[musical cue] [Voiceover] National Data Archive on Child Abuse and Neglect. 
[Erin McCauley]
All right folks we are going to get started. So just a few house cleaning before we begin. But this session is beginning at 12 Eastern right about now. During the presentation or workshop please submit any questions you have to the Q and A that way we can capture them for the purposes of creating our webinar afterwards where we turn this into a video series posted on our website and that way other folks can see the question. If things are put into the chat other participants can't see it. We also want to make sure everyone knows that this session is being recorded this is because we turn these summer training series kind of our live Workshop series into a webinar series that's been posted and preserved on our website. Throughout these sessions sometimes we'll refer to Prior summer series and so anything is available on our website from all of the prior Summers. And then you know if anyone's having any Zoom issues you can always go to the main support Zoom line or Andres who is a superstar behind the summer training series working behind the curtain getting all this done is available for help and His email is listed up there. But other than that we are going to get started as folks Roll On In. But as I said this is the summer training series by NDACAN or the National Data archive on child abuse in neglect. This is the first session of our summer series and this year this will focus on an introduction to NDACAN and all that we do. But mostly focus on the administrative data set. We have our resident statistician Sarah Sernaker here and she's going to be leading us through today's presentation. If you've ever been to our office hours you know that she also meets with folks in one-on-one or small group settings to help support data users. And then behind the curtain she's doing a lot of the actual data archiving so she has a ton of knowledge on all of the data that's available here at NDACAN and she's a great resource I hope everybody uses. And the NDACAN data archive is hosted at Cornell University and Duke University. If you been with us over the years we were originally at Duke or excuse me originally at Cornell and moved to Duke and right now we are co-hosted by both institutions along with having affiliations with different organizations and groups like UC Francisco where I am and Mathematica. And we are here and all of this is possible because of Children's Bureau they are the ones who are are really behind behind the curtain and we have Tammy White here to to introduce us in in just a minute and she is from Children's Bureau so they are really behind it all. So here is an overview of the summer. As I said today we're doing an introduction to NDACAN administrative data series. Next week where you're gonna talk about a new data acquisition so this is really exciting especially if you've been a long-term user to see kind of what's what's coming down the pike. And then we're going to be talking about causal inference using administrative data so you know if there's something that you see here that's really of interest and you're interested in learning about causal inference or seeing it applied I'd recommend coming back for number three. We're also going to talk about evaluating and dealing with missing data in R if any of you have used administrative Data before you know that missingness is kind of a persistent issue that you always have to deal with. We're going to talk about time series analysis in Stata and then we're going to end with a workshop on data visualization in R I think about a year ago maybe a year and a half ago we had a workshop on data visualization in R where we just kind of began the Journey of it and it was really popular so we've brought it back with a little bit more dedicated time this summer.
[Sarah Sernaker]
Yeah that was a office hours to jump in. 
[Erin McCauley]
Yeah and office hours is a great way to get continuing support throughout the year you can meet one group or one-on-one with our statisticians and staff members and then we also have many more informal workshops. It's also a great way to network and get any professional development support. And so here is our session agenda and before we dive in I'm just going to pass it to Tammy White to do an introduction to the series from Children's Bureau.
[Tammy White]
 Great thanks Erin. Hello everyone I just wanted to take a few seconds to welcome you all to this training for the summer. It's been around for a few years and I'm excited that it has been so popular that the archive is continuing annually now. It looks like a great season this year so I encourage you all to attend as many as you can. It's a really great resource for researchers and students and anybody else who's kind of interested in our administrative data set. So I appreciate you all being here I hope you can continue to attend more and I will turn it over back to Erin and or Sarah who's ever next on that agenda. Welcome everyone!
[Erin McCauley]
Thank you Tammy Sarah will now be taking it away and you will see Alex and I in the chat and answering questions throughout the session.
[Sarah Sernaker]
 Yes thank you for the introduction Erin to NDACAN and Children's Bureau and Tammy for that little introduction. And so my name is Sarah Sernaker I'm a statistician and part of NDACAN. And so this is the first time I'm doing this presentation so bear with me as I get warmed up here. But so today this is just an introduction mostly to NDACAN as a whole and in general what we do what we can offer you as a potential data user or current data user and data sets that we archive. And we archive I think over 200 data sets from the start of NDACAN so we're only touching like the tip of the iceberg here but these are the data sets that I'll get into that are used the most and most widely used so. So archiving data why do we even bother archiving data in the first place? So the whole point of archiving data is to have a place to keep and store data and to be data disseminators and be able to have the infrastructure to share data with users and help support them in using it so in that sense we maximize the utility and exposure of data, we allow the data to be analyzed from multiple perspectives including people from very many different fields. And we importantly preserve the history of child welfare. If you stop to think about it if these data didn't have a place to live, they'd kind of be all disparate and maybe not as easily accessible so we're preserving, hopefully enriching the history of child welfare. And through all that we increase the research volume on issues of child welfare so by all of these things we are implicitly pushing forward the research of child welfare and hopefully enabling the breadth and encouraging it the field to grow. So archiving is really important. It's a lot of like stuff that happens implicitly but it's all very vital to a larger data picture. So particularly when you archive with NDACAN some benefits to the data contributor so you might be here thinking that you have a data set or data collection that you're doing and you don't know where to ultimately put your data and some contracts do require data collectors to archive their data somewhere. And so as a data contributor NDACAN offers the following benefits. So we preserve the original data so we we take the data as it is you know under some guidelines of personal identifiability and risk management sort of things. But under those considerations we preserve the original data. We have created a large network of people in the field and in adjacent fields and other fields you might not even think of. And so the NDACAN network is great for connecting interested users, people who might not know about your data or have heard of it. So we increase the potential of number of author citations and just general data usage and importantly not just as a data contributor but also as a data user our staff provide a lot of help all around to all the people we interact with data users and data depositors to really make sure the data that we're archiving and distributing are of high quality, high standard, that they're cleaned and compatible amongst each other or within itself. So there are really hard working staff behind the scenes who really make sure that our data are living up to the appropriate standards of the field and requirements.
So benefits to the research Community not just at an individual level but NDACAN helps bring a greater understanding to the study of child well-being through replication and extension of previous research. So by that we maintain a database of literature that uses our data so that helps connect users with existing literature that uses our data and existing research and research goals and questions. So we do you know beyond the just connections of people in the field and users of our data we do try to keep track of the actual research and research questions and those sorts of things we keep track of those through what's called our candl bibliographic citation database sort of thing. And so and we also offer events and support as you all here are aware this is one of our events one of our popular events this is just one series of our summer training series events. As Erin has mentioned we have office hours in the fall we have a trump summer training summer research oh my God that's that's all right summer research institute there's too many acronyms here summer Research Institute. So we have a lot of offerings and like our goal here is to help you and to help data users and to archive data and you know we we just we do a lot to help our users because we want to encourage the connections in the Child Welfare community and we want to support users to help them make sure they're doing high quality research and using our data appropriately. So that is I think a key benefit or offering of our archive that you know I don't I've not seen in other archives I guess. We do offer a huge amount of support one-on-one or in these organized events.
So like taking a step back to the gen the larger archiving picture. Who are we here at NDACAN? So NDACAN I've been using it generously but that stands for the National Data Archive on Child Abuse and Neglect. And so our mission is to promote secondary analysis of child abuse and neglect data by providing researchers with high quality data sets,  documentation, and technical support and encourages collaboration within the scientific community. So basically a nice concise summary of all of what I was trying to say and that is our core mission. So the background of NDACAN is it was started at Cornell University or that's where it was initially housed and we've expanded that to Duke University so we are part of two core organizations Cornell and Duke. It was founded in 1988 so we've been here I haven't been here that long but NDACAN has around so what like 20 my math's right let's say like 25 years or so exactly 25 years you should do like a 25th birthday event. Anyway so NDACAN is funded and supported by the Children's Bureau so we are in a way government contractors with Children's Bureau who is part of ACF and larger HHS so the Department of Health and Human Services largely. So that is who keeps us running Keeps the Lights On keeps us all here doing this fun stuff. So the people that make up NDACAN so we have the director who's not here with us today but that's Christopher Wildeman so he's at Duke we also have co-directors Erin McCauley who you heard from before she's at University of California San Diego and we have Maria Fitzpatrick who is at Cornell. So the three we have our directors all at different places which is beneficial to users and depositors because it just really expands our network of colleagues and the NDACAN Community. Our senior advisor is John Eckenrode some who have been with us for a while may have known him formerly as our Director but I think he ultimately retired and Chris took on that position. Marie Cope is our project manager manager who's always working really hard behind the scenes. We have various research Associates Alex Roehrkasse who's at Butler University it's not University Butler University yeah and Frank Edwards who's really knowledgeable. I think both of these guys are doing presentations at the Summer Training Series. Frank is doing missing data I think I can't remember but they're both both both really great and knowledgeable. We also have Clayton Covington who's been with us for a few years he started with us as a research aide and now he is a PhD student at Harvard so he still is around in helping us. Then there's the statisticians sorry I realize I'm jumping around this columns so we have the statisticians me Sarah Sernaker and Michael Dineen who's been around here much longer than I have and is really an expert in especially our administrative data sets. So maybe some of our participants have already talked to him through our ticket system or who have gotten help in other various forms. We have our archiving assistant Andres who's here with us today also great behind the scenes get it making sure everything runs smoothly and working and then Alexandra Gibbons is our research aide. So I've gone through a list of people this is all just to say we have so many people and different backgrounds and different expertise and this isn't even this doesn't even touch on you know like the background like network of NDACAN folk who we also have. So we really do have an eclectic team with a lot of background and expertise who again are all here to help you. So NDACAN activities kind of like the sequence of our functioning. So we acquire data sets, we collect and assess child welfare data, along with that we protect so we protect the confidentiality none of our data all of our data go through processing and rigorous review processes to make sure we are really at minimal risk of any personal identification especially when we're talking about child welfare protection and protecting these key individuals these vulnerable individuals is really our highest one of our highest goals and things we strive for. We transform, so we create documentation to make data usable so we're not just you know passing along Excel files. We spend time to put together, you know, meticulous and thorough documentation and the form of users guides, code books, any supplemental documentation that we see would help users. So we put together whole data packages truly with the intention of the ease-of-use and understandability from the data user. We disseminate, so we license and distribute data sets in multiple formats so that's just us distributing our data appropriately. Some of our data do require more more requirements than others such as IRB's or additional data use agreements again all with the purpose of protecting these vulnerable populations.
Um we support we support child welfare researchers through direct one-on-one help through our events such as these, through our SRI the Summer Research Institute, to office hours like I said our key responsibility here is to support researchers, support users, support data depositors. We are here to help. And expand so expand the scope of child welfare research and as I've mentioned numerous times we do that implicitly and explicitly through events through just making connections in the community and encouraging research.
So acquire, we acquire administrative and survey-based data to publicly archive. We will serve anyone with child welfare related data who want to or are required to archive their data. As I briefly mentioned sometimes Grant requirements require the archiving of data and so we are an option if that is something you're required to do in your grant. So generally speaking we take data that is related to child welfare that could fall under the general umbrella of child welfare or you know people servicing the Child Welfare industry, or like we have data set on child welfare workers. So as long as it's you know touches on child welfare data, we would generally take it. There are some restrictions to that but the general umbrella of child welfare data.
Um so protect; so as I mentioned before we review all data submissions to ensure participant confidentiality and remove all PII. So for example some of our data you're probably familiar with has masked geographies. So for example in our AFCARS data set which I'll introduce in a little bit, we mask counties that have fewer than a thousand cases. Additionally we will do things such as recoding birthdays to make them not identifiable. And then we you know make permutations or adjustments to the data to retain things like dates or date spans or any other permutations small but to reduce risk of identifiability. Additionally to obtain data we don't just give data to anyone it's like hey I'm interested in your data send it to me no like we we do require users to submit an application. Some data do require an IRB, we have data use agreements. So all of these things again are with the goal to maintain the highest protection of these groups and children and acknowledging that these data, while hugely beneficial to the research community, are really sensitive and have you know like pretty serious data involved with them. But again we do our best that there's minimal risk of identifiability in our data sets and that is due to these rigorous review processes and high standards that we keep. So transform, as I mentioned before we don't just have the data we create users guides and codes books code books and this is all while we're putting it together we add content that we think will help the users. I process data a lot and I am always reviewing data as a potential data user and trying to think of things to add to the documentation to really help make using the data as easy and understandable as possible with as little confusion or misunderstanding. So things like that we've also added variables to some of our data sets to make things easier for researchers things that over the years we've noticed people just kept asking or came up a lot in research. So variables that we've created are things like age, age at the start of a fiscal year, age at the end of the fiscal year, we've created we've derived length of stay variables, we've categorized counties into a rural urban continuum code. These are all just examples of us creating variables again with the goal to help users and variables that we have seen that researchers are that are of most interest to researchers. Another thing we do to help users is we offered the data in multiple formats so we don't just send everyone SAS files for example which would drive me crazy. But we offer SAS,  SPSS, Stata, text files you know again if this if you can't work with one of those formats and you contact us directly we could you know that's something I could put together an R file for you. But by default we offer users four different formats of data when you receive a data package from us. And we provide sometimes syntax code so not just the data itself but helpful syntax in these languages to help read in the data and help transform like missing codes or add labels things like that. So again if and if there's anything out of scope from the documentation or the files like you should always feel free to reach out directly to us and we can provide help especially making sure you can at least open the data. So disseminate, so we license data to eligible eligible researchers so kind of what I was talking about before we don't just give the data to anyone. We distribute the data in multiple formats and we also conduct ad hoc special analyses. For example if you're not so technically savvy but you know you need a table derived from NCANDS or AFCARS for instance maybe like number of reports, number of child maltreatment reports by the report source by race. You know you need that but you don't know how to do that technically or you you don't know how to program you don't know programming languages. There are cases where you could reach out to us and what we call a special data request and we can all actually put those tabulations and numbers together for you. My list a fire truck went by okay I don't know if you can hear that it's pretty loud on my end. So if that's something you're interested in I would just I'm not going to go into too much detail but know that it exists and you can reach out with more questions about that and we can point you to the website that has more details with that. So support again we offer a lot of support we have a electronic mailing list serve so this would be like announcements from NDACAN we encourage scholarly scholarly discussion. I think this is a list serve that anyone could post to so like general research like any anyone who's on the list serve who's trying to initiate scholarly discussion or has postings that are appropriate to share. We have a course. 
[Erin McCauley]
Another good thing I want to jump in yeah and say too is our our list serve is also moderated, and so you really only get important information that gets through because anyone who wants to post you know sends the email and then we have a staff member look at the email and make sure that it's you know super relevant and appropriate for everyone before we send it out. So that's a real big strength of that   list serve  because I don't know about you guys but I have many  list serve I'm on where you just get too many emails and eventually you have to leave but this one's moderated.
[Sarah Sernaker]
 Yes thank you for adding that Erin because I am I'm not as familiar with the Lister I'm also on the community list serve which gets out of hand so it's good to hear there's moderation. So yeah as Erin said so it is more of a community like what is it like Town Hall but it is moderated so that you're not just receiving junk. So in addition to that we have this Updata e-newsletter so that's a quarterly newsletter that tells you of events going on, of any data sets that are released or coming up or like soon to be released. This is the reference database I was talking about before briefly this child abuse and neglect Digital Library we call candl so this is like a a database of citations and literature that uses our data so we can keep track of research, continuing research particularly for the data sets that we offer. And so users are interested data users can go to this to see you know what might have already been done with the data or you know it's also helpful when you're making data decisions and cleaning you can see what others might have done also. So in addition to that we have a summer Research Institute so this is by application. It's a week it's an intensive weekend long sort of Workshop where you apply with a project goal, you come with the intention of using one of our major data sets and basically you receive some intensive one-on-one help so this is really it's open to anyone professors, PhD students, grad students practitioners, but it's really great if you have a project and research goal in mind and you just get a lot of help, you get a lot of help and you make connections that have led to Publications and things in the past. Our summer training series you're experiencing live right now we have a really great hosting of presentations this year I'm really excited about it I think a lot more applied topics this year which is always fun and coding examples which I like 
[Erin McCauley]
We also collect all of the topics based on what the participants tell us is interesting or what they want to see more of and so at the end of the summer you'll get an email either from me or from Clayton Covington who's the host this year we'll be back next week and in that survey link it asked you know what subjects would you really like, or are there any sessions you'd like to see next year, and almost always all of our topics come from participants. So if you're seeing this year and you're like oh I really wish you know we had X Y or Z just let us know in the survey and we'll incorporate it into future years.
[Sarah Sernaker]
 Yes we really do take into account user or participant responses so please respond to your survey. As data researchers you should know the importance of surveys. But so where was I? So then we have NDACAN office hours which are once a month in the fall and they're just the first half hour is structured so you can come they're open rooms open Zoom rooms so you can come and go ask your questions about data or your research question or coding questions whatever. I'll I'm there I sit there and offer help so I know I've talked to users through the ticket system and I've met them through office hours so really great. And then the second half of the office hours are usually these informal presentations which you can join you could be a part of if you are a child welfare researcher. We do have office hours where we post our users and they put together small presentations and it's kind of a way to put your research out there again form connections and just do like a run through maybe on like a an oral presentation practice sort of. And then quickly.
[Erin McCauley]
I'll also emphasize that the the office hours are just over the top fantastic. In the next probably month we'll be announcing next year's topics for the workshops or presentations. And then I just want to re-emphasize what Sarah said about the ability to give presentations yourself. So we've had folks give research presentations but we've also had Dr Ivis King came and talked about how she uses NDACAN data for teaching and learning and she went through you know how she uses it in the classroom. And so if there's a topic that you're passionate about that you think our users would like you know also feel free to approach us because the office hours are are very informal and really about what is most helpful to the community.
[Sarah Sernaker]
 Yeah we've had some people come for like career advisement so they came and told us about how they got into their career and this was not just professors. I think we had Svetlana no Svetlana is a professor we had someone from CB we had someone from the practitioner's side um. 
[Erin McCauley]
Yeah we had a a workshop or a panel on non-academic data career. So basically just careers in in data and child welfare where we had folks from Mathematica and RTI and different more industry type research organizations so it's fantastic.
[Sarah Sernaker]
 Yes yeah office hours have been really great. But so those are like all of our formal stuff but we are always here to offer General support and offer resources. We have support and resources on our website so I put this link here [https://www.ndacan.acf.hhs.gov/user-support/user-support.cfm] we have past postings of previous summer Research Institute presentations. So things like data linkage like I don't think we're doing data linkages here we have plenty of past presentations on that and in various programming languages. There is a presentation last year I gave them like structural equation modeling. So we've gone we've done various different topics and we have all these resources and recordings and transcriptions on our website available right now. And this is our support email [ndacansupport@cornell.edu] so this creates like quote unquote ticket that someone on the other end will offer support. Sometimes that's Andres if it's questions of like licensing or whatnot if it's technical sometimes you'll be talking me. So this is this email is if you have a direct question you should always feel free to reach out like no matter how big or small we are really here to help.
So expand the last point. So we Foster interdisciplinary research teams as I've mentioned numerous times we've had people from all different fields, economics, child welfare, psychology, social work. Really again these we encourage it and we Foster it. So we've drawn researchers from new disciplines through conference workshops and presentations. I know Erin and Clayton are usually out on the conference scene doing poster presentations I believe particularly for NDACAN that is drawing in new people. We have these summer training webinar series which are open to the public and you know General most of the topics are General enough that we hope to attract new users, draw them in and show them what we can offer. And then our office hours which are also open to the public and although you might not have questions to come to office hours with, we have great presentations to see. So definitely if you're not already on our newsletter you can sign up for that and that's where you'll get notifications besides following our website and then we have Twitter. I don't do Twitter so I don't know what Twitter offers. Yeah 
[Erin McCauley]
We definitely have Twitter it's at @NDACAN_CU we advertise all of our events and then we even update for every data release or updates to Prior data releases.
[Sarah Sernaker]
Yes thank you Erin I was like I know we have Twitter but I don't know what gets on Twitter. But yeah so those are that's the basic NDACAN background so now I'm just gonna dive into some of the data sets that we offer especially our widely used ones. So just to jump right in. So a lot of our data sets and not all of them but a lot are federal data collection efforts. And so particularly we have administrative data and administrative data just means that it's data collected a lot of these are federally-mandated data collections so they tell the States you need to collect this information. So it's it's it is mandated to be collected but it's not collected with the intent of like statistical analyses or that it's more like record keeping. That's quote unquote administrative data but anyway so we have the NCANDS data system. So the first one at this National child abuse and neglect data system so we call that NCANDS. We have the adoption and Foster Care analysis and Reporting System we call that AFCARS. And we have the National Youth and transition database which is called NYTD and I have blurbs about all them which is why I'm not going into too much detail right now. So those are administrative data then we also have survey data so National and cross-site so. Our huge one is the national survey of Child and Adolescent well-being AKA NSCAW and so this is a longitudinal data longitudinal survey about two for our NSCAW folk there is NSCAW three which is coming out probably by the Fall. I just kind of noticed that my internet connection is unstable so bear with me let me know if you can't hear me at all. We also have the longitudinal studies of child abuse and neglect AKA LONGSCAN so this was a longitudinal study and followed children from ages 0 to 18 in the early 2000s I believe it was. And that one is really great because it just follows participants for such a long time and asks a lot of questions of them. And then we have the national incident studies of child abuse and neglect AKA NIS I guess I'll not say this one N I S I usually say. So this is these are probably our most widely-used data sets. We do have additional data sets just to name a few we have this Rockwool Duke Global child welfare data set so this one's relatively new this one is an aggregation of data from around the world trying to get a sense of children entering foster care or out of home care and also number of children experiencing child maltreatment reports from different countries. So there's information National level information from like England and France and the Netherlands for example. Another res relatively recent data set is this voluntary Cooperative information system so that's more of a historical data set looking at historical data of children entering out of home care I think it is mostly like foster care -related. Then another really popular one is the state child abuse and neglect policies database AKA scan so this one's really great. This one came out recently this one is great if you're using AFCARS or NCANDS particularly because it's state level information about or just how they word definitions what's included in a definition of for example a reporter for example. So it's like state-level sort of metadata on their policies and definitions related to child welfare and this can really help like help you understand the nuances and differences in how States report and record and collect data. The only caveat to that is there's only two years of data from 2019 to 2021. So it's really great data the only drawback is there's not more of it. So again this is truly just a handful of the data sets that we offer. If you go to our data if you go to our website and our data listings you can see we have so much data to offer and that can also mean it's hard to navigate. So again if you think we might have something to offer you feel free to reach out and we can help you know point you to a data set or help you Wade through our offerings to try to find one that would be relevant to your project.
[Erin McCauley]
 That's also a shout out a good time to drop by office hours so you can chat with us informally about these things. And next week we'll be talking about another new data set so come on back.
[Sarah Sernaker]
Oh yeah I should have put that here that's CCOULD data set. That one is Medicaid data from Florida and Kentucky. So very cool very very large data set. So that was like the high level just briefly touching on the data that we offer. So I'm gonna go into a few into detail for a few of our data sets these again are like our most or most commonly used data. Not to say that it you necessarily need to use it a lot of people use these data. So the NCANDS national child abuse and neglect data system. This is the compilation of child maltreatment reports across the United States so if a child has a child maltreatment report, for example with CPS, whether it's substantiated or not, that is ultimately a record that comes back through the NCANDS. So it's information about a report of child maltreatment including such as child demographics, what the alleged or substantiated abuse was, who the perpetrator was and their relationship with the child, information about the perpetrator, and this was a federally-sponsored National data collection effort and the whole point was to track the volume and nature of child maltreatment reporting. So it's to understand who's most affected whose experiencing these type you know the different types of abuse and it's all with the purpose of understanding it better so that we can inform policy and help again those most vulnerable and help you know allocate funding to where it's most needed. And so like I said data consists or sorry let me take a step back so the child file is the individual-level data as I was just talking about so those would be individual-level maltreatment reports and we have data that go back to 2000. So every year we get more and more data so for each year that is. So we have it going back to 2000. We also have what's called an agency file and that goes back to 2009 those are state-level Aggregates. So usually we direct people to child file because that is just a more granular look and if you have the savviness to deal with it we suggest dealing with the child file. Agency level is all state-level aggregate information. And we do have the code books for these online so I definitely recommend checking out the code books to see the exact contents and variables. But yeah so as I mentioned data consists of all investigations or assessments of alleged child maltreatment that received a disposition in the reporting year. So basically a child maltreatment investigation within a fiscal year. And the child file data are provided at the level of each child on a report so for every child that shows up on a report there is that is what defines a unique row. So each row is a child on an investigation and an investigation could have multiple children and children could show up on multiple reports. So it gets a little little tricky to deal with. But so the data include as I said demographics the type of abuse risk factors and then Services provided and then this this last point I have so Children's Bureau puts out what's called the child maltreatment report so if you're interested in NCANDS and want to get a sense of what's in it I would definitely check out the child maltreatment report that's accessible on the Children's Bureau website. And that is exactly derived from NCANDS data. It's just a bunch of state and National level summary statistics. Tables and statistics that I think are of most interest to policy makers and researchers so it's it's pretty curated.

So that's NCANDS and I know I'm looking at like a wall of text here so bear with me. So NCANDS is all of our child maltreatment data. AFCARS the adoption and Foster Care analysis and Reporting System is the federally-mandated data collection on children going into out-of-home care or most common or who were adopted. And that does not include private adoptions. So there are two files that come in AFCARS there's or sorry two different files we have the foster care data file and that goes back to 2000. So this is information on a child level so each row is a an individual child who was either in foster care or entered foster care at the start of the year. And there's information of demographics, race, removal reasons so why were they removed into foster care in the first place, their current settings so are they in a group home? Are they with it Foster family? Are they with a family a foster family of relatives or non-relatives? I think we have a distinction of the dates of removal, the dates of leaving foster care. So all the information regarding an out-of-home placement so that's the foster care file and then we also have What's called the adoption file. So each row would be a child who was adopted in the fiscal year and there's information again on demographics, and demographics on the adoptive parents. So for example we've seen a lot of research interest in like cross-racial adoptions comes up pretty frequently. We have dates such as when the parental rights are terminated and other information. Contents and variables are and these are separate data offerings so if you wanted the foster care file and the adoption file those are two separate data files that you would have to apply.
Maybe you could do on one Andres would know and all this. But they are two different data files. And the AFCARS collection and structure so AFCARS has been around for you can see since 2000 and I think even a few years before that but it wasn't clean and there was issues but Long story short AFCARS is undergoing a huge change after 2021. So they updated the fields and variables that they'll be collecting. So the change is going to the implementation and distribution of this new data what we call AFCARS 2021 might be slow to roll out because there's just so much to change and adapt to. But it's going to be hugely beneficial to researchers I mean this has like education of the child removed. So like what grade are they in? Are they currently going to school? Have they been removed with their siblings? It's just like a wealth of information and so once we get the infrastructure for, you know, intaking the data and distributing it this is going to be you know hugely beneficial to people interested in AFCARS. Five slides okay. So NSCAW so those two previously NCANDS and AFCARS so those are our admin data sets. So those are federally-mandated and collected at state level. The national survey of child and Adolescent well-being there's two cohorts so there's NSCAW one which was at the early 2000 there were five waves. There was NSCAW II which was three waves and that's about 2008 so about 2010s early 2010s. And then there's a third cohort NSCAW three this is only going to be two waves and right now we're in the process of cleaning and processing wave one. Wave one was collected in 2018 to 2021. They're currently collecting wave two so this is like very new and like current and ongoing. But the previous two nscaws are available to researchers. These are these are all longitudinal studies which examine child well-being and outcomes. There's information about families information about interventions and services. It's really like NCANDS on a granular level so it asks there are questionnaires asked of the child, questionnaires ask of the caregiver, questionnaires asked of the caseworker. There's even questionnaire I think in only NSCAW one asked of the teacher. So it's really trying to understand any risk factors, any fact or characteristics of the maltreatment report itself that there was any subsequent maltreatment. And then it follows them for years so it's also to understand you know the impact and the like the results are the outcomes rather of these children who have touched the Child Welfare the Child Welfare field or yeah. So this is sponsored by the ACF Administration for Children and Families which is part of Department of Health and Human Services. Research Triangle Institute so RTI are the people who actually do the data collection and are the quote unquote data depositors. And so I must have just kept this from previous years there's a research example here that's included of people using the NSCAW. But something about NSCAW is that we do not provide the code book and documentation on our website. It is available by request so please reach out if you're interested as a researcher and using NSCAW we don't have it publicly available because this is one of our more restricted data sets and so we want to make sure qualified researchers are accessing the materials. Trying to think there's so much to say about NSCAW but there's not the time to do so. There is a great page on our website for NSCAW support and I think under the user support and this has things about the survey design, survey weights, programming, using the survey and a bunch of stuff there's a lot of resources through our website alone for NSCAW. 
And then there's LONGSCAN which is the longitudinal study of child abuse and neglect so another survey this is
[Erin McCauley]
Sarah we just have a few minutes so after this I think we'll move to the Q and A okay?
[Sarah Sernaker]
Yeah yeah I think I'm almost done anyway so well I will be almost done.  So LONGSCAN is another survey. It follows children from 4 6 8 12 14 16 and 18. So it follows them for a long time. This was a survey given at five different sites so something I didn't mention is NSCAW is a nationally representative sample, LONGSCAN not so much. So there is a lot of great information here. It doesn't come with survey weights or anything like that so. And then here's an example of some research that uses LONGSCAN. And I believe there is documentation available on our website so if you want more info there be warned there are very many measures they use so there's a lot there's a lot to Wade through. And then we have VCIS our historical data. As I said SCAN has information on state definitions and policies and that is co-sponsored or what I don't know Co-released with Mathematica. So they actually have a site for SCAN just if you come across it it is the same data. So let me just stop there before I keep going. So again I'm Sarah here's my email address directly but again if you email the NDACAN support I will see it there too. And then next week I'll just leave this up while we end there and go through questions.
[Erin McCauley]
 Great Sarah thank you so much. Alex is here and she's going to moderate the Q and A for us but before you move to that I just want to again hype next week we hope to see you back and then give a big thank you to Sarah for leading today's presentation. And to Andres who's been answering some of the questions in the chat and he's really the person that does all of the behind the scenes work to make this series run at all and to turn it into that webinar Series. So big shout out to both of you and thank you to our participants for being here. Alex do you want to take it away with the Q and A?
[Alexandra Gibbons]
 Yes thank you Erin and thank you all for being here today. So our first question in the Q and A is: the definition of neglect is confusing neglect means failure by a caregiver either deliberately or through negligent or inability to take those actions necessary to provide a child with minimally adequate food, clothing, shelter, Medical Care, supervision, emotional stability and growth or other essential care. To make any consistent analysis my question is what does adequate care mean quantitatively or qualitatively?
[Sarah Sernaker]
 Oh that's a great question. So that's a question that I don't know that anyone can answer. So this is the the thing that comes up when you're working with these data sets especially like NCANDS and AFCARS each state defines things differently and sometimes vaguely. And that's really up to you as a researcher to try to understand and at least acknowledge that there are differences and that's something reading the child maltreatment report for example with NCANDS particularly can help understand because there are little blurbs by the states that you know speak to any differences they might have in definitions and Reporting. So that's one thing to try to understand the differences. Another is using our SCAN data set which helps like I said to try to understand how States May Define things differently and how their policies are affecting the data they collect and these sort of nuances. But at the end of the day we have to sometimes just accept that there are differences and make an assumption that the differences aren't significant enough to you know affect analyzes like at a national level for example. But this is something again if you're working with a particular state it might even be worth reaching out directly to the state. This is something that just you know it is something to keep in mind working with our data. It's we're talking about like legal definitions here so I think sometimes they are purposefully vague. That's my naive interpretation. Erin do you have anything?
[Erin McCauley]
 No I think that's great I was gonna say the SCAN data has definitions which I think are super helpful but also you know one thing that we discuss a lot here and especially during office hours if you want to continue the conversation I recommend coming by is the trade-offs between having such an incredible breadth of data in the administrative data sets that we really can talk about you know huge time Trends in in the whole country. But what we lose for that breadth is some of the depth of data that allows us to parse you know these exact differences in a way that's I think easier when you're potentially doing a different type of data collection like focusing on one County for example. And it's just kind of part of what we wrestle with. But you know the reason one of the reasons Sarah went through the staff at the beginning is because you know to share a little bit about our backgrounds and the fact that most of us are academics or are researchers who do research in this area and so we grapple with this in our own work in addition to our work with the archive. And so we're definitely always up to continue this conversation during office hours.
[Sarah Sernaker]
 Yes that's a great office hours discussion.
[Alexandra Gibbons]
Thank you so much for that answer. We have one last question about the NSCAW data. So the question is are the separate numbered NSCAW data sets different children?
[Sarah Sernaker]
Um yes that's a great question. So they are totally different children. So NSCAW one is just a totally different set of children than NSCAW two and they're not linkable between them. You could do if you're working with NSCAW one and NSCAW two you could do like cross analyzes sort of but given the differences in the survey design it's a it is a little tricky and requires like pretty careful consideration. But the short answer is they are different children in the different NSCAW data sets.
[Erin McCauley]
All right well those are all of our questions so thanks everyone for coming by today. Next week we'll be talking about a new data set that CCOULD data so we hope to see you back next week or later in the summer.
[Sarah Sernaker]
Yes thank you all.
[voiceover]
The National Data Archive on Child Abuse and Neglect is a collaboration between Cornell University and Duke university. Funding for NDACAN is provided by the Children's Bureau an office of the Administration for Children and Families.

[musical cue]