[Archival Research: Using Data in Social Science Research]
SEAN KELLY: My name is Sean Kelly.I'm Professor and Chair of Political Scienceat California State University Channel Islands.My area of specialty is American politics, specificallyAmerican political institutions.Most of my work is on the study of Congress and the presidency.My research, which is both qualitative and quantitative,
SEAN KELLY [continued]: draws very heavily on archival resources.I'm the co-editor of a book calledDoing Archival Research in Political Scienceand a member of the National Advisory Boardfor the Dirksen Congressional Center.In this video, we're going to focus on three topics.
SEAN KELLY [continued]: Using archival data-- once you have your data in hand,what do you do with it?Citing archival data-- just like any other research,your research may need to be replicated by others.They need to have the information on how to do that.Finally, some common critiques of archival research
SEAN KELLY [continued]: and how to respond to those critiques.[Using Archival Data]So how do you use archival data?Well, archival data, which is qualitative,it's words written down on paper,
SEAN KELLY [continued]: can be turned into quantitative data.This is something that my co-author and Ihave done quite frequently.For instance, back in the 1970s, a political scientistwrote a book about how members of Congressget their assignments to committees in Congress.
SEAN KELLY [continued]: Well, he used some data to show what committees were requestedand which committee members were assigned to.He wrote a book, and that was the end of ituntil we came along and we said these data are
SEAN KELLY [continued]: going to be in archival collections,let's go get more better data.And so we found the request that individual members of Congresswere making and turned them into data.We had data on what they had asked for,and we had data on what they actually got.
SEAN KELLY [continued]: And so we were able to address the questionof whether or not members of Congressare very successful at getting committee assignments.Now that may sound kind of boring,but in the study of Congress, it'sactually a very important questionbecause the committees are the places where policy is made.
SEAN KELLY [continued]: One of the live questions in the study of American politics,specifically the study of the presidencyis whether presidents are legislative leaders.Can they get legislation passed that they'reinterested in passing?President's actively engage with members of Congress
SEAN KELLY [continued]: to try to get their support for their policies.The question is whether or not those presidents are actuallysuccessful in what they're doing.By looking at some of the papers from the presidency of JimmyCarter, we were able to identify which members of Congresshe was talking to, who changed their mind,
SEAN KELLY [continued]: and who didn't change their mind.And the finding was kind of interesting.That is that Jimmy Carter, who's typicallyseen as a fairly weak president, wasvery effective at getting members of Congressto support him on a particular set of votes.
SEAN KELLY [continued]: So you can take qualitative data and turn itinto quantitative data and then usetypical statistical kinds of techniquesto address those data.Now archival data can also be used as qualitative data.
SEAN KELLY [continued]: So no numbers, just words.Archival resources are particularlyuseful for determining how a process works, whatthe order of events was, who the most important actorsin a particular process were because much of that,
SEAN KELLY [continued]: if not all of that, is going to be written down on to paper.By using letters and memos, you'reable to bring life to what sometimesis a pretty lifeless process when you read it in a textbook.When you actually see it written down on a page,
SEAN KELLY [continued]: it becomes all the more lively.It really breathes life into somethingthat maybe is just boring old history or boring old politics.For instance, here's a letter that's in a collectionthat we hold here at CSU Channel Islands.
SEAN KELLY [continued]: In this particular letter, the writeris writing to his member of Congressabout a piece of civil rights legislation thatis before the Congress.And as you can see by reading this letter,this constituent has some very strong feelingsabout his member of Congress and how his member of Congress
SEAN KELLY [continued]: should vote against civil rights legislation.In reading through this letter, youget a real sense of how strongly the opinionis held by the constituent.And it gives you some real insightinto the politics of the 1960s in a way that if you just
SEAN KELLY [continued]: read in a textbook that there were many people whoopposed civil rights legislation for a variety of reasons.When you read this letter, you reallyget a sense that, oh hey people really did feel stronglyand here's a person from northern Californiawho's expressing these ideas, which to many of us
SEAN KELLY [continued]: in this day and age seem antiquated.We've talked about using archival datafor quantitative research.We've talked a little bit about it for qualitative research.Where I find it's most useful, however, is in triangulating.That is by bringing both the qualitative
SEAN KELLY [continued]: and quantitative together.For instance, if we're looking at a quantitative question,did the president influence members of Congress?Well, we can run the numbers and produce coefficients,and they will answer the question quantitatively.
SEAN KELLY [continued]: But by turning to the qualitative to the letters,to the descriptions of what was going on at that time,we can either confirm that yes, in fact,the statistical results we're seeingare consistent with how people were observing it at the time
SEAN KELLY [continued]: or if we turn to the qualitative dataand they seem to contradict the quantitative data,then maybe we need to go back and takea look at our quantitative data to make surethat the results we're getting are really real.So the more angles you have on your results,
SEAN KELLY [continued]: the more you bounce the qualitativeagainst the quantitative to get a better sense of the storythat you're trying to tell in your research, the moreconfidence you're going to have in your findingsand the more confidence people aregoing to have when they read your research.[Citing Archival Data]
SEAN KELLY [continued]: It's really important with archival research justlike any other kind of research that you properly citeor document your research effort.That means providing enough informationto other researchers so that they can understandand maybe replicate what it is that you've done.
SEAN KELLY [continued]: This means keeping careful track of where specific materialscame from, what box, what folder, what university, whatcollection.All of those things are important.But it's also the case, that whenyou report that information, sometimesit's helpful to another researcher down the line.
SEAN KELLY [continued]: When I was working on some researchabout congressional committee assignments,I came across an article from the 1970swhere somebody had done something similar.I wrote to the archivist at the University of Georgiaand said this researcher did some research on your campusabout 20 years ago, and I'm trying to get the same data.
SEAN KELLY [continued]: Well, she went and checked against the requests thatwere made 20 years before, and she said, well,I know the exact boxes and foldersbecause they were written down during his research process.And within days, the documents were sitting before me.So what that means is that if you document your research,
SEAN KELLY [continued]: you may be helping somebody in the future.You may also be helping yourself because you'llknow where to go back and look when you inevitablylose that thing that you were so interested in.It's also important to document your researchfor qualitative purposes.In a project that I'm working on right now,
SEAN KELLY [continued]: I'm using a lot of letters from constituentsto their members of Congress.And if somebody wants to see these letters,I'm not only going to need to provide a box and a foldernumber, but I'm also going to needto provide the name of the person writing the letter,who the letter was aimed at, and what the date was
SEAN KELLY [continued]: because when it comes to these letters,it's one page out of hundreds.And we don't want to read back through all those hundredsof pages, we just look for the person and the date,and now we found what we're looking for.A couple of other things to be aware of.One is copyright restrictions.
SEAN KELLY [continued]: When a person writes something thatbecomes their intellectual property.And we have to be careful about using other people'sintellectual property.For most of the collections that I work in,because they are the papers of people who are public figures,
SEAN KELLY [continued]: the issue of copyright is a little bit more broad.So if a member of Congress working on the public dimewrites something, that's pretty much public information.But if there are private documentsin a collection like constituent letters,
SEAN KELLY [continued]: you're going to want to take a look at copyright issuesto see if you can use that material.Also if you are going to be publishing materialon to the internet, you should probably get permissionfrom the repository to post that information on to the internet
SEAN KELLY [continued]: because they may have some copyright claim.The second issue is confidentiality.When it comes to these collections, many of themcan be quite old.Anywhere from 10 years old to 100 years old or even older.For the most part, when it comes to confidentiality,
SEAN KELLY [continued]: if the document is outside the area of about 40 or 50 yearsold, there's probably no need to be treating itwith strict confidentiality.And certainly, if it is a public figure,there's no need for confidentiality.You just really have to use your judgment here.
SEAN KELLY [continued]: And in using your judgment, ask yourselfif revealing this information coulddo any harm to an individual, thenyou might just want to cite it asconfidential as an anonymous source.[Critiques of Archival Research]
SEAN KELLY [continued]: There are several critiques of archival researchthat you should be aware of and beable to talk about and defend your work against.One is what I like to call the back to the future argument.If you remember in the movie Back to the Future,
SEAN KELLY [continued]: the character Michael J. Fox goes back in timeand sends himself a letter in real time suchthat he can change all the events thathave happened in the movie.Some people argue that what goes into archival collections, what
SEAN KELLY [continued]: a politician or their staff puts into those folders kind of actsthat way.They're sending messages to the future.Just like in the movie, this is kind of a fanciful argument.That somebody would think 30, 40, 50 years ago
SEAN KELLY [continued]: that some academic is going to be looking at their papersand they're going to shape the arguments of the futureis a little bit of a circular kind of an argument.In the same way that if you really stop and think about it,Michael J. Fox could not send himself a letter
SEAN KELLY [continued]: into the future.It's an almost conspiratorial kindof approach to criticizing archival research.In reality, as somebody who worked in politics for a while,you're so busy doing the things that you do that you're notthinking that you're making history or much less
SEAN KELLY [continued]: that you're going to shape future history.Another critique is that these collections are very one-sided.And I think that's fair to a point.Collections are one-sided in that the person or the group
SEAN KELLY [continued]: that has created these documents are definitelythinking about the world from their unique point of view.But there are ways to avoid telling one side of the story.And the way to avoid it is by finding another collection thatperhaps takes a different point of view or at least
SEAN KELLY [continued]: comes from a different point of viewso that you can run these two collections against one anotherand try to suss out what the real story is.But you do have to be careful that youare hearing only one side of a story in a collection.
SEAN KELLY [continued]: It is a reasonable critique, but notone that can't be overcome by just doing a little bitmore work.Another argument against archival researchis that the collections are incomplete.That is how do you know that you have all of the data that
SEAN KELLY [continued]: was generated in this process?And my answer to that is usually a pretty flip, welldo you have better data?Usually, the person who's critiquing you will say,well no I don't.Archival research is a little bit like that climate science
SEAN KELLY [continued]: you hear about from time to time where some scientists willsay go to Antarctica, and they'lldrill down into the ice.And this ice is hundreds and thousands of years old.And they'll pull out a core.And they'll go through the various layers of the coreand collect the data about the climate across time.
SEAN KELLY [continued]: Now that core that they took is incomplete.They didn't call the whole of Antarctica.They only cored a very small part of it.So isn't it incomplete?And the answer is well yeah, it probably is,however, it is better to light a candle than to curse the dark.
SEAN KELLY [continued]: We have some information, and that information can be useful,and we can learn something new from it.And next year, somebody else can go to Antarcticaand take another core, and we can look at that one.It is important to remember that social science is justlike science, it's a process and each time
SEAN KELLY [continued]: we take that new sample, each timewe dig into that new archive, we have an opportunityto enter into another discussion about what we think ishappening in a social process.[Conclusion]
SEAN KELLY [continued]: In my view, archival research is a very plastic,a very malleable type of research approach.It can be used for quantitative research.It can be used for qualitative research.It can be used in multi-method research where we triangulate.
SEAN KELLY [continued]: It can even be helpful for researchthat's based on interviews.While there's no doubt that archival research canbe critiqued in some ways, being incomplete,it gives us insights into processes that we really
SEAN KELLY [continued]: can't tap into any other way than by going into these papersjust like somebody taking an ice core in the Antarctic.There's no other way to answer that question.And so we do it, and we hope for the best.Here are some questions that you mightthink about and some activities that you might engage in.
SEAN KELLY [continued]: Do you imagine that you would usea quantitative or qualitative approach to your research?What materials in a collection you are familiar with mightfit with that approach?Choose the approach that you didn't choose in number one.That is maybe you said that you wouldtake a quantitative approach, now think about your research
SEAN KELLY [continued]: from a qualitative perspective.Imagine and describe and how the alternative approachmight actually improve your research.Describe the strategy of triangulationthat I talked about a little bit before.Do you think it might strengthen an archive of research design
SEAN KELLY [continued]: or do you think it's unnecessary?Explain your opinion.Which of the criticisms of archival researchthat I just talked about, do you find most compelling?Why do you find it compelling?Explain yourself.
Prof. Sean Kelly explains using archival research in the social sciences. Kelly discusses citing archives and critiques of archival data.
Looks like you do not have access to this content.
Prof. Sean Kelly explains using archival research in the social sciences. Kelly discusses citing archives and critiques of archival data.