<body><script type="text/javascript"> function setAttributeOnload(object, attribute, val) { if(window.addEventListener) { window.addEventListener('load', function(){ object[attribute] = val; }, false); } else { window.attachEvent('onload', function(){ object[attribute] = val; }); } } </script> <div id="navbar-iframe-container"></div> <script type="text/javascript" src="https://apis.google.com/js/plusone.js"></script> <script type="text/javascript"> gapi.load("gapi.iframes:gapi.iframes.style.bubble", function() { if (gapi.iframes && gapi.iframes.getContext) { gapi.iframes.getContext().openChild({ url: 'https://www.blogger.com/navbar.g?targetBlogID\x3d14462747\x26blogName\x3dLIS+950:+Libraries+and+community\x26publishMode\x3dPUBLISH_MODE_BLOGSPOT\x26navbarType\x3dBLUE\x26layoutType\x3dCLASSIC\x26searchRoot\x3dhttp://lis-950.blogspot.com/search\x26blogLocale\x3den_US\x26v\x3d2\x26homepageUrl\x3dhttp://lis-950.blogspot.com/\x26vt\x3d1224404710664714099', where: document.getElementById("navbar-iframe-container"), id: "navbar-iframe" }); } }); </script>

LIS 950: Libraries and community

The purpose of this seminar is to explore an important topic in library and information studies in depth — in all its intertwined historical, cultural, philosophical, and political aspects — through a graduate reading/discussion seminar. The topic varies each time the course is taught; this time around, we will focus on "libraries and community"

Tuesday, September 06, 2005

Those Census Records

OK, this is as much a heads up as anything. On page 182, Williams cites "Groff" (and the name is listed this way in the end notes, too) and his work to help validate his use of these census figures. I have to admit, I was initially a little perturbed that Williams got the name wrong. Harvey Graff is a major (MAJOR) figure in literacy studies. His book The Literacy Myth has become a current (well, 1979) foundational text in my other field of study. But ultimately, what bothers me more, is that Williams, after discussing Wisconsin Census Data, uses Graff to validate this work, stating only that Graff worked on a different period and a different area. The fact of the matter is that Graff's census data was not U.S. census data. Graff's work used Canadian census data.

So my question is, methodologically, is this a fair comparison? Can Williams validate his use of U. S. Census data for this study based on the validation of Canadian census data? How does/can this knowledge affect how we view Williams' study? Or doesn't it?

8 Comments:

Blogger Soojin Park said...

In his article, William uses three research cases which are not related to libraries or LIS in order to justify his usage of the census data. His intention is that there are already research works that are based on data that are some errors but assumably not serious. He tried to get validation of his work, not the census data, from previous practices.

Rather, I'm questioning how to deal with errors in data. IS there any agreement about which extent data errors can be acceptable?

11:01 AM  
Blogger k8 said...

I think that ultimately I wanted more explicit explanation for how he was using these other studies to understand and work with possible errors in the data he was using. Just a sentence or two would have been great. However, since he doesn't seem to clearly explain how he is doing this, I am suspicious. I probably don't need to be, but it is this type of omission that sends up warning flags for me. I guess I'm not a very trusting reader.

1:17 PM  
Blogger Jom said...

I think so, and also I am wondering that by pursuing a national-level data to apply to a study, rather than looking at geographic differences in various parts, what else could be considered necessary, and why? Can any upcoming and changing differences and factors reflect a condition in a particular region and in “social conditions theory”, and then how can the theory be visualized by us as a student and a researcher?

5:58 PM  
Blogger Barbara Walden said...

Re: the Williams article
I had several quesions about this article, both as to conclusions and methodology.

First, what is the definition of a "public"library in this period? I think he glosses too easily over the fact that what we might call "private"libraries are sometimes part of his data. He fails to take sufficient cognizance of the change in status from slave to free for some Americans during this period, or give any attention at all to the potential impact of the Civil War on libraries.

He gives little attention to regional differences, which I think (based on what I read in Wayne Wiegand's Books and Reading class)were quite marked in this period, and I'm not really sure that the data he used from the various censuses are actually comparable (maybe they are ?)

I thought his conclusions were mundane: industrialization, education, -- these are kind of big, standard explanations for lots of things in the 19th century and so I wished he had been able to tease more out of his data.

What this article suggested to me is that, if a quantitative approach is what the researcher wants, perhaps this approach might be more fruitful on a more micro-level in an older time frame, rather than using national-level data so unquestioningly.

And, to answer the question, I think that if this author was using Canadian census data without telling his readers, then he is being dishonest.

I know I'm making strong comments here but of all the things we've read in our LIS seminars til now, this article has bothered me the most.==Barbara

8:26 PM  
Blogger Jacob said...

My simple answer to the second questions is that Williams cannot validate his use of U.S. Census data based on validation of Canadian census data by another researcher. This knowledge weakens the foundation upon which Williams can set his conclusions. If we connect validity with reliability it seems even Williams agrees that we cannot depend on other researchers as he says, on page 181, "Given these discrepancies for both years, it was concluded that the issue of reliability cannot be tested satisfactorily using the published work of other researchers."

8:31 PM  
Blogger Jacob said...

Following up on Barbara's post.

I also wondered why Williams did not even discuss the possibility that the Civil War might have some impact on libraries in the US that would show up in the 1870 Census. Related to this, Williams does not note the large jump in the 1870 Census mean number of Sunday School and Church Libraries per 10,000 Pop. from the 1850 or 1860 Census. This article bothers me but it illuminates the shortcoming in quantitative research when many questions surround the data set itself.

8:40 PM  
Blogger Awa said...

Oh does Williams really use the work on the Canadian census data to validate his own? It’s unbelievable! Maybe Graff uses the Canadian census data mostly but mentions something about the reliability of the US data set?

It seems that people are highly critical about this article! :P Well, as far as I’m concerned, I tend to view it as a text to understand the use of quantitative methods in historical research. And of course, the quantitative method is far from perfect in understanding humanity issues. If conbining with the analysis of the real social conditions at that time (impact of civil war, for example, of course), it might get much less criticism.

10:23 PM  
Blogger Kate said...

Whoa, what a blow to the research if indeed, the work of the Canadian census was used to validate his work. I, too, found the variables (education, mass communications, etc.) less than satisfactory. Of all the readings for the week, this one caused the most frustration in terms of the findings. One of Williams' conclusions is "Library development is a complex phenomenon." Well, I'm pretty sure we all could have came to the same conclusion without the census data.

12:36 AM  

Post a Comment

<< Home