Evidence Summary


Americans’ Tolerance of Racist Materials in Public Libraries Remained Steady between 1976-2006


A Review of:

Burke, S. K. (2010). Social tolerance and racist materials in public libraries. Reference and User Services Quarterly, 49(4), 369-379.


Reviewed by:

Leslie Bussert

Head of Instruction / Literature and Humanities Research and Instruction Librarian

Cascadia Community College Campus Library

University of Washington Bothell

Bothell, Washington, United States of America

Email: lbussert@uwb.edu


Received: 1 Dec. 2011                                                                    Accepted: 28 Jan. 2012



Description: cc-ca_logo_xl 2012 Bussert. This is an Open Access article distributed under the terms of the Creative Commons- Attribution-Noncommercial-Share Alike License 2.5 Canada (http://creativecommons.org/licenses/by-nc-sa/2.5/ca/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one.





Objective – To determine the general public’s levels of social tolerance toward public library materials containing racist content in order to present opinion data to librarians within a framework of scholarly perspectives that they can use for making decisions about intellectual freedom and controversial materials in libraries.


Design – Percentage and regression analysis of the General Social Survey longitudinal trend study dataset.


Setting – United States, 1976-2006.


Subjects – Random samples of 26,798 primarily English-speaking adults aged 18 and up.


Methods – The author analyzed responses from the well-respected and frequently used General Social Survey (GSS), which has been conducted by the National Opinion Research Center since 1972. The GSS is a closed-ended survey including a variety of demographic measures. Between the years 1976 and 2006, it also included a question to gauge the support of removing a book with racist beliefs about African Americans from the public library. The surveys were conducted irregularly over this thirty-year span, and in total the question was asked nineteen times garnering 26,798 responses. Spanish speakers were not included until 2006.


The author examined the data in multilevel cross-tabulations using percentages, and calculated chi-square for independence using frequencies. A multiple regression analysis was conducted to determine the predictive value of the independent variables examined on opinions of book removal. The author examined different variables, including education level, race, age, parental status, sex, geographic factors, religious affiliation, political party, and political conservatism. Occupation was not used in the regression analysis because sample sizes in some categories were too small. The two ordinal variables, age and education level, were available as ratio level data that are most appropriate for regression calculations.


Due to the large sample size, very small differences in percentages are significant at the .000 level. In these cases the author made judgment as to whether these differences were meaningful, or divided the data into multi-layer cross-tabulations to reduce the sample size and make the significance test more informative.


Main Results – Analysis revealed the most influential predictors of support for book removal from the public library were education level, religious affiliation, and race. Age was particularly influential for older respondents, while occupation and living in the South were moderately influential. Variables with only slight correlations to support of book removal included political party affiliation and conservatism, parental status, and sex.


Across all years of the study only 35.3% of respondents supported removal of racist materials from the public library. Levels of support only changed slightly over the decades: in 1976, 38.1% supported removal while in 2006 only 34.5% did.


The mean age of respondents was 44.1 years and the median was 42 years. Respondents over 57 years old were more likely to support removal (43.5%) compared to younger ages whose support ranged from 31.1–34.1%. The largest change over time was seen from respondents 57 years and older, whose support for removal dropped in later years of the study.


Education level had a strong impact on opinions; the lower one’s education level, the higher their support for removal of the racist book from the public library. Of those with less than a high school degree, 50.6% supported removal versus 35.8% of high school diploma holders. Respondents with junior college, bachelor’s, and graduate degrees supported removal at 29.2%, 20.5%, and 15.3%, respectively. Over time, those with high school degrees maintained their level of support for removal while those with higher levels of education increased their support for removal.


Race was strongly related to opinions on removing offensive items from the library. While half of African American respondents supported removing a racist book, only one-third of white respondents did. However, in all but a few subcategories of analysis, the majority of African Americans did not support removal, indicating a great deal of social tolerance on their part despite the possibility of being more sensitive to the implications of having racist materials in the library. When cross-tabulated with education level, the same pattern of support for removal was reproduced. There was little variation over time in white respondent’s opinions while African Americans’ varied slightly.

Geographic factors affected opinions supporting removal of racist materials, though place size only had a small impact on opinions. Respondents in the South were most likely to support removal (42.1%) and those in New England were least likely (25.2%). About one-third of respondents from the Midwest (33%), Mid-Atlantic (36%), and the West (29.8%) supported removal. Opinions over time remained the same in all regions but the South, whose support of removal dropped to 38.8%.


Religion was found to correlate with opinions on removing racist books from the library. Protestants showed the highest level of support for removal (39.5%), followed by Catholics (32.3%), Jews (21.7%), and respondents unaffiliated with religion (20.5%). Race had a strong impact within some religions on supporting removal, particularly among Methodists and those claiming no religion. When opinions by religion were cross-tabulated with education level, at every level Baptists were more likely to support removal than other groups, while Jews and those without religious affiliation were least likely.


Other demographic variables had little effect on opinions concerning removal of racist materials from the library. Parents supported removal (37%) while nonparents were less likely to (30%), and men and women were almost equally likely to support removal (33% and 37% respectively). Political affiliation and level of conservatism only showed slight effects on opinions supporting removal. By a small margin Democrats were most likely to support removal (39.2%) followed by Republicans (34%) and independents (32.5%). Across the conservatism spectrum, moderates were most likely to support removal (37.7%) followed by conservatives (36.4%) and liberals (29.9%).


The author also examined whether a respondent’s occupation influenced their opinions and focused this inquiry on the professions of library workers and educators. Librarians were overwhelmingly against removal of racist materials while library paraprofessionals were less likely to support it than other workers with a similar level of education. College and university teachers in disciplines other than library and information science were divided but in comparison to other similarly educated professions they were less likely to support removal. School teachers were significantly more likely to support removal than other occupations also requiring a bachelor’s degree.


When contrasted with controversial materials of other types, such as those by openly homosexual or communist authors, different patterns of support for removal over time were observed. Support for removal of books by homosexuals and communists declined significantly over the decades. Similar to the support of the removal of racist materials, education and religious affiliation were the variables most highly correlated to support of removal of these other types of controversial books.


Conclusion – The discomfort among Americans over the free expression of exclusionary speech about African Americans remained relatively consistent over the years of the study (1976 – 2006) despite some shifts within particular demographic categories. Tolerance toward free expression by homosexuals and communists increased over time, demonstrating exclusionary speech may be perceived as a different type of social threat. Librarians can use this information to: better understand how non-librarians view intellectual freedom in the context of materials with offensive content; inform collection development decisions and predict likelihood of challenges based on the demographics of their user communities; and to educate the public and library stakeholders of the implications of challenging these kinds of items within a library’s collection through upholding their professional values. Librarians should continue to serve their communities by acting as champions of intellectual freedom and to uphold the profession’s rigorous standards. The author suggests future research could: address attitudes about materials with racist views of populations other than African Americans; look for differences in opinions among library users versus non-users; and differentiate between adult and children’s materials containing controversial topics.





This study offers thorough analysis of a longitudinal dataset spanning thirty years and covering a variety of variables potentially impacting opinions on removing books from public libraries. The literature review includes other disciplines, such as speech communications, public policy, and race studies, for broader context.

Limitations of the study are acknowledged and based in the dataset itself. For example, African Americans are the only race included in the survey prompt analyzed. The author

concedes additional questions would be necessary for a comprehensive analysis from the library and information science perspective, and recognizes one’s opinions do not necessarily correlate to action.


The brief summary of data from related GSS survey prompts about support for removal of homosexual or communist materials is valuable, as it helps identify whether opinions on racist books translate to other types of controversial materials, and in this case they did not. This comparison could be expanded by also analyzing the GSS data about removal of socialist or militarist materials and more recent prompts about books by anti-American Muslim clergymen and general anti-religious materials (ICPSR, 2011a, 2011b).


The author’s findings are also illuminated when placed within the broader study of censorship and social tolerance. Boyer (2002) has traced shifting patterns of book censorship in American legal and cultural history allowing for further contextualization of these findings. Meanwhile, Harell’s (2007, 2010) work examines the effects and consequences diverse democratic societies have on social tolerance and how exposure to social diversity impacts attitudes toward exclusionary speech. Like Burke (2010), Harell also presents evidence that individuals are less likely to tolerate exclusionary speech compared to other controversial speech, and argues it is because exposure to social diversity increases empathy toward the victims of such speech (2007). This may help explain why individuals with more education, along with college professors or librarians, were more tolerant of racist materials in libraries. Perhaps exposure to diverse ideas and people via education also impact one’s tolerance for exclusionary speech. However, librarian opposition to removal of racist materials exceeded all other educators, indicating they are stronger advocates for free speech.


The findings of this study reaffirm the profession’s stance on intellectual freedom and the author offers several useful ideas for applying them to practice. Additional ideas for future research include revisiting the GSS data over time for additional analysis and collecting opinions on this issue from non-English speaking populations across the United States.




Boyer, P. S. (2002). Purity in print: Book censorship in America from the gilded age to the computer Age. Madison, WI: University of Wisconsin Press.


Burke, S. K. (2010). Social tolerance and racist materials in public libraries. Reference & User Services Quarterl,y 49(4), 369-379.


Harell, A. (30 Aug 2007). Tolerance judgments in multicultural democracies: Target groups and the role of social diversity. American Political Science Association Annual meeting, Hyatt Regency Chicago and the Sheraton Chicago Hotel and Towers, Chicago, IL, USA. Retrieved 11 Jan 2011 from http://www.allacademic.com/meta/p209670_index.html


Harell, A. (2010). The limits of social tolerance in diverse societies: Hate speech and political tolerance norms among young youth. Canadian Journal of Political Science, 43(2), 407-432. doi: 10.1017/S000842391000003X


Inter-University Consortium for Political and Social Research (ICPSR). (2011). General Social Survey, 1972-2006 [Data file]. Available from http://www.icpsr.umich.edu/icpsrweb/ICPSR/access/index.jsp


Inter-University Consortium for Political and

Social Research (ICPSR). (2011). General Social Survey, 1972-2008 [Data file]. Available from http://www.icpsr.umich.edu/icpsrweb/ICPSR/access/index.jsp

Evidence Based Library and Information Practice (EBLIP) | EBLIP on Twitter