Media Literacy

What is an "Academic" Book?

Generally speaking, academic or "scholarly" books differ from other types of nonfiction books in that they are, among other things:

• Thoroughly researched, often with a comprehensive bibliography
• Written by an expert of some sort (a practitioner, researcher, etc.)
• Published through a recognized academic press specializing in the subject. This is often a university press (i.e., Princeton University Press) but others include:
  • Routledge
  • Sage
  • Taylor and Francis
  • and others

These are the main elements you are going to want to consider when evaluating whether or not a book would be considered academic or scholarly. Some other potential considerations:

• They may have been reviewed by other academics, experts, or librarians
• The press indicates a strict editorial, fact-checking, or peer-review, or evaluation process 

To Learn More:

• University of Texas at San Antonio's "Scholarly vs. Popular vs. Trade" research guide
• University of Nevada, Reno's "Determining Whether a Book is Scholarly" research guide
• The University of Sydney's "Academic Literature" research guide

How Can I Find Reliable Data?

There are a variety of factors that can impact whether or not a data set is both reliable. Some aspects of data reliability include:

• The data collection method(s)
• The sample (including sample size, demographic makeup, etc.)
• Completeness (e.g., are all critical variables available?)
• Replicability (e.g., does the data produce similar results when the similar tests are repeated?)
• The metadata, Readme, or other data manuals/codebooks
  • For example: can you accurately identify what each variable explicitly means given the information provided? Are the questions or rich variable descriptions provided? Or are you guessing based on vague variable names?
• And more...

Moreover, even if a data set is reliable, it may not be applicable to your research. Finding reliable and applicable data is an excellent conversation to to have with your subject librarian, but, to get you started on finding potentially reliable sources, take a look at the following links:

• Princeton University Library Data Catalog
  • An overarching list of data sets Princeton has access to. Note that there may be some access limitations depending on the topic you need, and definitely take a look at the quick links on the lower-left side of the page
• IPUMS
  • Census and survey data from across the world on a variety of topics
• Princeton Maps & Geospatial Information Center
• Our Research Guides! Select your discipline or search and see if one has been created that focuses on data. Two particular highlights:
  • Finding and Using Financial Data at Princeton University
  • Political Opinion & Survey Data Sources

Corrections

Corrections occur post-article publication. While the policies differ from publisher to publisher, and even amongst publications (e.g., The Article Correction Policy for the journal Communications on Applied Electronics typically will only publish significant changes as Errata, Corrigenda and Addenda), typically Corrections from major publishers occur based on the following bases:

• The changes to be made are significant but the scholarly integrity of the piece remains intact
  • E.g., the authors aren't just correcting their spelling, but they could be fixing the data presented in their tables
• The conclusions of the article will remain the same, even with the changes corrected
  • E.g., miscalculations may have been made but the overall significance of the results remains reliable
• The changes are directly related to an error or omission in (but not limited to) funding acknowledgments, the data availability statement, author names, conflicts of interests, etc.

To see examples of these publisher policies, see the editorial policies of PLOS, Taylor & Francis, and Wiley.

Corrections are typically shorter pieces that may include edits that you should read in tandem with the original piece or that have already been applied to the original piece, depending factors that could range from anything to journal policy, the severity of the error, or whether or not the error was made on the side of the publisher or the author(s).

The process for indexing and publicizing corrected articles depends on the publisher. It's often the case that the journal or the larger publisher will include a correction notice or banner close to the top of the original article that links to a new, corrected version or to the published correction that lists how the online version has been amended from what it once was. For example, the online version article listed below was corrected from the original and a correction was released explaining the reason for these changes was due to a publisher error. The amended article contains a banner just underneath the citation information linking the reader to the correction:

• Caron, P., Tudor, C. Grunenwald, S. (2025). Levothyroxine Absorption Test With the Daily Levothyroxine Dose in Patients With “Refractory Hypothyroidism”. Journal of the Endocrine Society, bvaf017. 9(4), https://doi.org/10.1210/jendso/bvaf017

• Correction to: “Levothyroxine Absorption Test With the Daily Levothyroxine Dose in Patients With ‘Refractory Hypothyroidism.’” (2025). Journal of the Endocrine Society, 9(6), bvaf076. https://doi.org/10.1210/jendso/bvaf076

For publications that do not amend the original article This may now be listed in the journal with an amended title consisting of the original title with the addition of the word "Correction" or "Corrected". PLOS takes this approach with their publications, publishing corrections with the phrase "Correction" and the original versions of corrected articles with large, visible banners:

While corrections may not often show up in your database searches, if you click on an article and you see a banner stating that an article you're considering has been corrected, it's definitely worth your time to look into the correction before you proceed. You may find the correction alleviates some concerns you had with the original article.

Retractions

Retractions occur when a published article is removed from a journal due to a significant concern or error. These errors include, but are not limited to: 

• Major errors that make the findings unreliable, such as miscalculations or experimental errors, fabrication of data, or falsification of findings
• Undisclosed conflicts of interest
• Copyright infringement
• Plagiarism
• Unethical research practices

While an investigation is taking place, journal editors may publish an Expression of Concern to indicate that a paper or sections of a paper may not be trustworthy or validated (See "Retractions: Their Effect on the Scholarly Record, and What's Being Done About It"), but this is not required while an investigation is ongoing or before a retraction takes place. An Expression of Concern May also be published as a stand-alone post-publication notice (See the publisher's notes for PLOS, Wiley, and Taylor & Francis)  For more complete information on retractions, check out the following sources:

• COPE Council. COPE Guidelines: Retraction Guidelines. November 2019. 

  https://doi.org/10.24318/cope.2019.1.4 © 2019 Committee on Publication Ethics (CC BY-NC-ND 4.0) https://publicationethics.org

• Walden University Library's "Journal Article Retraction" Research Guide

• And to give some context about the actual state of retractions: Petrou, C. (April 18, 2024) Guest Post--Making Sense of Retractions and Tackling Research Misconduct. The Scholarly Kitchen. https://scholarlykitchen.sspnet.org/2024/04/18/guest-post-making-sense-of-retractions-and-tackling-research-misconduct/

If you'd like to see some papers that have been Retracted, their reasons for retractions, the journals they are in, and the ultimate outcomes of the papers, take a look at the following resources:

• Retraction Watch 
  • An online website that self-describes as "Tracking retractions as a window into the scientific process". It regularly publishes blog posts in an investigative style on all things scholarly publications, retractions, and more--from the positive stories to issues that you may want to be aware of if you want to one day publish or be involved in the scholarly publishing industry
• The Retraction Watch Database
  • Database itself
    • As a tool--keeps track of various papers that have been investigated, the reason for the investigation, outcome of the investigation (correction, reinstatement, retraction, etc.)
    • Make sure to look at the README before looking at the data!
  • User Guide
    • For various source tips & tricks, FAQs, and the data appendices that define how the database is structured

If in your research you find a retracted article, it may be amended similarly to a corrected article. While it likely won't be removed from the site itself, it's record will be amended. This will likely include a change in indexing so that it's title will now include the original title with the addition of some variation of "Retraction" or "Retracted". This will often be included at the beginning of the amended title, but may be added to the end. There may also be other indicators in the database or publication that the article you are considering has been retracted.

In Web of Science, your search result list are amended so that retracted publications will show up with the words "Retracted:" preceding the initial title. But in case you miss that, underneath the title you'll see a warning symbol (an orange outline of a triangle containing an orange exclamation point) followed by the word retracted. If you were to click on this article, you'd receive the same warning notice above the title and in the article's information section, under "Document Type" you'll see that Web of Science has now added the category of "Retracted Publication". and the publisher itself has made the retraction even more explicit

PLOS appears to rely more on the word "Retraction:" in their search results to get their message, although the article pages themselves do include a small warning label above the title similar to Web of Science. If you find a pre-retraction version of the article you'll see a clear warning banner alerting you to the retraction and its reason. If you try to download a PDF of a retracted article from their site--even the original version--you will see a large red watermark with the word "Retracted" spanning across each page. If you'd like to see an example:

• Vasiliauskaite V, Evans TS (2019) Social success of perfumes. PLoS ONE 14(7): e0218664. https://doi.org/10.1371/journal.pone.0218664

In other words, if you pay enough attention to your databases and any accompanying warning signs (and perhaps don't try to just automate scraping results for your research when you shouldn't be...) then the chances are quite high that you'll notice a source has been retracted before you use it. But do make sure that you're paying attention to your search and the database and publisher context clues so you don't miss this important information.