Image 01

Archive for the ‘database tips’ Category

Fact-Checking 101: Read Widely

Friday, December 16th, 2016

This entry is part 2 of 2 in the series Fact-Checking

Fact-checking is hard, but it is almost impossible to assess statements or stories you know nothing about. That’s part of what propagandists play on. There isn’t really any substitute for being broadly informed. What follows is some discussion of what I suggest.

Advice

  1. Read widely.
  2. Read more than you watch or listen. TV, video, and radio are usually limited in scope to what looks good/sounds good, and also because most people can read much more than they can see/hear in a given amount of time.
  3. Read outside your Overton Window. Lots of things European and Asian governments do would be unthinkable in the US…and vice versa. Having a frame of reference external to whatever political or social conversation you’re fact-checking helps you decide what’s really outrageous and what isn’t.
  4. Make sure that you’re seeing things you disagree with or are puzzled by daily.

News sources at Tufts

Newspapers work differently in libraryland for a variety of not very interesting reasons, mostly related to money. We don’t usually have access to the newspaper’s website, instead we get them through databases which are mostly designed for search rather than browsing.

Option 1: Web for discovery, library for delivery

As with journal articles, the easiest way to find out if we have something when you hit a paywall is to do a search in Jumbosearch. Be aware that the title of the article sometimes changes a little between the website and the various editions of the paper.

Option 2: Search for discovery

If you’re looking to do comparative analysis of how stories are covered across sources, you’ll want to use one of our newspaper databases.

Lexis-Nexis and Factiva are large collections of mostly English-language newspapers. They’re designed to allow you to readily compare coverage from paper to paper and region to region. Lexis-Nexis also includes transcripts of television and radio network broadcasts.

Tufts also has access to the Boston Globe, The New York Times, Washington Post and LA Times are findable in Lexis-Nexis. For conservative British coverage, you might try the Economist. If you’re looking for less mainstream news coverage, you can try Alt Press Watch or Ethnic News Watch, which focus on smaller publications.

Fact-checking 101: Check the sources

Thursday, December 15th, 2016

This entry is part 1 of 2 in the series Fact-Checking

As a librarian and citizen and concerned human I have spent a lot of the last year consuming news in a variety of formats, online and off. In the aftermath of the election I have some thoughts on tools and approaches I take to making sense of it, and to making sure what I’m reading makes sense. This is part 1 in a multi-part series. First: what to fact-check.

There’s a lot of discussion of “fake news” since the election. It’s not fake news, it’s propaganda. Fact-checking propaganda is usually a waste of time, and making opponents waste their time and energy trying to disprove things often serves the propagandist’s purpose. Another tactic is to flood the airwaves with so much nonsense that it’s impossible to have a conversation about certain topics, like climate change. Zeynep Tufekci, a professor at the University of North Carolina, points out this is a form of censorship. There is also the pragmatic point that people receptive to conspiracy theories are not often receptive to fact-checking.

Don’t try to fact-check something like Pizzagate. Let the New York Times do that–they have a staff and a budget and will work more comprehensively than you likely can.

What I recommend instead is picking the topics you can quickly and easily check on. Most misinformation evaporates if you look at it closely at all. Jumbosearch is the big search box on our homepage. It contains the full contents of the library catalog plus the full text of about 150 million articles. It’s really good for solving a set of common fact-checking problems.

Example 1: new scientific discovery discussed in a news article. Question: is that actually what the study says? First Step: read the article.

  • How: use Jumbosearch to quickly identify if Tufts has an online copy or interlibrary loan it if we don’t. Usually an author’s name and a few distinctive words from the title are all you need. In this case the New York Times has linked to it, but in other cases you’ll need to search.
  • Note: Jumbosearch is also the easiest way to get past paywalls from off-campus. When you hit one, search for the article in Jumbosearch and you’ll be prompted to log in for full text if we have it.

Example 2: an author is discussing their new book with a spectacular claim. Question: is this reasonable? First step: check what reviewers say, especially scholarly ones. Next step: read the book yourself and draw your own conclusions.

BookReview

Example 3: a friend on Twitter or Facebook or in your email inbox brings up something from a website you’ve never heard of. Question: Is this true? First step: see how other media are talking about the issue.

JS News

Additional useful tools

  • Oxford Reference and Credo Reference are both scholarly Wikipedia equivalents; Credo looks a little bit better on a mobile device, but they otherwise solve the same problem. It’s like having a shelf  of reputable scholarly encyclopedias and dictionaries.
  • BLC Worldcat is a library catalog of library catalogs, essentially books (and videos) and other things in print for several thousand research libraries worldwide, including Tufts. If Tufts has it you’ll see how to get it; if we don’t, you’ll be able to have it delivered in a few days.

These are the tools I reach for first if I read something dubious or doubtful or overly excited. Do this much, and you’ll be better prepared than most.

Next in this series: some suggestions for a balanced media diet, and a few news sources available to the Tufts community.

Dyabola Tips

Wednesday, May 30th, 2012

Dyabola is an essential resource for archaeology research, but can be a little baffling. In no particular order, here are a few comments about how it works based on my work with it so far. It reminds me a lot of the CD-ROM interface I used for it in 1993.

  • It includes several collections of data we don’t subscribe to–they are in grey on the list and marked “get license”.
  • You’ll need to check the box labeled “Activate IP access” and hit “start”. Select the language of the interface by clicking on the flag you like.
  • Dyabola keeps a list of successful searches (i.e., ones with results) on the right; you can get back to a previous search by clicking on it, or return to the main menu by selecting Search Result Options.
  • What it’s searching is authors and titles; it’s also possible to browse by subject or by DAI record number.
  • Simple keyword searching appears to work fairly well
  • It is possible to combine lists of search results; it’s necessary to search multi-word phrases individually and combine them into a new result…which you have to name.
  • It is possible to print a list of citations
  • Author searching is gender-neutral. Which is to say, use initials rather than first names.
    • When you do a search it opens up a separate window without the usual browser back button and other controls
    • you can right-click (Windows) or option-click (Mac) to get the Back option
    • you can also click on one of the terms in Session Results on the right to get back to a set of results

Questions? Just ask!

JSTOR Redesign

Tuesday, September 7th, 2010

FindIt at Tufts button crossed out

JSTOR has made some significant changes to how it works, in addition to adding a new coat of paint to its design. Traditionally JSTOR has not covered the current issues of the journals it includes, with anywhere from 3-5 years of current issues usually excluded. It’s also traditionally been all full-text–anything you found in JSTOR you could read immediately. Over the last year JSTOR has been doing deals with scholarly publishers to include abstracts of current issues of some of its journals, about 174 so far, of the 1200 or so we subscribe to from JSTOR. (full list of titles).

This makes JSTOR more like a service along the lines of Academic Onefile, which includes lots of full text and some abstracts. What’s tricky is that JSTOR so far doesn’t support the technology behind our FindIt@Tufts feature, which points you in the direction of subscriptions elsewhere in our collections. We subscribe to most of what they’re indexing, and if you’re on campus and follow a link that looks like this you’ll likely have no trouble reading and downloading. From off-campus you may be asked to pay for access to an article–don’t do this before you check and see whether the library catalog has the journal title and date. You can also request things we don’t have through ILLiad , our interlibrary loan service, and get a PDF copy emailed to you, usually in two days or less.

JSTOR Search Options

JSTOR search options

If you want to search JSTOR the way you have in the past, uncheck the “include links to external content” box on the search screen.

If you want to search just the current material, uncheck the “include only content I can access” box. But note that this is a few years of less than two hundred of the 40,000 or so titles we subscribe to.

Confused? Just ask if you get stuck.

L’année philologique gets a facelift

Tuesday, May 18th, 2010

Home page of l'annee philologique

Home page of l'annee philologique

L’année philologique continues to resolutely remain ten years behind where the rest of the web is, but there are some substantial changes in addition to the new paint job.

The big news: L’année will now include notices of articles and other works prior to their appearance in the print volume at the end of the year in a special tab next to search results called “Interim Records”. A few sample searches show results from as recently as last year (!). I mock, but this is actually a big change from the print edition, which has traditionally run about three years behind the current date. There is a current list of new material added, updated monthly. But there is no convenient way to get notice of this, other than to visit this page. (I’ll see if I can come up with a solution for this over the summer). One minor nuisance: interim records can be printed, emailed, or saved…but not sent directly to Refworks.

Annee Results

Interim records displayed on results pages

A few minor additions:

  • users can set up an account to keep track of searches and save results between visits (that’s 99% of what you get if you set up an account)
  • there is a modest increase in the number of forms you can export citations in (text, PDF, Refworks file, Refworks direct export)
screenshot of preferences panel

screenshot of preferences panel

Changes to the search interface

The underlying structure of the database appears not to have changed, but some of the mechanical difficulty of doing a search has been decreased.

  • Limit by language in Advanced Search (sadly this is limited to English, French, Italian, Spanish, and German–no Bulgarian or Czech). Go to “Advanced Search” and select the “Filters” menu at the bottom
Advanced search screen with location of search filters

Advanced search screen with location of search filters

Questions or concerns? Let me know.

New in JSTOR

Thursday, February 5th, 2009

JSTOR is currently running a trial of a new type of information, 19th-century British pamphlets, at least through June 30th, 2009. The collection consists of several thousand pamphlets scanned by British universities, on a variety of subjects, mostly political and religious, with a good mix of government documents, reports, and speeches mixed in. Looking for contemporary commentary on a speech by Disraeli? Are priests sumptuous livers, wine-bibbers, and thorough idlers? Interested in an 1825 list of grievances from coal miners?