Comparison of Wikipedia and other encyclopedias for accuracy, breadth, and depth in historical articles

The Authors

Lucy Holman Rector, Library & Instructional Resources, Harford Community College, Bel Air, Maryland, USA

Abstract

Purpose – This paper seeks to provide reference librarians and faculty with evidence regarding the comprehensiveness and accuracy of Wikipedia articles compared with respected reference resources.

Design/methodology/approach – This content analysis evaluated nine Wikipedia articles against comparable articles in Encyclopaedia Britannica, The Dictionary of American History and American National Biography Online in order to compare Wikipedia's comprehensiveness and accuracy. The researcher used a modification of a stratified random sampling and a purposive sampling to identify a variety of historical entries and compared each text in terms of depth, accuracy, and detail.

Findings – The study did reveal inaccuracies in eight of the nine entries and exposed major flaws in at least two of the nine Wikipedia articles. Overall, Wikipedia's accuracy rate was 80 percent compared with 95-96 percent accuracy within the other sources. This study does support the claim that Wikipedia is less reliable than other reference resources. Furthermore, the research found at least five unattributed direct quotations and verbatim text from other sources with no citations.

Research limitations/implications – More research must be undertaken to analyze Wikipedia entries in other disciplines in order to judge the source's accuracy and overall quality. This paper also shows the need for analysis of Wikipedia articles' histories and editing process.

Practical implications – This research provides a methodology for further content analysis of Wikipedia articles.

Originality/value – Although generalizations cannot be made from this paper alone, the paper provides empirical data to support concerns regarding the accuracy and authoritativeness of Wikipedia.

Article Type:

Research paper

Keyword(s):

Reference services; Encyclopaedias.

Journal:

Reference Services Review

Volume:

Number:

Year:

2008

pp:

7-22

Copyright ©

Emerald Group Publishing Limited

ISSN:

0090-7324

Introduction

Wikipedia, the collaborative work of 67,000 contributors, is currently the largest single encyclopedia, with 2,045,007 English articles as of October 12, 2007 ( Wikipedia, 2007). While many (Benkler, 2006; Janes, 2005; Lessig, 2004; O'Leary, 2005) see Wikipedia as a valuable resource, others, particularly scholars and librarians (Achterman, 2005; Binkley, 2006; Ishizuka, 2004; Jacso, 2002; Read 2006a, b), question the accuracy, comprehensiveness and writing quality of Wikipedia entries and the authority of its contributors.

Interestingly, although many question Wikipedia's strength as an encyclopedia, few (Giles, 2005; Rosenzweig, 2006) have actually compared Wikipedia to other sources to provide evidence supporting their concern. Research is needed to determine Wikipedia's place among reference works. This study will conduct a content analysis of Wikipedia and three other reputable history and general reference sources to compare their coverage of historical topics in terms of accuracy, breadth, and depth.

Literature review

Most reference works rely on scholars to write and edit their articles and essays. Wikipedia, however, allows anyone to contribute and edit encyclopedia entries. In fact, Wikipedia welcomes amateur contributors and notes that no formal training is required for posting an entry ( Wikipedia, 2007). Unlike other reference sources, individual Wikipedia entries do not list authors' full or even real names, and authors do not post their credentials in terms of expertise in the field of their contributions. Without full disclosure of authorship, readers cannot verify the expertise of the author or even conduct further research on his/her credentials. Furthermore, while Wikipedia encourages authors to cite references, it realizes that many articles do not include sources ( Wikipedia, 2007). The encyclopedia readily admits that there are certainly opportunities for vandalism, and the lack of an editorial board or governing process may result in under-coverage of certain topics. However, the encyclopedia believes that its openness allows for greater breadth and depth of many subjects, and any inaccuracies, inadvertent or malicious, may be corrected in a matter of hours, and new discoveries quickly added. It is precisely this open peer-production process that champions of Wikipedia cite as its greatest strength.

Is Wikipedia a reputable source? Should educators and librarians be embracing this living, growing text or should they continue to caution students and other users to evaluate Wikipedia content more carefully than that of other reference sources? In order to measure its relative coverage and scope, several studies have compared Wikipedia entries to those in well-respected reference works. Most (Giles, 2005; Lih, 2004; Rosenzweig, 2006) have found Wikipedia to be as accurate as other sources, although they have found minor factual mistakes and criticized the level of writing.

In a recent study conducted by the journal Nature (Giles, 2005), reviewers compared 21 science entries from both Wikipedia and Encyclopaedia Britannica and found that each encyclopedia contained four major errors. Wikipedia had 162 omissions or minor factual errors, while Encyclopaedia Britannica had 123. Rosenzweig (2006) compared 25 Wikipedia biographies to those in two other online resources (American National Biography Online and Encarta) and found Wikipedia to be less comprehensive and detailed than American National Biography but more so than Encarta.

Others, such as Alexander Halavais, have informally tested the encyclopedia by purposefully submitting erroneous material to see how quickly, if at all, the material is corrected (Read, 2006a). Halavais discovered that all of his 13 errors had been corrected within three hours of his postings. Weaver et al. (2006) examined relationships among Wikipedia entries connecting individuals, organizations, and places (for example, an article on US presidents would state that Woodrow Wilson was the 28th president and then would link directly to his biography, which would confirm his position as the 28th president). Their research found that of 200 statements in 73 entries, 97 percent were true. Another study (Korfiatis et al., 2006) analyzed relationships among Wikipedia contributors and how they respected one another's work. These last two studies, however, actually do not verify the accuracy of facts within Wikipedia or evaluate the writing quality.

Certainly more research is needed to measure the accuracy and authoritativeness of Wikipedia entries to satisfy academics' concern about the quality of the work. This content analysis attempts to expand that body of research.

Methodology

This study evaluated nine comparable entries in Wikipedia, Encyclopaedia Britannica (print and online), the Dictionary of American History, and American National Biography Online to compare the accuracy and thoroughness of Wikipedia entries to those of well-respected print reference sources.

To conduct such a comparison this study used two of the most reputable history sources and the most respected general encyclopedia, both print and electronic. The Guide to Reference Books (Balay, 1996), the bibliography that most academic libraries use to build their collections, list both prominently as subject-specific encyclopedias. The Dictionary of American History (3rd edition, 2003) is considered a required reference work in most academic libraries. The latest edition of ten volumes includes 4,434 articles with 1,200 illustrations and 252 maps (Ward, 2003). Subject experts write all entries. Several noted reviewers (Coutts and McConnell, 2004; Ward, 2003; Rettig, 2003; Browne, 2003) find that the new edition improves upon the success of previous editions and continues to be a standard historical reference. Balay (1996) states that the earlier revised edition (1976) thoroughly covers the social, political, economic and cultural history of the USA, omitting only biographies that are the purview of the Dictionary of American Biography.

For American biographies, the long-standing pre-eminent source was the Dictionary of American Biography (DAB), starting in the 1930s and continuing through 1988. The more recent American National Biography (1999; ANB) has replaced the DAB as the library standard. The American Library Association selected it as an Outstanding Reference Source in 2000 (Dalinsky, 2000), and the Association of American Publishers awarded the 24-volume work as its publication of the year ( Library Journal, 2000). One reviewer considers it “an essential purchase for all academic and most large public libraries” (Coutts and Richard, 2000). The online version contains 17,400 biographies, including updated and new entries from the 1999 print edition, and 80,000 cross-references and external links ( American National Biography, 2007).

The Encyclopaedia Britannica has a long established history as the premier general encyclopedia. Balay (1996) considers Britannica “the most famous encyclopedia in English, and for some purposes the best” (p. 107). The 14th edition, the New Britannica, made significant changes to its structure and organization. Balay (1996) continues, “All in all Britannica remains the most comprehensive, authoritative international encyclopedia in the English language” (p. 108). The 15th edition, first published in 2002, contains 64,862 entries in 32 volumes; 4,300 authors contributed articles. Its electronic counterpart, Britannica Online, has 118,270 entries with almost 18,000 illustrations and maps and includes video and audio clips as well as 135,000 links to other online references (Awe and Bibel, 2002). The new 2005 edition includes more than 65,000 articles and 24,000 images and maps ( Encyclopaedia Britannica, 2006a). As of December Britannica Online has expanded to 122,264 articles ( Encyclopaedia Britannica, 2006b).

Lastly, as mentioned earlier, Wikipedia is the largest online reference, with 75,000 contributors and 5,300,000 entries in 100 languages ( Wikipedia, 2007). Wikipedia stands on three basic principles:

articles should present an unbiased or neutral description of the entry;
articles should be derived from reliable published sources and be verifiable; and
articles should not represent an author's original research ( Wikipedia, 2006).

Wikipedia encourages contributors to cite sources used ( Wikipedia, 2006), but numerous articles have few if any citations. Authors are anonymous and may not be experts in their subject area.

There are a number of challenges in conducting a content analysis of internet resources. First of all, the size of Wikipedia and other online sources makes it difficult to find truly representative samples. In addition, the fact that the online sources contain graphic, video and audio files as well as text complicates a direct comparison to print resources (Weare and Lin, 2000). Thirdly, web content almost invariably contains a number of links to additional material. While these links may be considered similar to “see also” references in a print source, the immediacy of the linked content encourages the reader to break from a linear reading of the content at hand with options for additional reading (Weare and Lin, 2000). Lastly, the dynamic nature of internet resources adds the challenge of capturing content that may be edited or completely rewritten in minutes or certainly days (Weare and Lin, 2000). This makes any particular snapshot of the content somewhat arbitrary.

Because this content analysis is partially qualitative in nature, White and Marsh (2006) recommend that the sampling be purposeful rather than completely random. In order to select a representative variety of historical articles, this study used a modification of stratified random and purposive sampling. The researcher first identified four types of historical entries:

place;
event;
movement or phenomena; and
biography.

The sampling began with the specialized history reference, the Dictionary of American History (DAH). Within each volume the researcher randomly selected an entry, alternating among the four types of entries so that there were two biographies, two events, two places, and three movements/phenomena in total.

The nine selections were:

Badlands and Sand Creek (places);
The Chautauqua movement, the Free Soil party, and the Niagara movement (phenomena);
The Harper's Ferry raid and the Mexican-American War (events); and
William Kidd and Harriet Tubman (biographies).

Once the researcher identified examples of each type article in the historical encyclopedias, she found the comparable article in Encyclopaedia Britannica and Wikipedia. For the two biographies, the researcher then selected the equivalent biographical essay in the American National Biography Online.

As mentioned, one difficulty of analyzing web content is its constantly changing nature. Of course, Wikipedia believes that its strength is its dynamicism. To address the rapid rate of content change, this analysis also briefly reviewed each Wikipedia article's history to note the quantity, type and scope of editorial changes over a specified period of time.

The study compared each text in terms of size, accuracy, detail, references (verifiability) and opportunities for further reading. Some of these concepts were quantified by such operational measures as word and fact counts and number of references. Other more qualitative concepts were measured by the accuracy of factual materials, the level of detail and the ability to verify entry content.

The study considered each entry (without links to external internet content or “see also” references) a unit of measure. The analysis included a word count and fact count for each entry and tallied the number of facts included in two or more sources as fact matches. The analysis also rated the reliability of those facts by checking them against additional historical and biographical sources (see list of references for fact checking). The analysis included a count of “see also” references in the print materials and contextual links in the online sources. The researcher checked authorship and bibliography citations. Each entry was compared to the other two in terms of matched facts, factual discrepancies, additional facts that were not included in the other sources and facts omitted that did appear in the other sources. The researcher calculated an overall accuracy rate by dividing the number of incorrect or partially incorrect facts by the total facts as well as by the total verifiable facts (subtracting the unverifiable facts from the total fact count).

Results

Badlands

The entries on badlands were three of the shortest essays examined. Table I shows the comparison among the four sources (including the print and online versions of Encyclopaedia Britannica) in their description of badlands. Britannica actually includes two entries on the badlands; one in terms of the geological phenomenon in general and one specifically on the Badlands National Park in South Dakota; the other two sources combine the two into single articles. Here Britannica Online (BO) has by far the longest of the entries (625 words); the print version and Wikipedia are next with slightly more than 250 words (253 and 258, respectively). All the facts are verifiable and correct; the difference here lay in the facts included. Wikipedia lists six additional facts concerning other badlands areas and national parks in North Dakota and Canada. The Dictionary of American History (DAH) provides history of the settlement of the area, and the history of paleontological discovery. Britannica, particularly BO, includes many new facts concerning the size, location and geographical description of Badlands National Park and its flora and fauna.

Because each entry focuses on different aspects of the phenomenon and region known as the badlands it is difficult to compare the four in terms of comprehensiveness. However, certainly Britannica Online has the most extensive entry. In terms of level of detail, both Britannicas and the DAH contain more specific data. Furthermore, only the DAH includes a bibliography (two entries) and two “see also” references.

Chautauqua

The four essays on the Chautauqua movement vary considerably, and several facts here are unverifiable or erroneous. Table II details the evaluation of the entries. The two Britannicas are virtually identical. Ten facts are shared by all four sources. The Dictionary of American History (DAH) shares six each with Britannica Online (BO) and Wikipedia, while BO and Wikipedia share one. Britannica includes six otherwise unmentioned facts, DAH contains 69 unmatched facts and Wikipedia contains 94 additional facts.

In this case, Wikipedia and DAH are relatively equal in their comprehensiveness, discussing the original Chautauqua as well as the local and circuit or tent Chautauqua movements. DAH also covers more on the history of the first Chautauqua and the growth of its offshoot, the Chautauqua Literary and Scientific Circles (CLSC), while Wikipedia has more breadth in the areas of music and other entertainment features and more recent history. Wikipedia has more unverifiable facts (eight) and inaccuracies (nine) than does DAH, but given that Wikipedia has 94 additional facts for a total of 110, it still holds a higher ratio of accuracy. Most of the factual errors are in the size of the movement (Scott, 1999; Bohn, 1826). Both entries are relatively succinct and tightly written. While the Wikipedia article does list four general references, its last paragraph on the comparison of Robert Pirsig's work, Zen and the Art of Motorcycle Maintenance, with Chautauquan themes contains quotes without attribution. Furthermore, the essay contains a quote attributed to Theodore Roosevelt with no reference/citation and another quote seemingly from Sinclair Lewis that was unverifiable and unattributed.

Here, the Britannica (both print and online) are the shortest and least comprehensive entries with 203/202 words and 23 facts. All of Britannica's facts are verifiable; the only error is in the number of permanent, local assemblies at the height of the movement.

Free Soil Party

In the case of the entries on the Free Soil Party, the small political party of the 1840s and 1850s, none of the articles are particularly comprehensive and detailed, although the Dictionary of American History is the longest essay with 576 words and 49 facts (see Table III). Again, the two versions of Britannica are identical, and all four essays share nine facts. DAH shares nine additional facts with Britannica and six with Wikipedia, while Wikipedia and Britannica share two. DAH has 25 additional facts and discusses the origins of the party and its influence in politics in the years leading to the Civil War.

Wikipedia and Britannica are relatively equal in size (372 and 316 words, respectively) and number of facts (23 and 29). Only Wikipedia has a single fact error – the number of party representatives in Congress after the 1848 election. Additionally, Wikipedia contains a quote apparently from the party's platform with no attribution.

Harper's Ferry Raid

The DAH entry on the Harper's Ferry Raid differ fundamentally from the other three articles in that it focuses only on John Brown's 1859 attack on the arsenal in Harper's Ferry, while Wikipedia and the two Britannica articles include both earlier and more recent history on the town and general geographic features (see Table IV). The research here only reviewed the sections of the latter three articles relating to Brown's Harper's Ferry raid. With that said, the DAH article is much lengthier (1,631 words) with many more facts (122). The article provides details of John Brown's pursuits prior to the raid, explaining how his abolitionist efforts culminated in the attack. It describes the individual events of the raids and the actions of both Brown's men and the militia and US Calvary. The Wikipedia and Britannica entries are much shorter, sharing a majority of their facts. Despite the disparity in length and depth of the two articles, the Wikipedia entry has only one fewer incorrect fact that DAH and has one more unverifiable fact. Still, the Wikipedia article has a rather long direct quote in the last paragraph of the John Brown section without attribution, and the one citation included in that section does not accurately reflect the original source (the Wikipedia entry lists a different date than the original).

The total article size is 1,659 words for Wikipedia and 432/549 for Encyclopaedia Britannica/Britannica Online.

William Kidd

Of the four entries for William Kidd (pirate), Wikipedia is by far the longest article with 1,978 words specific to Kidd's biography. As one can see in Table V, although Wikipedia provides many more particular details about Kidd's pursuits, ships, crew members and bounty, both it and DAH include all major incidents leading to his trial and death. Many of these details in Wikipedia are either incorrect (14) or unverifiable (37) as compared to the four unverified facts in DAH and the three partially incorrect facts in the two Britannicas. For example, the Wikipedia article includes several details of Kidd's childhood that were not confirmed in three principal biographies (Harris, 2002; Ritchie, 1986; Zacks, 2002), and the essay suggests he turned to piracy before most of the biographers indicate. One paragraph of the essay references a children's book on Kidd, which may oversimplify the man's career. In addition, the Wikipedia entry includes 38 facts concerning references to Kidd in literature, music and popular culture that are not directly related to his life, but may be of interest to readers. Again, the two Britannica articles are considerably less detailed and comprehensive, with only 19 facts not mentioned in either of the other two entries.

The total word count of the Wikipedia entry is 2,415, with 38 additional facts.

Mexican-American War

Table VI illustrates the comparison of the entries on the Mexican-American War. Wikipedia has the longest article by far, more than five times as long as Britannica and more than three times as long as DAH. Each entry shares between 25 and 30 facts with one or the other (or both) source(s); Wikipedia has the most original facts and the most that are incorrect or unverifiable. After consulting 34 historical and biographical sources, this study found that Britannica has one unverifiable fact and one fact of questionable accuracy (there is slight disagreement among a number of sources – Meed, 2003; Selby, n.d.; Bauer, 1974 – as to the proposed initial US offer to buy California and New Mexico). There are two minor inaccuracies in the DAH entry, related to battle dates, and one unverifiable fact. In Wikipedia ten facts are unverifiable, 12 facts are only partially correct or verifiable, and 28 facts are in direct dispute with other sources. This total does not include facts, such as number of casualties, which varied within a number of sources. Most of the inaccuracies are minor ones, such as incorrect dates or lengths of battles or size of military forces and number of casualties. In other cases, the inaccuracies are more problematic; in one case involving a skirmish at Dominquez Rancho, Wikipedia states that the Californios (or Mexicans) retreated, when actually the Americans retreated (Denger, n.d.). Furthermore, the sentence in Wikipedia is exactly the same as another source about a different California skirmish at Rancho Natividad (California State Military Museum, 2002), where the Californios did retreat.

But perhaps even more problematic, this entry in Wikipedia is written so that small skirmishes have more prominence than major battles (and some major battles were missing entirely), and readers may have difficulty determining the chronology of the war without reference to specific dates because the entry did not sequence events in chronological order. Unlike the other two samples evaluated, this entry obviously is written by multiple authors with little apparent concern for the overall flow and readability of the article. Unfortunately the citations, too, are very problematic. There are two footnotes each marked [1], [2] and [3], with only one each listed in the footnotes. Secondly, the footnote number [2] textually referenced the author of the second footnote [1]. The article includes six additional citations parenthetically in the entry; one of those cannot be verified.

The DAH entry seems to be the most comprehensive and accurate, even though the Wikipedia is longer and much more detailed. Although it does not provide the level of detail that Wikipedia did (77 facts as compared to 234), it offers an overview of the war that is proportional to most other historical accounts. It contains information on each of the major military campaigns in the war as well as the diplomatic efforts made during the war. Furthermore, it offers a brief description of the events leading up to the war and the economic and political aftermath of the conflict.

Niagara Movement

The entries describing W.E.B. DuBois' Niagara movement are among the shortest; all of the articles are less than 500 words (see Table VII for details). Here the DAH is the longest at 470 and the most comprehensive with 55 facts, 39 of which are not mentioned elsewhere. Its depiction of the movement includes the first three conventions and the absorption of the movement into the larger NAACP. The new Britannica articles are identical and relatively brief, as is the Wikipedia entry. Although all of the entries are brief, there are still disputable facts; Wikipedia has the most with four unverifiable facts and three containing errors; DAH has three unverifiable statements and one statement containing errors, and Britannica has two factual errors.

Sand Creek

As one can see in Table VIII, the three entries (here again, the two Britannicas were identical) on Sand Creek and the Sand Creek Massacre in 1864 vary in length and number of facts. Wikipedia's entry is by far the longest, with 1,010 words on the origins, events and consequences of the massacre (not including references to the attack in music, film and literature). However, the DAH article contains a relatively large number of facts (two-thirds of Wikipedia's total) given the shorter length (43 percent of Wikipedia's). Furthermore, Wikipedia again has a much higher proportion of unverifiable (three) and factual errors (six) than does either DAH (two incorrect) or Britannica (one incorrect). Some of these include the actual location of the reservation where the Cheyenne agreed to settle and particular battalions of soldiers who participated (Greene and Scott, 2004; Greene, 2004; Hatch, 2004; Brown, 1971). Moreover, the Wikipedia entry includes a long quote apparently from the Joint Committee on the Conduct of the [Civil] War that is neither attributed nor fully verifiable.

Harriet Tubman

The four essays on Harriet Tubman vary considerably, and several facts here are unverifiable or incorrect. Table IX details the evaluation of the biographies. In total, 14 facts are shared by all four sources; Britannica Online (BO) and American National Biography (ANB) share nine, BO and Wikipedia share one, and ANB and Wikipedia share 12 facts. Britannica includes two otherwise unmentioned facts, ANB contains 35 unmatched facts and Wikipedia contains 30 additional facts.

The most complete and accurate entry is in ANB by Cornell history professor Margaret Washington (Cornell University, 2006); it is the longest with 1,375 words and contains the most facts about Tubman. Of its 70 facts, all except three of are verifiable; two events are confirmed but vary in terms of the age at which they occurred (Clinton, 2004; Quinn, 2003; Bradford, 1886).

Wikipedia's entry follows ANB in terms of length and comprehensiveness; it includes 67 facts. All except two are verifiable. Of the remaining 65, ten contain errors. Some of these are minor errors, such as year of birth and number of siblings; others are more significant, such as the number of rescue attempts she made and the number of slaves she helped escaped (Clinton, 2004).

Britannica (both print and online) are the shortest and least comprehensive entries with 300 words/19 facts and 385 words/26 facts, respectively. All of Britannica's facts are verifiable, with only one factual error.

Discussion

Certainly this research is too limited in scope to make broad generalizations about the credibility and authority of Wikipedia entries. However, the study did reveal inaccuracies in eight of the nine entries and exposed major flaws in at least two of the nine Wikipedia articles. If one combined the number of facts listed for the nine entries within each source (seven for DAH and two for ANB) and compared them to the number of inaccuracies and unverifiable facts, Britannica has a 96 percent accuracy rate (96.5 percent excluding unverifiable facts), DAH and ANB combined have a 95 percent rate (97.5 percent excluding unverifiable facts), while Wikipedia has an 80 percent accuracy rate (88 percent excluding unverifiable facts). Similarly, the first two sources have a 99.6 percent and 97.3 percent verifiability rate, respectively, while Wikipedia has a 90.7 percent verifiability rate. Based on these nine articles it would seem that Wikipedia is less reliable than the other three sources.

Only the Dictionary of American History and American National Biography credit their entries. Wikipedia stands on the anonymity of authors, and Britannica failed to list authors as well. All except one of the essay authors are researchers in the particular area; many are renowned scholars with multiple publications on the topic. The author of the badlands essay is Doane Robinson, former state historian of South Dakota (WGBH, 2002). Andrew Rieser, the author of the Chautauqua piece, was an associate editor of the DAH (Gale, 2007) and a published author on the movement. Robert Ritchie, the author of the Kidd essay, is a pre-eminent Kidd biographer and Director of Research and Education at the Huntington Library ( American National Biography, 2007).

Eight Wikipedia articles contain unattributed quotes and at least five cases of possibly plagiarized content (material found verbatim elsewhere). The researcher found these cases easily without intense scrutiny; more text may have been copied. This lack of attribution is of serious concern; without an evaluation process or peer review, plagiarized content on Wikipedia may go unchecked.

This analysis shows reason to continue to measure and compare Wikipedia entries to those of authoritative references sources in history and a variety of other academic disciplines. This comparison is particularly necessary for larger, more comprehensive entries with longer and more detailed histories.

Another issue that this study revealed is the difference in writing quality and point of view in the various sources. The shorter entries in Wikipedia are well written and seem to flow seamlessly as if written by one or few writers, but longer entries, particularly the Mexican-American War article, are disjointed and show variation in voice from the numerous contributors. While it is important to include different perspectives on historical places, people and events, these entries should exhibit consistent voice, literary flow and cohesiveness. The writing quality of Wikipedia entries appears to vary greatly on this point. Furthermore, some of the sources Wikipedia uses (when mentioned) may be outdated or lack credibility themselves. For example, the Kidd essay references a juvenile biography, which may provide an overly simplistic treatment of the man.

One area of further study may be to examine and evaluate Wikipedia's histories themselves. In the brief review of the more detailed history of the Mexican-American War, for instance, this study found that references were deleted and correct facts were modified incorrectly. Also, it would appear that some inaccuracies may have been typing errors. A more detailed study of individual histories may lead to deeper analysis as to the nature of Wikipedia's contributors.

While Wikipedia provides a wealth of information and is a model for non-proprietary peer-production of reference materials, it does not fare as favorably as do other reference resources under scrutiny for accuracy, comprehensiveness and reliability. Academics may question students' or colleagues' use of Wikipedia as a scholarly resource. Certainly, the lack of attribution should be of great concern to scholars and Wikipedians alike. Research is needed in other disciplines comparing Wikipedia against trusted reference sources to further evaluate its reliability and value.

Table IComparison of badlands entries

Table IIComparison of Chautauqua entries

Table IIIComparison of Free Soil entries

Table IVComparison of Harper's Ferry entries

Table VComparison of Kidd entries

Table VIComparison of Mexican-American War entries

Table VIIComparison of Niagara Movement entries

Table VIIIComparison of Sand Creek Massacre entries

Table IXComparison of Tubman entries

References

Achterman, D. (2005), "Surviving Wikipedia: improving student search habits through information literacy and teacher collaboration", Knowledge Quest, Vol. 35 No.5, pp.38-40.