we thought about this idea and we think there might be something about us having an impact. 370 00:41:53,170 --> 00:42:01,140 The interesting thing is that our sample is a random sample and that actually allow us to think about different ideas, 371 00:42:01,140 --> 00:42:05,340 which we have to reconstruct a control sample. 372 00:42:05,340 --> 00:42:15,930 One idea would be to obtain a new random sample from the original universe or do some kind of matching with communities are nearby there. 373 00:42:15,930 --> 00:42:21,480 The the large challenge there is that you have to be able to migration because we will need to 374 00:42:21,480 --> 00:42:26,820 go to these communities that are similar or are located closer to the regional communities. 375 00:42:26,820 --> 00:42:34,530 And we will need to look not only for adolescents that have the same age as our cohort as an adult. 376 00:42:34,530 --> 00:42:40,740 But we will need to search for those adolescents that were born there that migrated. 377 00:42:40,740 --> 00:42:47,700 I mean, we think this is feasible but is very expensive. And you know is it reminds us of an important question for us. 378 00:42:47,700 --> 00:42:58,160 If we have an impact, we would like to happen now. So then a lot of these in the future. 379 00:42:58,160 --> 00:43:05,670 Thank you so much for the wonderful presentation and congratulations on all the decades of work and you have done. 380 00:43:05,670 --> 00:43:14,360 And I was just curious about the design of the younger and older cohorts and the points you mentioned about, 381 00:43:14,360 --> 00:43:23,660 you know, the possibility of, for example, comparing an eight year old in the comfort zone of the risk of heart with somebody 382 00:43:23,660 --> 00:43:28,790 who ages into the same group that later covered maybe some at an older age. 383 00:43:28,790 --> 00:43:32,540 But you kind of how you think about this. 384 00:43:32,540 --> 00:43:40,640 I guess whether you're thinking, Oh, perhaps you have done this, you know, being able to achieve you detect certain level changes. 385 00:43:40,640 --> 00:43:48,080 And I guess also, you know, if we think about if we do believe that the measurement affects either how people respond or in fact, 386 00:43:48,080 --> 00:43:50,840 their experiences or both, 387 00:43:50,840 --> 00:43:56,480 whether that there might be of how you think about comparing somebody who's doing the story for the first time, for example, 388 00:43:56,480 --> 00:44:05,240 for somebody we think a few times already and they probably already have these papers, obviously necessary advisor and so on. 389 00:44:05,240 --> 00:44:13,730 And so we what we have, I mean, the the maybe the the the main, uh, 390 00:44:13,730 --> 00:44:21,680 a use we have made to this component of our design has been especially useful with it in our relationship with the policy makers 391 00:44:21,680 --> 00:44:29,780 because they are always interested in trying to understand what what has changed over time improve for for children and our listeners. 392 00:44:29,780 --> 00:44:34,910 So because of this design, we can say, you know, our child has the age of 15 now. 393 00:44:34,910 --> 00:44:38,540 If you compare him, compare them with children the same age. 394 00:44:38,540 --> 00:44:42,620 But seven years ago, you can see certain things that have improved. 395 00:44:42,620 --> 00:44:51,170 In the case of blue standing have reduced considerably the years of applications that then have also improved considerably. 396 00:44:51,170 --> 00:44:56,000 Cognitive achievement has improved, but slowly, maybe not enough. 397 00:44:56,000 --> 00:45:01,670 We can say that with the data. So this is why it's particularly useful for us. 398 00:45:01,670 --> 00:45:16,030 And also, as you're saying, for for for the studies, for analysis, it can become an obvious control group that maybe, maybe not that, 399 00:45:16,030 --> 00:45:19,340 not that what maybe not the other one group is not affected, 400 00:45:19,340 --> 00:45:24,530 but they're affected by the fact that different age so that you can actually do analysis with this. 401 00:45:24,530 --> 00:45:31,670 There's been some studies, but actually, that's the nice thing about the press, as you were mentioning to see that is is publicly available data. 402 00:45:31,670 --> 00:45:38,530 So this, you know, this can be done even if it has to be done by accident. 403 00:45:38,530 --> 00:45:42,490 I think one of the things that's really interesting about the into what comparison was, 404 00:45:42,490 --> 00:45:52,780 this was one of the ways in which we discovered the pace of increase of rates of obesity in adolescents wasn't bearing the way we did, 405 00:45:52,780 --> 00:46:01,030 where we're by the younger cohort as they were in adolescence, where the rates of obesity were much higher than they were. 406 00:46:01,030 --> 00:46:03,580 They have been able to cope with the same ages. 407 00:46:03,580 --> 00:46:10,960 And this is this is really opened up a whole new area of research because it's obviously becoming such an important consideration 408 00:46:10,960 --> 00:46:19,780 in so many countries where historically that the research was always focussed on undernutrition and stunting and so on. 409 00:46:19,780 --> 00:46:26,410 And now we're seeing this dramatic shift, particularly that shift in adolescents, which again, I think takes. 410 00:46:26,410 --> 00:46:32,500 Then it takes the whole framing of of the understanding of human development into a different 411 00:46:32,500 --> 00:46:37,270 age group where so much of it historically was thought to be about early childhood. 412 00:46:37,270 --> 00:46:42,190 We're now looking at the passage of what analysis we can really see what's happening in adolescence 413 00:46:42,190 --> 00:46:47,650 that we should be concerned about as a sort of second period of major change in human development. 414 00:46:47,650 --> 00:46:59,830 Major sort of transition point. Yeah, I think we've also been able to contribute to it with this without any intervention, 415 00:46:59,830 --> 00:47:02,950 able to see that some children who were not necessarily condemned, 416 00:47:02,950 --> 00:47:08,830 to be sure all their lives that actually were able to catch up a little bit later on, 417 00:47:08,830 --> 00:47:14,260 which was not kind of like a sort of belief that everything happened. 418 00:47:14,260 --> 00:47:19,330 I think in the first three years of life in California to be the end, and there is, 419 00:47:19,330 --> 00:47:27,210 you know, it's not surprising little bit more flexibility later to say. 420 00:47:27,210 --> 00:47:33,870 Hi. Well, I've got a security compliance team, I've got a couple of questions. 421 00:47:33,870 --> 00:47:38,400 He spoke a little bit about attrition to age came up in there, 422 00:47:38,400 --> 00:47:48,810 but were there any other factors and at any point where you was able to sort of ask other household members like why one of them had migrated? 423 00:47:48,810 --> 00:47:55,890 So treat it as an outcome because in some circumstances you might as actually something that was beneficial 424 00:47:55,890 --> 00:48:03,040 and at a certain point is kind of a labour opportunity in a different country or something like that. 425 00:48:03,040 --> 00:48:08,960 And also, I don't actually know the origins of the young life study, 426 00:48:08,960 --> 00:48:14,970 so I don't know whether you got funding that all of 15 years follow up at the beginning. 427 00:48:14,970 --> 00:48:23,650 And did you at any point consider increasing your sample size at a given point to study certain rare outcomes? 428 00:48:23,650 --> 00:48:35,840 It sort of became more interesting. All of policy makers started to sort of really kind of become interested in saying, 429 00:48:35,840 --> 00:48:40,710 maybe you want to offer some relief of playing a couple of hundred ActionScript. 430 00:48:40,710 --> 00:48:53,140 Yeah, I mean the. The stuff you got from this originally was defeat, defeat originally. 431 00:48:53,140 --> 00:49:00,940 We have we have moved into different funders. We developed a different fund, a different tone. 432 00:49:00,940 --> 00:49:08,600 Yeah, I mean, it's also we started four five units I mentioned and then reviewed and for various reasons. 433 00:49:08,600 --> 00:49:12,310 And took over the same institutions. 434 00:49:12,310 --> 00:49:24,680 I think in the in their countries were moving in right to the beginning of what intelligence do you have to think about increasing in some society? 435 00:49:24,680 --> 00:49:29,210 But the two time studies were halfway through. 436 00:49:29,210 --> 00:49:40,590 Yeah, right. So it's about saying we want to bring in another and number of young people a certain age to match this kind of area, 437 00:49:40,590 --> 00:49:49,930 to see whether we look at whether we might have had an impact on is really just incorporate them and use the same questionnaire in those participants. 438 00:49:49,930 --> 00:49:53,470 But you just have more people in the study. Yeah. 439 00:49:53,470 --> 00:49:57,790 Well, the answer is no, we haven't. 440 00:49:57,790 --> 00:50:06,400 I mean, we haven't ever considered increasing the original, I mean, adding in more 15 year olds or anything like that. 441 00:50:06,400 --> 00:50:10,420 I think partly because we haven't felt we've really had to because the numbers are quite large 442 00:50:10,420 --> 00:50:16,330 and they would be difficult to see as it would be difficult to do pairing as well because some, 443 00:50:16,330 --> 00:50:25,600 you know, just difficult to match. I'm not quite sure how we would find a group that was sort of similar that 444 00:50:25,600 --> 00:50:30,640 would then contribute without distorting the the group that we originally had. 445 00:50:30,640 --> 00:50:34,010 I'm not quite sure how we would do it, but we have anyway. 446 00:50:34,010 --> 00:50:37,990 So the answer is that we haven't done that. You have another question. 447 00:50:37,990 --> 00:50:43,550 Did you say something else? Yes. Yeah. Just to just to compliment to that. 448 00:50:43,550 --> 00:50:46,750 So it gave us funding for 15 years for the four. 449 00:50:46,750 --> 00:50:51,970 For the first time around since the beginning, we knew that we were going to be able to restrict over time. 450 00:50:51,970 --> 00:51:00,800 We have had our founders because there are other sources of funding to introduce a service, for instance, or to analyse the data. 451 00:51:00,800 --> 00:51:03,640 So but if you have to be the main contributor. 452 00:51:03,640 --> 00:51:10,690 Yeah, I mean, it's true that we I mean, we probably thought about this, but we actually didn't make it happen to increase a sample size. 453 00:51:10,690 --> 00:51:15,370 I guess, you know, we have to look with the fact that, you know, our studies are very, very complex. 454 00:51:15,370 --> 00:51:16,840 From a logistical point of view, 455 00:51:16,840 --> 00:51:24,130 because of the high rates of internal migration and I guess probably we didn't want to get into more trouble, to be honest there. 456 00:51:24,130 --> 00:51:30,910 But but I I see I see your point. I would have been useful, you know, in the way we think when we think about the future, 457 00:51:30,910 --> 00:51:36,130 because if you if you see a number, you realise that it is the younger corporate lawyer. 458 00:51:36,130 --> 00:51:38,650 So eventually we might need to, 459 00:51:38,650 --> 00:51:46,900 which officially which we know will no longer follow the older cohort because the sample gets even even if the attrition rate is low. 460 00:51:46,900 --> 00:51:50,680 Yet the sample gets. Also, the sample is worth lots of money. 461 00:51:50,680 --> 00:51:56,440 So we we might have to think about not tracking them at some point. 462 00:51:56,440 --> 00:52:02,740 Whereas in the case of her younger cohort, we think there will be able to leave for many, for many more years. 463 00:52:02,740 --> 00:52:10,060 I mean, that is true. I mean, I think we thought about it, but you know, we need to obtain more funding. 464 00:52:10,060 --> 00:52:14,650 And also, the complexity of the fieldwork will have increased and these are very complex. 465 00:52:14,650 --> 00:52:20,230 I have to say about that. You also have a question about attrition. 466 00:52:20,230 --> 00:52:28,510 And the reason for that reason. I mean, we're working on this with a buddy with a colleague in terms of understanding the weather. 467 00:52:28,510 --> 00:52:35,200 Attrition is so driven by several characteristics and that hope is going to be released soon. 468 00:52:35,200 --> 00:52:43,960 But what I can tell you that for this specific case of Peru in particular, but I think it may apply to some extent to other countries, is that I mean, 469 00:52:43,960 --> 00:52:49,360 one important thing that you see that the attrition rate is higher for the older cohort 470 00:52:49,360 --> 00:52:58,780 and that in that you see the larger increase in attrition rate happens from age 15 to 19. 471 00:52:58,780 --> 00:53:05,170 So that's why it is not a huge increase in attrition, but you see that it's more than what we used to have before. 472 00:53:05,170 --> 00:53:13,300 It was very linear that the increase in the rate of attrition and then it jumped between 15 and 19. 473 00:53:13,300 --> 00:53:17,770 That's between rounds a four or five. 474 00:53:17,770 --> 00:53:23,320 And I'm sorry to round three or four, I'm sorry. And the reason in part two, I mean, 475 00:53:23,320 --> 00:53:34,720 multiple migration is one of the reasons so many of them complete the school and move two to the closest city to work. 476 00:53:34,720 --> 00:53:40,230 That doesn't mean we don't follow them, we follow them. But it is just for that fact. 477 00:53:40,230 --> 00:53:44,970 It is likely that we might have more problems in finding them. 478 00:53:44,970 --> 00:53:56,810 Then also, I think by 19 there, somehow it's one thing about the the the the the the ethical approval is that, 479 00:53:56,810 --> 00:54:01,410 you know, between that pact between these two rounds, they they became official. 480 00:54:01,410 --> 00:54:09,360 So we had to be in the past, we would've requested consent from the caregiver from them. 481 00:54:09,360 --> 00:54:16,350 But if that information, I think we only needed consent from them and you have somehow more empowered, thought the left to me. 482 00:54:16,350 --> 00:54:20,400 If I have to continue with this study, I might say no. 483 00:54:20,400 --> 00:54:28,830 I mean, and in another context, another thing connected with this eight you think resonate is that. 484 00:54:28,830 --> 00:54:35,280 If you became independent, it means you need to we need to ask you to answer the question. 485 00:54:35,280 --> 00:54:40,980 In addition to everything else, because when you were age 15, when your parents. 486 00:54:40,980 --> 00:54:45,000 The question that was less answered by the caregiver. But if you're independent. 487 00:54:45,000 --> 00:54:56,400 No, no, not anymore. It is interesting to see, though, that the next round at these very two, we managed to. 488 00:54:56,400 --> 00:55:01,790 Convinced some of the previous participants that. 489 00:55:01,790 --> 00:55:06,270 Not sometimes, not really declined to continue. 490 00:55:06,270 --> 00:55:11,700 But, you know, it's like when you knock at the door separately, very polite, and they're going to say no. 491 00:55:11,700 --> 00:55:16,710 So yeah, knock at the door and they say yes, but come back tomorrow and you knock at the door. 492 00:55:16,710 --> 00:55:23,670 The next day, come back tomorrow, please. When you've done that three times, this is kind of a refusal, right? 493 00:55:23,670 --> 00:55:30,630 So you don't you don't. You start knocking at the door. But it's not that the person has said, I don't want to be part of this day anymore, 494 00:55:30,630 --> 00:55:34,290 because if they say that they don't want to be part of it anymore, we would. 495 00:55:34,290 --> 00:55:39,570 That's fine, of course, and we don't don't try to trade them in their crown. 496 00:55:39,570 --> 00:55:46,200 But if it has been something like like a. polite refuse, I don't know how to call it and we knock at the door again. 497 00:55:46,200 --> 00:55:50,130 And it's interesting to see that by listening to some of them wanted to come back. 498 00:55:50,130 --> 00:55:54,000 I think has lots of people. The fact that the age, you know, you feel more empowered. 499 00:55:54,000 --> 00:55:56,520 You like, I want to something different. 500 00:55:56,520 --> 00:56:05,720 But then that these are my old friends from the United States knocking on the door again, I should probably open the door and talk to them. 501 00:56:05,720 --> 00:56:12,890 Can I just add something? I mean, I would love to just say that it's true we didn't expand the sample, but we did. 502 00:56:12,890 --> 00:56:15,530 We did add siblings data. 503 00:56:15,530 --> 00:56:23,120 I can't remember which round that was, which enabled us to do much more on intra household demand dynamics than we've been able to do before. 504 00:56:23,120 --> 00:56:31,010 And the other thing I think, which is even more powerful was the school effectiveness study, which we introduced in 2010, 505 00:56:31,010 --> 00:56:38,420 which basically came about because of policy interests and donor interest in the extraordinary effects 506 00:56:38,420 --> 00:56:46,790 of education and the incredible aspirations that this is some of children had towards education. 507 00:56:46,790 --> 00:56:52,280 So we introduced a week of extra funding to do this, and we were basically doing school based research, 508 00:56:52,280 --> 00:56:57,170 and that's actually over 30000 children across the four study countries. 509 00:56:57,170 --> 00:57:01,340 That really is much bigger than the household sample in total, 510 00:57:01,340 --> 00:57:08,730 and that's enabled improved that for us to make the link between the indexed child and the school of their own. 511 00:57:08,730 --> 00:57:12,050 Perhaps. I mean, you know, it's true. 512 00:57:12,050 --> 00:57:13,880 I mean, actually, 513 00:57:13,880 --> 00:57:23,210 the previous question about using the older cohort as a comparison of the younger cohort or vice versa and what thinking about majoring in this, 514 00:57:23,210 --> 00:57:27,680 but now your cousin might be able to see. So we have. 515 00:57:27,680 --> 00:57:33,560 So we basically follow one child in each household at but for the rest of the children. 516 00:57:33,560 --> 00:57:42,800 We collect very basic information six a level of education and same time use, which is not not not which is a lot, actually. 517 00:57:42,800 --> 00:57:49,010 But since the third round, we started collecting additional data for the younger sibling. 518 00:57:49,010 --> 00:57:54,650 So they're the student born right after the day, the child. 519 00:57:54,650 --> 00:58:01,880 And obviously the reason why we did this was to kind of look for a not to have a counterfactual within the household. 520 00:58:01,880 --> 00:58:08,660 And actually, some of the deaths have been more successful in terms of the journey that we still have, 521 00:58:08,660 --> 00:58:14,510 but just within siblings radiation to look at the impact of programmes. 522 00:58:14,510 --> 00:58:19,850 A bill would typically both seem to have been affected. 523 00:58:19,850 --> 00:58:25,910 Have we see the programme starting from very early or late in life? 524 00:58:25,910 --> 00:58:33,920 So the sample? Yeah. So actually, when I say that we for 2000 children in the younger cohort, approximately, in fact, 525 00:58:33,920 --> 00:58:41,030 is two thousand seven hundred, for example, maybe if you include the younger siblings and for them, 526 00:58:41,030 --> 00:58:53,450 we collect the anthropometric and one on one cognitive test and it social emotional competence to say so we can actually do that comparison. 527 00:58:53,450 --> 00:59:00,550 And just when I think there was one question that I forgot to answer about in migration. 528 00:59:00,550 --> 00:59:06,590 I mean, so. I mean, you know, we our data is publicly available. 529 00:59:06,590 --> 00:59:16,210 And what you can clearly see there is we have some anonymized variables, but you can see if the person has has moved. 530 00:59:16,210 --> 00:59:25,110 And what whether the person has moved, but within the same district also has moved to a different district. 531 00:59:25,110 --> 00:59:33,940 And there is also another information about the person who still lives in an urban area or a remote area and the name of the region and province. 532 00:59:33,940 --> 00:59:40,150 So if you combine all this, you can actually finish what migration means in Peru. 533 00:59:40,150 --> 00:59:44,290 You can have to establish where someone has migrated and obviously you can. 534 00:59:44,290 --> 00:59:50,810 And also, we have questions about for those who have moved to a new district. 535 00:59:50,810 --> 00:59:56,920 The reason why they did so, obviously in many cases, is searching for work opportunities. 536 00:59:56,920 --> 01:00:04,700 Lots of opportunities. And so you can use this information to look at the, you know, what drives migration. 537 01:00:04,700 --> 01:00:10,700 And also because let me try to study what are the impacts of migration on outcomes. 538 01:00:10,700 --> 01:00:16,840 That is in fact also a nice study by one member of our team, Simon Cowell. 539 01:00:16,840 --> 01:00:23,630 He's the former PI of the study, who looked before into the impact of the migration of them, 540 01:00:23,630 --> 01:00:29,690 of the mother of the young children and its impact on child outcomes. 541 01:00:29,690 --> 01:00:36,710 And for that in particular, in the case of Peru in round two. For those that are familiar with the dataset, if you go to run the dataset, 542 01:00:36,710 --> 01:00:43,280 there is a model that measures the migration of the mother so that you can actually do that. 543 01:00:43,280 --> 01:00:50,100 I used to have to say that we collect more information about migration, more specifically about the places. 544 01:00:50,100 --> 01:00:54,140 But because of the confidentiality agreement, we cannot release that data. 545 01:00:54,140 --> 01:01:00,460 But even with a publicly available version of this dataset, there are many opportunities to analyse this this this, 546 01:01:00,460 --> 01:01:12,560 this topic, which are how inequities on my case because there are so many things to do, but some of you can do it. 547 01:01:12,560 --> 01:01:21,290 Well, I have a question. I confess what I should say as we talk about attrition, the attrition rates of young lives are remarkably low. 548 01:01:21,290 --> 01:01:27,590 I mean, if you look at British based cohorts, you'll often see up to 40 percent attrition rate over the same period. 549 01:01:27,590 --> 01:01:33,170 So it really is. Congratulations to the team so that we maybe we may be talking about what predicts attrition, 550 01:01:33,170 --> 01:01:38,600 but we're really talking about tiny numbers relative to other studies. 551 01:01:38,600 --> 01:01:43,130 My question is about something which we have in all of our studies, which is a checklist, 552 01:01:43,130 --> 01:01:49,670 and the quick is the list of questions that we wish we'd asked but haven't found. 553 01:01:49,670 --> 01:01:54,770 And I wondered whether you had your own checklist, official, unofficial and things. 554 01:01:54,770 --> 01:01:59,660 That is particularly for other researchers who are considering working on starting cohorts. 555 01:01:59,660 --> 01:02:07,510 Things that you wish you had asked, but didn't. That means no. 556 01:02:07,510 --> 01:02:16,650 I know I might have asked about pets earlier on in the study, but you have to explain that, Mary. 557 01:02:16,650 --> 01:02:24,810 Well, actually, this is not just me through limbo is very keen on that pet, mainly dogs. 558 01:02:24,810 --> 01:02:30,520 It's almost exclusively dogs. And so it's not just me. 559 01:02:30,520 --> 01:02:37,630 And so we did introduce a question of what ground we introduced it in that did you have pets? 560 01:02:37,630 --> 01:02:47,770 And it has been quite interesting where we haven't fully analysed it yet, but we're working on getting some funding to analyse it from an early age. 561 01:02:47,770 --> 01:02:51,700 But it would have been quite interesting also. But this is that's a minor issue. 562 01:02:51,700 --> 01:02:59,650 I mean, one thing that is quite nice to know is the British people's hearts were you'd be surprised. 563 01:02:59,650 --> 01:03:07,310 I mean. I mean that just as an example, this is a Leamer example, but a very large number of families have dogs. 564 01:03:07,310 --> 01:03:11,940 But because it's a dog, we have a dog walker because we've improved. 565 01:03:11,940 --> 01:03:24,930 And somebody probably has to live in the house because it's got animals in the house, maybe dogs, some cats, guinea pigs, but mostly dogs. 566 01:03:24,930 --> 01:03:27,660 So people, you know, it's it's an expenditure. 567 01:03:27,660 --> 01:03:35,010 And even in rural areas, I'm not sure everyone are very different from, you know, if you live in a rural area, a farm dog and things, 568 01:03:35,010 --> 01:03:44,490 but limits certain things and not just middle class areas, lemur shantytown areas, a lot of people have have to have pets. 569 01:03:44,490 --> 01:03:49,590 So this is kind of like a the size of because it's not the most important aspect of their lives. 570 01:03:49,590 --> 01:03:55,110 But some of us think it's interesting and it's pretty complete, actually. 571 01:03:55,110 --> 01:04:02,660 But I haven't thought, Oh my goodness, I wish we'd really asked that. 572 01:04:02,660 --> 01:04:07,700 No. Yeah, I mean, I can put a specific thing. 573 01:04:07,700 --> 01:04:14,720 Perhaps if you ask me another thing, I'd remember something I'd ask that I would have liked of all, but it has been. 574 01:04:14,720 --> 01:04:23,500 Yeah, but maybe one reason why many cannot come up with questions that she wish we had added is 575 01:04:23,500 --> 01:04:29,080 because because of the design that we have this older cohort of the younger cohort and. 576 01:04:29,080 --> 01:04:39,310 You know this this has been established like this in a way so that we have the chance to introduce questions for the older cohort first. 577 01:04:39,310 --> 01:04:48,070 And you know, if there is anything to be modified or approved, then we have a few years so that pilot and the younger the same age. 578 01:04:48,070 --> 01:04:54,750 We actually have a very big question, everybody. Then they have the same age. So I would say probably that we have some regrets. 579 01:04:54,750 --> 01:05:00,670 Well, maybe amongst more of the older cohort, but not so much older than the younger cohort, 580 01:05:00,670 --> 01:05:08,000 because we already had a chance to think about this for many years. For instance, know I wish we had included people. 581 01:05:08,000 --> 01:05:16,440 The beta test since the first round for the older cohort, not for the younger cohort, because they were too young. 582 01:05:16,440 --> 01:05:22,770 But, you know, so this is not a regret for the younger cohort, but it is for the undercover eye, 583 01:05:22,770 --> 01:05:28,770 which we have not seen maybe a cognitive achievement just for the caregivers in the first visit. 584 01:05:28,770 --> 01:05:38,850 But you know, I see also that our question and reserves are so long already that I can see why there was a logic to keep giving it as it was. 585 01:05:38,850 --> 01:05:46,870 I wish we had it started with the civilians earlier. I wish we could start that with them in trying to. 586 01:05:46,870 --> 01:05:52,360 And, you know, we learnt our time, I remember with the seal, is you. 587 01:05:52,360 --> 01:05:58,090 I mean, because with the seal, we started only with dental metrics and the cognitive information. 588 01:05:58,090 --> 01:06:05,550 But then after we said, we should also ask about such emotional competencies from them. 589 01:06:05,550 --> 01:06:12,750 So, yeah, I mean, there are some things that we could have done differently, and also, I guess in terms of measuring early childhood development, 590 01:06:12,750 --> 01:06:17,100 I think many things have improved over time and discuss tools, 591 01:06:17,100 --> 01:06:23,250 a standardised measure of early childhood development that maybe were not available from one 2002. 592 01:06:23,250 --> 01:06:28,680 Or maybe they were. But I'm very happy with the information we've collected. 593 01:06:28,680 --> 01:06:36,390 But on that area of early childhood development, we could have collected more in from one or had an earlier round one. 594 01:06:36,390 --> 01:06:42,450 I think that's been one of the biggest challenges actually is initiation already at age one. 595 01:06:42,450 --> 01:06:48,810 You know, for us, obviously, some birth cohorts raised tons of questions that we'll never be able to answer. 596 01:06:48,810 --> 01:06:56,940 And I think it's really important to remember that because we've entered the nutrition debate and made a very strong case for this argument, 597 01:06:56,940 --> 01:07:06,750 that children can recover the time and they don't just recover physically, but they also recover cognitively in terms of educational attainment. 598 01:07:06,750 --> 01:07:10,920 So people come straight back to us and say, well, which children are recovering? 599 01:07:10,920 --> 01:07:18,000 And we're not able to answer the questions about whether it's the children who were malnourished. 600 01:07:18,000 --> 01:07:24,240 Maybe they weren't malnourished as long or as badly or as early as some of the other children. 601 01:07:24,240 --> 01:07:33,840 And maybe that's what's enabled them to recover. That, of course, is the most vital question when you when you post this new opportunity for recovery. 602 01:07:33,840 --> 01:07:38,160 People need to know that because otherwise, what are they going to know? What's how is it going to help? 603 01:07:38,160 --> 01:07:47,070 So I get very frustrated that we weren't actually really gathering data from in utero and also, you know, much earlier from the first. 604 01:07:47,070 --> 01:07:50,910 I mean, there are many studies which do start in the first phase of life. 605 01:07:50,910 --> 01:08:00,180 There were reasons why it was a good choice to make, but it's had that negative consequence, I think, in terms of kind of answers we can provide. 606 01:08:00,180 --> 01:08:06,930 But we didn't know we were going to make this discovery, so we didn't know we were going to need to go to answer those questions, which is your point. 607 01:08:06,930 --> 01:08:20,520 I think of another thing that would be. Quite interesting would have been that they're now changing the team a bit, but there are now. 608 01:08:20,520 --> 01:08:24,150 Things you put on your hands that would tell you how many pieces you are putting away. 609 01:08:24,150 --> 01:08:28,370 Tell me how many pieces you done and measure up physical activity, that would have been interesting. 610 01:08:28,370 --> 01:08:35,700 We don't really have any good measures of physical activity we've got so that we've got not crude measures where have done. 611 01:08:35,700 --> 01:08:41,940 We don't have to do very much on dietary intake either, really, but it is certainly physical activity would have been there. 612 01:08:41,940 --> 01:08:44,760 There are much more opportunities now to be able to. 613 01:08:44,760 --> 01:08:52,650 I'm not sure we never got it back again, given, but Buffalo, given that in the beginning, was a presence. 614 01:08:52,650 --> 01:08:57,330 In fact, as we read it and then moving it back to you or something, but um. 615 01:08:57,330 --> 01:09:02,910 But I think that would have been something that would be interesting because we don't have very good information on that, 616 01:09:02,910 --> 01:09:08,850 and it's probably quite different in rural and urban areas. Can they add something to this? 617 01:09:08,850 --> 01:09:13,560 I mean, you know, because we do this every three, four years, each round, 618 01:09:13,560 --> 01:09:18,240 and so each round we have a new discussion about what are we going to measure now? 619 01:09:18,240 --> 01:09:22,300 Of course, we need to keep a lot of comparability with the older cohorts. 620 01:09:22,300 --> 01:09:25,650 You know, there's a big part of the question that has to remain the same. 621 01:09:25,650 --> 01:09:32,820 We will have to assess in time and we have to cut this model because, you know, maybe people are not using this for research. 622 01:09:32,820 --> 01:09:36,870 Maybe you know what's exciting the first, but it's not really working. 623 01:09:36,870 --> 01:09:41,050 And then research, you know, a whole discussion about what we want to measure now. 624 01:09:41,050 --> 01:09:44,430 So this is an interdisciplinary group. 625 01:09:44,430 --> 01:09:53,130 I'm an economist from the office of a psychologist. We have an anthropologist comedies, but this is amazing. 626 01:09:53,130 --> 01:09:59,580 But also, we want because it is. We have to remember that is an international study for countries. 627 01:09:59,580 --> 01:10:08,910 We also want to have comparability with other countries. So every few years, we gather together to discuss what's going to be the question next week. 628 01:10:08,910 --> 01:10:12,960 And everybody wants to let a question and we're at Molineux. 629 01:10:12,960 --> 01:10:20,520 So if that is a tough negotiation. So we have at some point we have reached an agreement in which if you if you say you want to add something, 630 01:10:20,520 --> 01:10:27,070 you also have to say what what you want to drop. So that the time remains the same. 631 01:10:27,070 --> 01:10:35,200 And that's what maybe that's why, you know, America's really great ideas about things to incorporate and petroleum i the theme of 632 01:10:35,200 --> 01:10:41,480 let's give them illustration time a constant because it's a every very long question. 633 01:10:41,480 --> 01:10:48,400 One thing that I would probably like to have more, which would be about the food environment. 634 01:10:48,400 --> 01:10:57,100 So when when we've we've asked them, we have some information on some dietary intake and diversity and a few things not obviously, 635 01:10:57,100 --> 01:11:03,880 we can't do a full nutritional study, but it would be I think it would have been interesting and it would be interesting. 636 01:11:03,880 --> 01:11:12,450 Now I think to. Have some information on that food environment, not to say how far you are from shops, what shops are available, 637 01:11:12,450 --> 01:11:20,060 and this is a whole area which we haven't really and wouldn't be very difficult because we do have the community. 638 01:11:20,060 --> 01:11:27,080 You know, we work in the community as well. So I think that that's something we might think about. 639 01:11:27,080 --> 01:11:34,220 I think a few questions now, but it would have been would be it would be useful to have known a bit more access to food. 640 01:11:34,220 --> 01:11:39,200 Where do you get it from? Any other questions? 641 01:11:39,200 --> 01:11:49,370 I mean, you talked about the longitudinal survey and we know young life of mixed methods study, you know, fabulous qualitative research team in place, 642 01:11:49,370 --> 01:11:55,790 and they've been able to respond to questions as they arise, whether it policy, relevant issues or things that have emerged in the survey. 643 01:11:55,790 --> 01:12:05,930 And they've been able to follow this through qualitative studies in the online communities and where those experiences still exist in the sample. 644 01:12:05,930 --> 01:12:18,820 They can then go go outside of the community. And I was wondering here if you on what you think the qualitative research adds to the survey. 645 01:12:18,820 --> 01:12:24,290 Yeah. No, I mean, maybe one reason why I haven't thought much about the qualitative component of the 646 01:12:24,290 --> 01:12:29,350 study is because I have been more focussed on the design of the quantitative part. 647 01:12:29,350 --> 01:12:32,600 But but just for that reason, I've been shy about this. 648 01:12:32,600 --> 01:12:35,630 But now of course, we are. So, yeah, 649 01:12:35,630 --> 01:12:44,000 so basically we have kind of comparable grounds for quantitative and qualitative sample on the qualitative sample is much smaller for obvious reasons. 650 01:12:44,000 --> 01:12:50,990 We have a sample of approximately 50 participants in total from two cohorts of twenty five from the younger cohort, 651 01:12:50,990 --> 01:12:55,970 twenty five from the other cohort with a kind of a gender balance. 652 01:12:55,970 --> 01:13:04,940 And, you know, we've been following them through in-depth interviews and focus groups as if they were very young. 653 01:13:04,940 --> 01:13:07,010 And it's been extremely useful. 654 01:13:07,010 --> 01:13:14,020 I mean, if I have to highlight something because I could, I could highlight the, you know, the quality of the research. 655 01:13:14,020 --> 01:13:21,130 But, you know, because I haven't been very much involved in the in writing these papers. 656 01:13:21,130 --> 01:13:30,320 But what I have seen a very big impact is in the in the dialogue with the policy makers because, you know, for example, when you want to think about. 657 01:13:30,320 --> 01:13:38,990 I mean, just to give an example, in Peru, we have something called a bakery, which is a scholarship for higher education that is a national, 658 01:13:38,990 --> 01:13:43,110 a nationwide programme, but it's a very selective and very selective programme. 659 01:13:43,110 --> 01:13:50,720 So only those that have the highest court attempting to make it and assuming conditional on being from a public school. 660 01:13:50,720 --> 01:13:57,560 So even though only a few of them have utterly obtained this scholarship, 661 01:13:57,560 --> 01:14:05,180 you know we can interview those few people comparing them also with people that maybe even apply but didn't get the scholarship. 662 01:14:05,180 --> 01:14:12,090 And. It's really interesting because, for instance, the government had information about the impact of the caricature, 663 01:14:12,090 --> 01:14:17,000 but this is a completely different story when you have the chance to talk to them, 664 01:14:17,000 --> 01:14:26,400 to play to those that have used their scholarship and had the chance to really tell you why you might be working or why not or not. 665 01:14:26,400 --> 01:14:32,070 So in the dialogue with the policymakers especially useful, I was giving the example of a public programme. 666 01:14:32,070 --> 01:14:39,660 But in general, when we have discussions about child work and educational trajectories, 667 01:14:39,660 --> 01:14:47,040 trajectories to the labour market, so we have defined the quantitative data. 668 01:14:47,040 --> 01:14:59,990 But, you know, you really need to understand what is driving the results, and for that, the continuation of the qualitative data is huge. 669 01:14:59,990 --> 01:15:06,200 I know you've had some quite significant ethical concerns and considerations. 670 01:15:06,200 --> 01:15:17,900 And one of the one of the challenges I think that we're all facing now is that the whole world of ethics is really becoming very it's becoming a, 671 01:15:17,900 --> 01:15:25,400 you know, it's it's the culture of ethics is very strident about how you deal with certain problems in particular, 672 01:15:25,400 --> 01:15:26,570 for example, 673 01:15:26,570 --> 01:15:38,990 the requirement that you report any troubling situations and in troubling incidents and so on that you encounter in the field and in some context, 674 01:15:38,990 --> 01:15:44,300 reporting to authorities isn't necessarily going to make the situation any better. 675 01:15:44,300 --> 01:15:46,200 It may actually exacerbate the situation. 676 01:15:46,200 --> 01:15:53,090 So it's not as straightforward as it might seem when you're, you know, you or I, I'll be putting requirements on you. 677 01:15:53,090 --> 01:16:00,530 And I was just wondering if if you have anything to share about your experience of of the sort of ethics of the research over the years, 678 01:16:00,530 --> 01:16:07,220 whether they've been consistent, you know, typical problems or whether or how you've dealt with them, 679 01:16:07,220 --> 01:16:12,650 whether there are some that you felt were really unresolvable or I know you are 680 01:16:12,650 --> 01:16:17,030 for some have mentioned when you say goodbye to people at the end of the study, 681 01:16:17,030 --> 01:16:22,760 it's not really goodbye because they have your phone number and they phone up and they keep on fighting up and they have problems. 682 01:16:22,760 --> 01:16:31,340 So does this pose a challenge? But the people have had an even closer relationship than in longitudinal study, for example. 683 01:16:31,340 --> 01:16:33,320 There are lots of issues that have emerged. 684 01:16:33,320 --> 01:16:44,060 So I think what one aspect is that, you know, we've been with this family for 15 years and probably what is going to be more than that. 685 01:16:44,060 --> 01:16:52,010 So in the in the act, in fact, in the last visit, officially, we had to say goodbye in a way because why? 686 01:16:52,010 --> 01:16:55,760 Because it's the constant fear that we ask them to sign the wrong one. 687 01:16:55,760 --> 01:16:59,270 It said that this study was going to last 15 years. 688 01:16:59,270 --> 01:17:05,060 I mean, don't worry, we're only going to come here every three to four years and the longer that. 689 01:17:05,060 --> 01:17:07,580 And if we have something new, we're going to tell you. 690 01:17:07,580 --> 01:17:15,680 And when I get your reminder and be like, you know, it was not a surprise that we were coming back for 15 years, but during and the last visit, 691 01:17:15,680 --> 01:17:25,610 we kind of we kind of said goodbye because we had to, but we also told them we might come back and we would like to continue with this stay. 692 01:17:25,610 --> 01:17:33,230 We try to keep the door open in a way because we knew it was very likely that we were going to come back and say, 693 01:17:33,230 --> 01:17:38,690 No, no, we're going to be able to say goodbye. I mean, I guess at some point for the older cohort, 694 01:17:38,690 --> 01:17:46,330 they're probably going to take the decision for burial because this is becoming even though the attrition rate is low in the small sample as it goes. 695 01:17:46,330 --> 01:17:51,050 Maybe, maybe we're going to have to say at some point, OK, this is the last time a, you know, 696 01:17:51,050 --> 01:17:58,370 we we ask you for some time, but also the fact that we know we go, we visit them every time. 697 01:17:58,370 --> 01:18:03,650 So I mean, something important to many mentioned at the beginning is that we have this data set of ethical, 698 01:18:03,650 --> 01:18:11,840 not just only ethical cases, but in general difficult cases that feel you need to remember before you go and knock on the door again. 699 01:18:11,840 --> 01:18:17,990 You have to remember the last time we came, you know, even the small things like, you know, 700 01:18:17,990 --> 01:18:25,370 the person, for some reason we forceps and then with upset with the comments of that. 701 01:18:25,370 --> 01:18:34,100 So every time we go back again, that is a form that has gone Tusk's comments about ethical potential ethical concerns. 702 01:18:34,100 --> 01:18:44,600 I mean, it can be something as small. It can be something bigger like, you know, if we went if in the last round the mother asked us for help, 703 01:18:44,600 --> 01:18:47,120 it could be for many, for many, for many things. 704 01:18:47,120 --> 01:18:54,710 I know if there is a disabled person in the house and they want to know how they can access a wheelchair, 705 01:18:54,710 --> 01:18:58,610 and sometimes they don't know that it is possible to access one. 706 01:18:58,610 --> 01:19:05,540 They just need to register, but they don't know what to register. So we we make sure that. 707 01:19:05,540 --> 01:19:13,040 So what? We have this information. First of all, the numerator we ask him to ask him or ask her if it's something that can wait. 708 01:19:13,040 --> 01:19:20,870 Take notes and then get back to the supervisor in the field and their supervisor gets back to the field manager in Lima. 709 01:19:20,870 --> 01:19:26,570 And if in many respects, to the research team in Lima and we try to find our way, 710 01:19:26,570 --> 01:19:31,940 sometimes the problem can be solved locally because the numerator of experience experiences they know, 711 01:19:31,940 --> 01:19:37,540 they already know, for instance, how you can register to obtain a doctor in the wheelchair, for instance. 712 01:19:37,540 --> 01:19:41,870 So but they still have to let us know that that's happening, and then we take note of that. 713 01:19:41,870 --> 01:19:50,880 So the next time we come, we know that that happened. But also because we need to keep up with our work if we said. 714 01:19:50,880 --> 01:19:55,920 I cannot I am taking note of this. And then we don't follow up. 715 01:19:55,920 --> 01:20:04,110 I mean, that would be bad for us as individual families, but also it would be bad in terms of our strategy because if we come back, 716 01:20:04,110 --> 01:20:11,160 if four years later and they say, you promise to help me with this, with this problem and you didn't. 717 01:20:11,160 --> 01:20:16,770 So I mean, there are these two aspects that we need to to balance. 718 01:20:16,770 --> 01:20:22,290 I will say that in general, overall, I mean, of course, there are things that we might have improved. 719 01:20:22,290 --> 01:20:31,950 I think, you know, nowadays there's more. Maybe more and more attention in terms of the medical procedures you use in the field. 720 01:20:31,950 --> 01:20:37,440 I do. I feel very proud of the of the protocols that were established before I joined, to be honest. 721 01:20:37,440 --> 01:20:41,070 I mean, Mary and the team had to kind of re-establish this, basically. 722 01:20:41,070 --> 01:20:46,410 We've been trying to improve, prove something, but I have to say that, you know, if I seen nowadays, 723 01:20:46,410 --> 01:20:52,320 what what do you require compared to what we've been doing over the last since the beginning? 724 01:20:52,320 --> 01:21:01,920 I think, you know, there is more freedom we may improve. But overall, a man in the team did a great work. 725 01:21:01,920 --> 01:21:11,040 Perhaps we don't have more questions I would like to hear, and not just from from Marion Island, but from the rest of the OnLive team. 726 01:21:11,040 --> 01:21:18,510 Whether you would have just a sentence of advice for researchers who were embarking on a cohort 727 01:21:18,510 --> 01:21:25,140 or embarking on a work on a cohort and what you see from the breadth of your experience, 728 01:21:25,140 --> 01:21:38,690 what your advice and suggestions to them would be. Gosh, give you a moment, partnerships, relationships are absolutely fundamental, 729 01:21:38,690 --> 01:21:50,240 and I think the more I think about reflecting back on challenges that we faced that need to be, you have to be incredibly flexible. 730 01:21:50,240 --> 01:21:59,480 I often tell people that the Brexit referendum 2016 in this country was the single biggest 731 01:21:59,480 --> 01:22:04,520 threat to young lives because we were about to have our most expensive ever year of field work, 732 01:22:04,520 --> 01:22:12,500 which meant that we were. The funding was going from Oxford to the study countries and the pound collapsed overnight. 733 01:22:12,500 --> 01:22:17,210 It's not hundreds of thousands of pounds of budget. 734 01:22:17,210 --> 01:22:24,740 We couldn't have adjusted and dealt with. Fact if it wasn't for the fact that we had long term relationships and partnerships of trust, 735 01:22:24,740 --> 01:22:30,020 we immediately contacted everybody in the team and said, we're going to have to make cuts. 736 01:22:30,020 --> 01:22:36,350 What we can to do about this sort of flexibility, the trust, the adjustment and all of that was possible. 737 01:22:36,350 --> 01:22:45,590 And I'm really alarmed by studies that try to be longitudinal using companies, including companies that have flown in from overseas. 738 01:22:45,590 --> 01:22:51,110 I don't think we would have anything like the level of success in terms of low attrition in terms of the 739 01:22:51,110 --> 01:23:00,080 quality of the data in terms of the respect towards the study country of the participants in the study. 740 01:23:00,080 --> 01:23:05,390 Almost any aspect you want to mention is only really come about through the partnership approach. 741 01:23:05,390 --> 01:23:08,310 It's not always been easy. We don't always agree. 742 01:23:08,310 --> 01:23:15,590 Economists sometimes try to talk to them as apologists, and they don't understand each other at all because we all use funny words. 743 01:23:15,590 --> 01:23:19,710 Others don't understand, but could always tease out, well, they use. 744 01:23:19,710 --> 01:23:26,660 Yes, exactly. We at least we talk to people and we use words, and they just use all these numbers if somehow that mattered in the world. 745 01:23:26,660 --> 01:23:28,460 But anyway, we've yeah, 746 01:23:28,460 --> 01:23:38,720 it's really important and I think people underestimate and I'm really worried when I see requirement donors to set the study up and go with it. 747 01:23:38,720 --> 01:23:42,620 They just give you like a six month inception period. It's not long enough. 748 01:23:42,620 --> 01:23:47,480 You must have time to build good relationships and and learn about each other. 749 01:23:47,480 --> 01:23:51,980 I think that's to me, that's the single most important thing. I think that you're right about that. 750 01:23:51,980 --> 01:24:02,960 This is a very good point. There's a multidisciplinary component of there's been a drought, particularly in the last five rounds. 751 01:24:02,960 --> 01:24:06,620 Probably seen them. Yeah, that's been very, very positive thing. 752 01:24:06,620 --> 01:24:11,840 I mean, that makes you kind of feel that you've got a role in you contributing bits, you know? 753 01:24:11,840 --> 01:24:16,010 But you're lucky. We've had a good group to work with. 754 01:24:16,010 --> 01:24:19,640 The two institutions got all without being embattled together. 755 01:24:19,640 --> 01:24:24,370 But yeah, I think that's a good point. 756 01:24:24,370 --> 01:24:30,430 If I can, I mean, just a couple of ideas. 757 01:24:30,430 --> 01:24:40,330 I mean, I think it's very important that you work with people, I mean, offend our group from the point of view of research, it's excellent. 758 01:24:40,330 --> 01:24:45,590 But we wouldn't have survived if you know the group of people really didn't care. 759 01:24:45,590 --> 01:24:52,530 You know, like when I joined the team, to be honest, I didn't know that, you know, two years after we collected data, 760 01:24:52,530 --> 01:24:57,490 when received a phone call from a participant asking for information or asking for help. 761 01:24:57,490 --> 01:25:06,460 And we have a protocol to deal with the situation. But you know, you can say we are not collecting data this year, but you have to and and you know, 762 01:25:06,460 --> 01:25:11,290 they have our information, our contact information in the consent forms and in the consultation dates. 763 01:25:11,290 --> 01:25:14,970 So they contact us, you know, the. 764 01:25:14,970 --> 01:25:24,550 Enough to deal with the situations the ethical crisis due to the brought the moment in which we are collecting data. 765 01:25:24,550 --> 01:25:26,500 You have to deal with those situations. 766 01:25:26,500 --> 01:25:33,840 This has nothing to do with research, of course, in a way, but it's probably the most important thing you end up doing when you're collecting data. 767 01:25:33,840 --> 01:25:38,370 And actually, you have to care if you don't care. 768 01:25:38,370 --> 01:25:43,590 Maybe that's also why we're successful with attrition, but surely, but you have to care. 769 01:25:43,590 --> 01:25:54,430 And the second, yes, final advice is to take photos because I and I don't know exactly to what extent this helps. 770 01:25:54,430 --> 01:25:59,770 I'm pretty sure that it helps a lot like they know that they're receiving that photo from the last round and 771 01:25:59,770 --> 01:26:07,360 they want to keep that photo and we've seen in their walls that they they have a special place for the photos. 772 01:26:07,360 --> 01:26:11,560 So they argue that that's a very good idea, 773 01:26:11,560 --> 01:26:18,040 not just just obviously to show to anybody because it's because we couldn't shoot, because that would be the problem. 774 01:26:18,040 --> 01:26:26,690 But the photos to go back to the that we give back to the family that aren't people that don't whose idea it was, it was a bad idea. 775 01:26:26,690 --> 01:26:31,770 And any other colleagues will rise anything to. 776 01:26:31,770 --> 01:26:43,280 Or shall we close by saying a huge thank you and we look forward to reading even more than the 400 papers that have already ActionScript read this. 777 01:26:43,280 --> 01:26:52,690 They say a handful of small countries, when I feel some people are telling delegates if they start using the data, if they have in race is public. 778 01:26:52,690 --> 01:26:56,580 Well, a huge thank you. We look forward to the next in this series. 779 01:26:56,580 --> 01:27:03,180 And congratulations on the incredible success you've achieved both in Peru, but in my DNA study. 780 01:27:03,180 --> 01:27:07,646 Thanks, everyone. And.