Middlebury College: A Stand Against Wikipedia

Parent Cafe
61

<p>Stockmarket, the high income/high test scores correlation is valid over a large population but not necessarily for small pockets within it. I have known many families who are low income/ high education/ high parental involvement in children's educaton, and they will blow out any study. When you live in a university neighborhood, you will see this commonly but in the general population that is not the case. With many immigrant groups, the same can hold true, as many are educated, but without language skills and credidation to work jobs at their educational level here in the US. Or came from families with high educational values that are passed down but when educational opportunities from their native countries were closed.</p>

62

<p>


</a>
If so then that graph is crap. That whole site is a proponent of racism and segregation.</p>

63

<p>
[quote]
But how on earth, a low income white/asian student score better than a higher income URM, that does not make sense to me.

[/quote]
</p>

<p>Abigail Thernstrom is one scholar who has researched why that happens. She has a book </p>

<p><a href="http://www.amazon.com/No-Excuses-Closing-Racial-Learning/dp/0743204468%5B/url%5D"&gt;http://www.amazon.com/No-Excuses-Closing-Racial-Learning/dp/0743204468&lt;/a> </p>

<p>that is quite controversial because it points out facts like those. Not everyone agrees with her discussion of why that gap happens or what to do about it, but it is a genuine phenomenon. </p>

<p>For what it's worth, my son scores much higher than you would expect from our family income. It's a matter of family priorities in our family.</p>

64

<p>Some observations:</p>

<p>(1) First of all, re the fluid nature of Wikipedia: at the bottom of each page, there is a notation like "This page was last modified 20:56, 23 January 2007." (obviously the date/time will be different for any entry). I think that any citation of a web site -- and it goes for any site, not just wikipedia, should include both date of retrieval or last modified date, if available. Because it is a wiki, that means that historic versions can be retrieved -- so for example, I grabbed that date from the wikipedia Middlebury College entry. If I was looking at a paper that referred to the page as of 14 October 2006, it would be a simple matter of going to the history to pull up that date. So, here is the Oct. 14 entry: <a href="http://en.wikipedia.org/w/index.php?title=Middlebury_College&oldid=81467134%5B/url%5D"&gt;http://en.wikipedia.org/w/index.php?title=Middlebury_College&oldid=81467134&lt;/a&gt;&lt;/p>

<p>and here is the Jan. 23 entry:
<a href="http://en.wikipedia.org/wiki/Middlebury_College%5B/url%5D"&gt;http://en.wikipedia.org/wiki/Middlebury_College&lt;/a&gt;&lt;/p>

<hr>

<p>(2) My daughter included a Wikipedia reference in a paper she turned in last fall. I was surprised, but I did think that the reference was appropriate in the context she used it. It was a very in-depth paper that cited to a number of original sources; at the end of the paper she had two sections - one was "works cited"; and one was "works consulted". The Wikipedia reference was under the "works consulted" section. It was also clearly on a tangential matter, not the central thrust of her paper. That is, if she was writing a paper about apples and along the way ran across original source references comparing apples to to other fruits, including oranges.. then wikipedia was used to get some background on oranges. (Her paper was about history, not fruit -- but the point is that wikipedia was "consulted", not "cited", as a reference on a collateral issue). </p>

<p>It seems to me that it makes sense to "consult" Wikipedia in that context, especially when (as in my daughter's situation).</p>

<hr>

<p>(3) I edit sometimes on Wikipedia. Admittedly I started editing out of frustration with some inaccuracies in a particular article that I wanted to fix; but editing has made me very aware of their standards. </p>

<p>Because of copyright concerns, images posted on Wikipedia must have attribution; if not they are subject to removal. The SAT chart that was referenced before is here:
<a href="http://en.wikipedia.org/wiki/Image:1995-SAT-Education.png%5B/url%5D"&gt;http://en.wikipedia.org/wiki/Image:1995-SAT-Education.png&lt;/a&gt;&lt;/p>

<p>That page has the notation,
Made in Excel from this source: [1]</p>

<p>The footnote refers to:
<a href="http://www.lagriffedulion.f2s.com//testing.htm%5B/url%5D"&gt;http://www.lagriffedulion.f2s.com//testing.htm&lt;/a&gt;&lt;/p>

<p>This is the same as Mathmom's "scouting around" mentioned in the post above; I just wanted to explain exactly how that scouting can be done. </p>

<p>Having the source doesn't tell us whether it is valid or not, but it is probably BETTER info than you would get from many other sources,such as a news article or another web site, or a Britannica article. If you are on Wikipedia and you see a graph or chart or picture that is NOT sourced in some way.... well, what you ought to do is remove it, or at least tag it as unsourced. (At least once you are comfortable with the way Wikipedia and their tagging system works). </p>

<p>Wikipedia has a lot of problems, and as it has gotten more popular, it has also started to draw more vandals and jokesters. Nonetheless I have found it to be a overall a very good source of general information. I don't think it can be ignored. Whether or not college students should cite to it depends, as noted above, on the context of the citation. </p>

<hr>

<p>(4) Wikipedia encourages its authors to cite references, so one very nice thing is that you can go to a Wikipedia article, scroll down to the bottom, and have a lot of links that take you to other online references, including journal articles. So it is a great starting point for anyone doing research. Let's say, for example, that I was writing a paper about some aspect Global Warming -- I could go to the article at <a href="http://en.wikipedia.org/wiki/Global_warming%5B/url%5D"&gt;http://en.wikipedia.org/wiki/Global_warming&lt;/a> - get a good overview of the issues, and then start clicking the links to the notes & references at the bottom which seemed most promising in terms of reliability; for example, "^ Earth System Research Laboratory, Trends in Atmospheric Carbon Dioxide. National Oceanic & Atmospheric Administration. Retrieved on 2007-01-18" looks like a good bet. </p>

<p>I think the harm in a college "banning" Wikipedia is that it could cut off students from a valuable entry point to research. Also, one of the advantages of the collaborative nature of Wikipedia is that articles sometimes do a very good job of highlighting issues of controversy. I think it would be much more valuable to spend some time educating students on understanding how and when to use Wikipedia. "Citing" is better than reading it and not citing, because one way or another it is going to be used as a source of information -- when it is NOT mentioned, then what you get potentially is simply unsourced misinformation. At least if some prof. discovers through student work that there are inaccuracies in a particular article, the prof can go online and correct the article.</p>

65

<p>The SAT graph data also came from this
<a href="http://en.wikipedia.org/wiki/Education_outcomes_in_the_United_States_by_race_and_other_classifications%5B/url%5D"&gt;http://en.wikipedia.org/wiki/Education_outcomes_in_the_United_States_by_race_and_other_classifications&lt;/a&gt;&lt;/p>

<p>which in turns come from this
<a href="http://www.arthurhu.com/index/sat.htm%5B/url%5D"&gt;http://www.arthurhu.com/index/sat.htm&lt;/a&gt;&lt;/p>

<p>It claims data Source: 1995 College Board SAT Profiles</p>

<p>So far I cannot find the original source. I cannot find any challenge to these data either.</p>

66

<p>I can understand why historians, whose discipline often values engagement with primary sources and rigorous documentation, would have qualms about Wikipedia. But just today I came across a reference to a source on Wikipedia that I have never seen cited in print ever (it was to a small, privately printed edition). If this obscure book were only a little more central to my current project, it would've been the case that NOT consulting Wikipedia would have meant incomplete or flawed work.</p>

67

<p>Tokenadult:</p>

<p>It is also a matter of parents' educational levels. This is why in Cambridge, MA, reliance on economic data alone would be very misleading, as graduate students from MIT and Harvard often qualify for F/R lunch for their children. This puts them in the same category as new immigrants with limited English and children of parents with low income who may or may not have completed high school. Obviously that is wrong. And, from what I know about you, it would be wrong about your family as well. Ditto Mini and other posters on CC.</p>

68

<p>I don't think anyone is suggesting that banning Wikipedia is a good idea. But it should never be cited as an authorittive source because it is open to pranks, as Calmom states. Or simply sloppy scholarship. It's a good starting point for research & provides some valuable links.</p>

69

<p>Hi, Marite, I think what is considered anomalous is not so much high scores among low-income people (for me the archetypical example of those is scores of children of first-generation immigrants from China) but rather low scores among high-income people. If a family has resources to pay out of pocket for tutors, to buy books and to subscribe to magazines, and to otherwise raise the children's chances to score well, what's with the low scores? </p>

<p>To everyone in the thread: the PRIMARY source for data about family income and ethnic characteristics of SAT-takers is the College Board. Everyone who takes the SAT has the opportunity to fill out self-reported ethnicity and self-estimated family income data. The College Board gathers data like that with every administration of the SAT. I am not aware how often the College Board releases reports or studies of those data.</p>

70

<p>College Board releases the data, including correlations between income bands and mean SAT scores every year. Here's a link to download this year's national report. The correlation between income and scores is dramatic, to put it mildly.</p>

<p><a href="http://www.collegeboard.com/about/news_info/cbsenior/yr2006/reports.html%5B/url%5D"&gt;http://www.collegeboard.com/about/news_info/cbsenior/yr2006/reports.html&lt;/a&gt;&lt;/p>

71

<p>Re: post #70, yes, the national report every year has a few tables showing aggregate ethnic data about scores, but not the kind of table that shows income variations and ethnic categorizations on the same table, as in the table that sparked this partial thread-hijack. So far this is an example of Wikipedia including at least some data that are hard to find elsewhere.</p>

72

<p>Oh, and by the way, the better way to display bivariate data is with a bivariate plot (a "scatterplot"), not with the line graphs or tables that we are mostly seeing as examples in this thread.</p>

73

<p>re #70
Yes, but there is no income together with race data in this report or any report since 1996. Apparently the last one with that data is 1995 (which everyone is quoting) but it is not in the archive. I guess the data is too explosive so they stop reporting it after 95.</p>

<p>Another interesting data in the profile is the low SAT score of those who plan to major in education. No wonder our education system is in trouble.</p>

74

<p>Yup. Actually, there was a College Board scientist who, in the early 1990s, came up with an algorithm for correcting scores for income, race, and parental education bias. Worked, too.</p>

<p>But the main customers wouldn't have been too happy. He was fired just a few months after he completed the job.</p>

75

<p>Mini, a scatterplot would show (visually) why an "algorithm" like that produces junk results: there is variance in SAT scores at all levels of income. Central tendency is one thing, but there is no valid way to "correct" individual scores in that way. </p>

<p>Much the same applies to correlations between IQ and income, etc., etc., etc.</p>

76

<p>"Central tendency is one thing, but there is no valid way to "correct" individual scores in that way."</p>

<p>You should tell that to the CollegeBoard. They paid very good money for their research scientists to have it done, and they did it. Your beef should be with them, not with me.</p>

77

<p>Well, I presume that's why College Board doesn't make any such "corrections," because upon further analysis they found out that wouldn't make sense. </p>

<p>Do you happen to have any links to news stories about the guy who attempted to develop a correction formula? That would be something interesting to look up.</p>

78

<p>No, they discovered the opposite, and realized what it would do to their customer base. (You've read Nicholas Lemann's books, and the interviews and articles that followed, I presume.)</p>

<p>(But as to the validity, since research hasn't held up any significant premises of the SAT other than the one they claim - a predictor (but not a very good one, when it comes to minority students) of first-year college performance, I'm not sure it makes a heck of a lot of difference.)</p>

79

<p>I am a regular reader of The Atlantic, so I've read a lot of Nicholas Lemann's articles over the years. I've just requested The</a> Big Test from my local public library. I'll see what he has to say about these issues in the book that I haven't already seen elsewhere. </p>

<p>I'll quote here Stats:</a> Data and Models, a pretty good "reform" statistics textbook that I have at hand. </p>

<p>


</p>

80

<p>To go back to MarathonMan's comment that "Colleges at Middlebury's level should be teaching students how to evaluate sources." I think that is precisely what Middlebury's academic departments want to do. The informal, broad guidelines on the Midd library resource page provide only links to the care and feeding of internet sources and are out-dated, vague, and obviously not enough to do the trick. Professors are compelled to come to grips with wiki precisely because the nature of wiki is changing and proving to be a tremendous tool for the exchange of information- scholarpedia, which I mentioned before, is a fascinating offshoot. Midd's policy to bar the use of Wikipedia is not clearly not a ban - the distinction is quite a bit more than academic - students are warned that they cite wiki "at their own peril" and so this is really nothing more than a step up from an admonition. Calmom's excellent post touches on the how to cite wiki - and I would think that even under Middlebury's new policy an appropriate citation to a wiki source, especially under the rubric of consulted sources, would be not only acceptable but necessary if wiki was used either as a starting point or to bolster minor arguments. How to use wiki and, more importantly, how not to use wiki as a central,core, or pivotal reference is at issue.</p>