Reference Service, Government Information, and COVID-19

Abstract The authors designed and distributed two surveys to answer the questions of whether the 2020 federal response to the COVID-19 pandemic affected academic librarians’ trust in government information, and whether it affected the way they use government information professionally. The results showed a limited decrease in trust in 2020, somewhat mitigated by the administration turnover in 2021. Many librarians already discussed considerations such as bias, authority, and disinformation when recommending government information to students, and others indicated they are more likely to do so going forward. They will also use or suggest more caution with government documents than before. Reference interactions do not always allow the necessary time and space for a nuanced conversation about the limits of government information, so libraries and academic institutions should look for other ways to improve student political literacy as well.


Introduction
When COVID-19 reached the national consciousness in the United States in early 2020, it was accompanied by an infodemic, defined as "an over-abundance of information-some accurate and some not-that makes it hard for people to find trustworthy sources and reliable guidance when they need it" (Managing the COVID-19 Infodemic, 2020). The executive branch of the United States in particular drove the spread of disinformation and confusion as President Trump took to Twitter and other government outlets to provide assurances, dismissals, or unfounded advice regarding the pandemic (Dwoskin & Timberg, 2021;Egan, 2020). More disastrously, the Center for Disease Control (CDC) published science-based guidance and then rescinded it due to pressure from the White House (Bandler, Callahan, Rotella, & Berg, 2020;CDC, 2020). Other agencies and representatives of the executive branch continued to downplay and spread disinformation about COVID-19 and the safety measures enacted by state governments (Abutaleb, Parker, Dawsey, & Rucker, 2020). Trump attacked mask wearing, testing measures, the vaccine process, and the severity of the problem throughout 2020, which led to a measurable increase in the infection rate as people followed Trump instead of evidence-based recommendations (Evanega, Lynas, Adams, & Smolenyak, 2020, p. 7;Sheth, 2020;The Editors, 2020;Victor, Serviss, & Paybarah, 2020). The combination of all this further increased general distrust in government information, predicted to last well beyond the Trump administration (Freelon & Wells, 2020).
Reliable, credible information is vital during a global health crisis, and federal and state governments have direct involvement in pandemic response by funding research, passing mandates, and distributing information. An individual's degree of trust or distrust in these institutions affects their response to those mandates and announcements. The government is not a monolith, but a wildly complex set of multi-level structures staffed by a mix of long-time public servants, elected officials, and political appointees. Accordingly, people hold complex attitudes toward various levels of government, agencies, office holders, and political parties.
Academic librarians assist students and faculty with government information research in reference interactions and research consultations. The pandemic has made credible, reliable public health information vital for everyone, further complicating a post-fact, highly partisan environment. With this background, the authors hoped to answer the following two questions: Has the COVID-19 pandemic affected academic librarians' trust in government information? Has the COVID-19 pandemic affected the way they use government information professionally?

Literature review
When performing academic research, students commonly cite government information from both federal and state agencies (Dubicki & Bucks, 2018, pp. 38-39). This requires some degree of trust in the reliability of government documents, particularly if researchers use those sources to find accurate data rather than to trace policy decisions or note historical significance. Government information is not just limited to the more obvious Congressional bills and agency press conferences. Rather, it includes anything produced by any branch of the government, from official reports to social media content (Albert, Emery, & Hyde, 2020, p. 38). This can mean students sometimes use government sources without fully realizing it. Additionally, the availability of government information online can blur the distinction between publisher and format (Brunvand & Pashkova-Balkenhol, 2008, p. 202). Developing a better understanding of what government information is and how it is created and disseminated is important for students, and significant fluctuations in trust or distrust levels as discussed below could dramatically affect when and how student researchers use these resources.
Among librarians and information professionals, trust in government information was variable long before COVID-19. Following the 2016 election season, academic librarians published a flood of articles, blogs, and guides addressing fake news and the post-truth landscape. Flynn and Harnett detailed how "the 2016 election and the first year of [the Trump] administration have brought to light for the general public a host of issues related to the production, distribution, consumption, and preservation of government information" (2018, p. 209). Prior to the pandemic, librarians expressed concern regarding the "take-down or disappearance" of information from official federal government websites, and many felt they should inform students of this pattern when teaching about or using government information (Kubas, 2019, p. 136). While librarians, like everyone, tend to favor some information sources over others, they showed a decrease in trust in government information after the 2016 election and Trump's inauguration in 2017 (Bluemle, 2018, p. 268). The shift was significant, though not universal, and Kubas noted growing distrust from both academic librarians and their patrons, often directed by their political leanings (2020, p. 137).
Scrutiny of the Trump administration's information practices created more awareness of government entities. More people used and monitored federal government information than before and increasingly understood that sites like the post office and weather service were government publications as well (Flynn & Hartnett, 2018, p. 212). The combination of growing media partisanship, disinformation from a variety of sources, and Trump's pattern of communicating and governing via erratic, unclear tweets led to "heightened skepticism and wariness" among students and academic library patrons after 2016 (Flynn & Hartnett, 2018, p. 210). When it comes to outright false information that affects people's understanding of democracy and reality, libraries cannot adopt a stance of alleged neutrality that acts as if all sources are equally valid (Buschman, 2019, p. 220). Academic librarians must use and teach government information responsibly, which occasionally requires a more nuanced discussion than sending a student a link to a .gov website and ending the interaction. Political literacy is a necessary subcategory of information literacy, and librarians must wade into so-called political topics to assist in fact checking, gauging authority, and refuting incorrect information (Buschman, 2019, p. 221). However, some advocate expanding on traditional information literacy approaches while others argue that if those things could effectively combat disinformation, then librarians would not find ourselves in this position (Sullivan, 2019(Sullivan, , p. 1148. Trust and distrust are distinct concepts and may occur simultaneously (McKnight & Chervany, 2001, p. 29). Van De Walle and Six found that people do hold both views at once, and that both result in different consequences and approaches (2014, p. 169). A person could trust their local government while distrusting the federal government, or trust one political party and distrust another, for example. People who feel distrust toward the government come from a place of inherent suspicion and do not expect the government to work, while people who feel low trust toward the government may find it unreliable but fundamentally functional (Van De Walle & Six, 2014, p. 162). Even though Americans tend to complain about the federal government and often self-report distrust or low trust levels, most still behave in ways that indicate overall faith in the system (Intawan & Nicholson, 2018, p. 601). Houston and Harding found that US citizens generally viewed government workers and elected leaders as capable, but assumed they still acted in their own self-interest rather than for the public good or their constituents (2013, p. 69).
According to Bruno, both trust and distrust in the government and its political systems are valuable and necessary (2017, p. 296). A population that inherently trusts everything a government agency or official says shows too much deference for power and allows for corruption, neglect of governing duties, or betrayal of the public good for personal gain (2017, p. 303). A low level of distrust increases oversight and keeps government officials accountable. Despite that, a population overly skeptical of its government makes the structures ineffective at best. Thus, a balance of trust in the overall system alongside distrust of individual politicians is beneficial to running a country. Both lack of oversight based on too much trust or lack of support based on too much distrust would functionally end a democracy (2017, p. 304).
Current research shows some overall shifts in library reference during the pandemic which should be considered as well. Many academic libraries physically closed or reduced services during the pandemic, leading to a necessary reliance on virtual reference (Garvey, 2021, p. 132). Librarians tend to shorten or skip the traditional reference interview when answering virtual reference questions as both parties favored a quick answer over an in-depth one (Logan & Lewis, 2011, p. 224). COVID-era research does not suggest this limitation has changed much (Kitzie, Connaway, & Radford, 2021, p. 212). Additionally, the overall tone of chat reference questions trended negatively during 2020, especially as the pandemic stretched on longer than anticipated (Kathuria, 2021, p. 115). College enrollment also dropped overall, with an average of eight percent fewer undergraduates since Fall 2019 (COVID-19, 2021).

Methodology
The authors used a survey to collect quantitative and qualitative data regarding academic librarians' use of and trust in government information since the beginning of the COVID-19 pandemic in the United States. The authors created the survey in Qualtrics and distributed the anonymous survey link to members of the American Libraries Association (ALA) via ALA Connect. Specifically, the authors posted in the groups ALA Members; Association of College and Research Libraries (ACRL) Members; the Politics, Policy and International Relations Section (PPIRS) of ACRL; the Anthropology and Sociology Section (ANSS) of ACRL; the Government and Documents Round Table (GODORT) of ALA; and the ACRL Instruction Section. The authors contacted colleagues to request survey link distribution to the listserv or mailing list of the Black Caucus American Library Association (BCALA) and the Tribal College Librarians Institute (TCLI) as well. The authors also shared the survey link publicly via Twitter and in an Academic Librarians group on Facebook. Most of those listservs distribute primarily or exclusively to academic librarians, and both the request for participation and informed consent specified that group as the target demographic for the study. The survey was open from July 22, 2021, through September 12, 2021.
Trust and distrust are difficult concepts to define. Trust can be conceptualized as something someone feels, something someone does, or a way someone is (McKnight & Chervany, 2001, p. 29) The definitions of trust and distrust in government information used in the surveys are scaffolded by definitions or discussions in different disciplines. To trust is "to rely on the truthfulness or accuracy of " or "to place confidence in, rely on" ("Trust," 2021). Trust may also mean confidence that someone will keep a commitment (Kelsall, 2021, p. 290). People may also hold "trusting beliefs," which refers to the extent that one believes another person has the power and ability to do what needs to be done, is motivated to act in another's interests, and will behave consistently (McKnight & Chervany, 2001, p. 36). Philosophy of science distinguishes epistemic trust, or trust in someone's "capacity as a provider of information" (Wilholt, 2013, p. 233). Trust in government may be based on competency and care, or that the government will act rationally and in the best interests of the people it serves (Houston & Harding, 2013, p. 55).
Going beyond the dictionary definition of distrust as a mere lack of trust ("Distrust," 2021), the survey's definition is based on the idea that, when people distrust public officials, they have "an attitude of suspicion rooted in the judgment that a given office-holder or institution is unlikely to serve the public interest, to act as duty requires, or to withstand the temptations of misrule" (Bruno, 2017, p. 296). The authors adapted these definitions, which describe trust or distrust of people or institutions, to learn about academic librarians' trust in government information. It is possible that one's feelings toward the content of a source is entangled with one's feelings toward the person or office that created the information, so the definitions provided to respondents acknowledge both. The first survey presented these definitions to respondents ahead of questions about trust or distrust in government information: "Trust in government information is defined as having a reasonable degree of confidence in its factual accuracy, and in the integrity of the information creation and publication process. Distrust is defined as viewing government information from a place of inherent suspicion and skepticism regarding its accuracy and integrity." Government information is a broad category that may include any number of print and online sources. The literature suggests that, in practice, researchers may cite government agency web pages more than published reports (Hackstadt, 2021, p. 8). Furthermore, Trump's use of his personal Twitter account to share or amplify official US government information has called into question the use of social media by public officials and whether a Twitter account, for example, should be considered a public forum (Durkee, 2021). The pandemic also affects the very landscape of information needs and library users may seek assistance with government information resources for personal use. The survey provided respondents with examples of government information, but the researchers in this study chose to rely on respondents' expertise, so did not define it.
In the initial survey, 52 respondents answered that they consented to be contacted for further information about their responses; 51 provided contact information. The authors sent a follow-up survey to these 51 individuals; 33 responded. The second survey was open from October 7, 2021, to October 21, 2021, and asked more detailed questions about the use of government information in one-on-one interactions with patrons, such as reference interviews or research consultations. The authors used tools embedded in Qualtrics to analyze quantitative responses and NVivo software to code and analyze qualitative responses. Each author independently coded the qualitative answers, then compared results. The authors discussed differences in coding results until consensus was reached. This study is limited in scope to the federal government of the United States with a focus on the executive branch. It is further limited by the collected data as the authors focused on the experiences of academic librarians within the United States. Survey distribution to ALA Connect and professional listservs limited data collection to those who had access to these online spaces.

Results
Respondents could skip questions or end the surveys at any time, so the number of responses to each question varies. The first survey asked respondents to identify primary job responsibilities; more than one selection was possible. Reference was the most common primary responsibility, followed by Instruction, Subject Liaison, Collection Development/Management, Government Documents, and Information Literacy (Table 1). Approximately 98% (n = 147) of respondents report that they interact with government information to some extent as part of their job (Table 2).

Trust and distrust
Reflecting on their pre-pandemic trust levels, most librarians reported either a moderate or high degree of trust in government information prior to February 2020 (47.86% and 38.57%, respectively; n = 140) (Table 3).    Just over half of the respondents (56.43%, n = 140) reported that they have the same level of trust in government information now, approximately 18 months into the pandemic, as prior to the pandemic (Table 4). Most librarians (76.86%, n = 121) reported their likelihood of using government information in their own research or study has not changed since the beginning of the pandemic (Table 5).
Where respondents indicated a change in trust or distrust levels since February 2020 (Table 4), the survey asked them to share what event caused the change in trust or distrust. Open ended responses indicated that the 2020 election affected academic librarians' trust in federal government or government information. Most of these statements were not clear about how their trust or distrust changed. For example, respondents provided answers like "Trump lost," "2020 election," or "Biden elected" as a reason for a change but did not clarify if the event caused more or less trust or distrust. Only two responses were specific about how the change in administration affected their trust: My trust seriously dipped during Trump's time as President. It is slowly recovering since President Biden has taken office Some of my trust has increased due to the new administration, but particularly I don't trust federal public health information very much anymore Accordingly, the follow-up survey attempted to gauge how respondents' trust changed, if at all, in relation to the 2020 US election and other high-profile events during the pandemic. The literature demonstrates that the 2016 US election had an impact on the field, so the survey asked librarians about how both the 2016 and 2020 US presidential elections affected their level of trust in government information. 61.29% of respondents either agree or strongly agree (32.26% and 29.03%, respectively; n = 31) with the statement, "I feel less trust in government information now due to the election of Donald Trump in 2016. " 19.35% disagree with this statement and 3.23% strongly disagree; 16.13% neither agree nor disagree ( Figure 1). Short answers indicated that people were unsurprised with Trump's continued pattern of disinformation. One respondent pointed out, "The issues started under the [Trump] administration as much of the material that my folks needed was getting scrubbed so we had to discuss ways in which we could locate material beyond the scrubbed climate data, et cetera." The second survey also asked librarians to rate their level of agreement with the statement, "I feel less trust in government information now due the election of Joe Biden in 2020." 46.67% (n = 30) disagree with this statement and 30.00% strongly disagree. 20% neither agree nor disagree with the statement; 3.33% agree with the statement; no one responded that they strongly agree (Figure 1).
Librarians also shared how the change in administration affected their trust in government agencies. Over half strongly agree or agree (23% and 35%, respectively; n = 31) with the statement "My trust in federal agencies regarding COVID-19 has increased since President Biden took office in 2021. " 10% disagree and 6% strongly disagree. 25% neither agree nor disagree (Figure 2).   The second survey also asked about two specific events that drew criticism of the federal government's COVID response to see if they affected librarians' trust in federal agencies. First, the CDC changed COVID data reporting at the direction of the president, leading to widespread confusion and uncertainty early in the pandemic (Bandler et al., 2020). Librarians were asked to rate how strongly they agree or disagree with the statement, "The changes in COVID-19 data reporting ordered by the Trump White House in 2020 caused me to have less trust in the Centers for Disease Control (CDC)." Slightly less than half agree or strongly agree (29.03% and 19.35%, respectively; n = 31) with this statement. 19.35% disagree and 6.45% strongly disagree; 25.81% neither agreed nor disagreed (Figure 3).
Another noteworthy event was Pfizer's announcement of an effective COVID-19 vaccine just days after the 2020 election. Donald Trump and Donald Trump, Jr. tweeted that the Food and Drug Administration (FDA) purposely waited until after the election to make the announcement to make Trump look bad (Dean & Dunn, 2020). The survey asked librarians to rate their level of agreement or disagreement with the statement, "Trump's assertion that the FDA purposely waited to announce a successful COVID-19 vaccine until after the 2020 election caused me to have less trust in the FDA." Most respondents strongly disagree or disagree (41.94% and 35.48%, respectively; n = 31) with the statement. Only 9.68% neither agree nor disagree; 9.68% agree; and 3.23% strongly agree ( Figure  3). Approximately 71% of librarians strongly agree or agree (29.03% and 32.26%, respectively; n = 31) with the statement "My trust in one or more US federal agencies decreased during the Trump administration because of the handling of the COVID-19 pandemic by federal agencies." 19.35% neither agree nor disagree and 19.35% disagree (Figure 4).
The survey also asked about experiences with patron trust or distrust of government information relating to the federal COVID-19 response. Kubas found that patron distrust grew leading up to 2020, although both librarians' and patrons' political leanings directly correlated with reported trust levels (2020, p. 317). Most respondents (85.71%, n = 126) to the first survey reported no noticeable change in the level of interest in government information or in requests for government documents from students since February 2020 (Table 6). 76.80% (n = 125) stated they have perceived no indication of a change in trust on the part of library patrons due to COVID-19 and 64.29% (n = 126) have seen no change in patrons' trust levels for other reasons (Tables 7 and 8). Most respondents to the follow-up survey (92.59%, n = 27) said in interactions with patrons, they are usually the one to initiate any conversations about the reliability or accuracy of government information. However, open-ended responses indicated that sometimes patrons do express concerns about bias in government information.

Discussing government information with library users
Reference was the most reported responsibility of survey respondents ( Table 1). The follow-up survey asked for more details about the kinds of interactions librarians have with individuals and how respondents talk about government information. Most librarians (93.75%, n = 32) stated that one-on-one interactions with patrons is a regular part of their job. These interactions include reference duties, research consultations, and informal conversations with library users (Table 9). Most survey respondents    indicated that they sometimes or frequently (46.67% and 30.00% respectively; n = 30) assist patrons with finding government information in oneon-one interactions (Table 10). This assistance includes helping patrons locate a known government information source or helping patrons locate a variety of government information sources ( Figure 5). When asked in the first survey if they are more likely to discuss bias, authority, information creation, or disinformation when recommending government information to patrons since February 2020, responses were split. 47.11% reported they are more likely to discuss these issues with patrons and 48.76% reported no change (n = 121) ( Table 11). The survey also asked librarians if they addressed federal government misinformation or disinformation in their work with library users both prior to February 2020 and since February 2020. Responses indicate little, if any, change (Table 12). 60.00% (n = 30) of respondents stated they sometimes address government information authority, bias, and credibility in one-on-one interactions with patrons; 23.33% frequently do so, 13.33% never do, and 3.33% always do ( Figure 6).
The follow-up survey also asked academic librarians if and how they address the creation processes behind government information with patrons individually. 46.67% (n = 30) of librarians said they discuss aspects of  government information creation in one-on-one interactions with library users sometimes, but 36.67% said they never do so; 13.33% frequently and 3.33% always discuss government information creation with individual patrons. 40.00% of respondents sometimes and 30.00% frequently explain to individual patrons that political climate can affect the content, distribution, or presentation of government information. 26.67% never and 3.33% always do so ( Figure 6).   Most librarians said they always (23.33%), frequently (46.67%), or sometimes (23.33%) recommend that patrons supplement government information with other sources, such as peer reviewed scholarship, news articles, or gray literature; 6.67% (n = 30) never make this recommendation. 46.67% (n = 30) of librarians reported they review source evaluation criteria with patrons seeking government information sometimes; 23.33% never do, 20.00% frequently do, and 10.00% always do ( Figure 6).
Open-ended responses in the first survey indicated that COVID-19 safety measures, like closures, limited services, or remote work, may have affected the nature and amount of patron interactions between February 2020 and September 2021. The follow-up survey asked respondents to describe patron interactions during this period and if they had more, fewer, or about the same number of patron interactions compared to before the pandemic. Most librarians (57.58%, n = 33) reported fewer patron interactions after March 2020 and that these interactions were predominantly virtual (Table 13).

Discussion
According to the survey results, after the first 18 months of the COVID-19 pandemic in the United States, academic librarians reported a range of trust levels toward the federal government and a greater decrease in trust after February 2020. It appears the Trump administration's pandemic response did not have an overwhelming effect on academic librarians' views toward government information. Rather, it largely confirmed or deepened attitudes and practices they already held. Public trust in the government fluctuates depending on the party that holds the presidency (Pew Research Center, 2020b, p. 14). The change in administration affected trust in government information for roughly half of the respondents, but it was not wholly clear whether that was related to how Trump's executive branch responded to COVID or the pre-established pattern of false claims and missing information. Scrutiny of the Trump administration's handling of government information began in January 2017 with the removal of LGBTQ information from the White House website (Garnar, 2018, p. 193) and removal of climate change information from the EPA website (Bichell, 2018). Librarians also reported a decrease of trust in federal agencies under the Trump administration, but an increase of trust in federal agencies since President Biden took office in January 2021. In general, people tend to trust agencies more than individuals who hold elected office, such as the president or representatives (Houston & Harding, 2013, p. 55). Pew Research Center surveys conducted at the beginning of the pandemic show 83% of people were either confident or somewhat confident the Centers for Disease Control (CDC) was doing a good job with COVID-19 response, compared to 45% and 48% who felt President Trump and Vice President Pence, respectively, were doing a somewhat or very good job (Pew Research Center, 2020a, p. 5). As described above, the CDC's changes in COVID-19 data reporting led to a measurable decrease in trust among academic librarians. However, accusations that the FDA delayed a vaccine announcement until after the election did not have the same effect. This comparison indicates that abrupt changes and lack of transparency in data reporting and public health guidance is far more detrimental than politically motivated criticism from elected officials and their relatives.
It is also worth noting that, while the first survey took care in distinguishing between trust and distrust, the distinction did not seem to transfer to librarians' descriptions of their own views toward government information. Given the choice, a few respondents admitted some degree of distrust in the reliability of government information prior to the pandemic (Table 3). However, librarians did not use the word distrust in free text responses to describe their own feelings or positions. Rather, they described fluctuations in degrees of trust. A small number of respondents referred to faculty or community members as distrustful toward the Trump administration, though. It is unclear if this is simply a matter of conversational use of the terms that may be less rigid than the survey's definitions, or if respondents simply did not feel active distrust toward government information at the time.
Many librarians addressed elements like bias and information creation when discussing government documents and resources before the pandemic began, and they indicated they continue to do so, perhaps with somewhat greater caution than before. Interestingly, almost half of respondents said they were more likely to address "bias, authority, information creation, or disinformation when recommending government documents to patrons one on-one." However, a follow up question about whether librarians "addressed federal government misinformation or disinformation in [their] work with library patrons" showed almost no change from before and after February 2020. Multiple explanations might account for this. The first question asked about a broader selection of topics such as bias and information creation, while the second focused specifically on mis-and disinformation. Some may feel more comfortable with generalized discussions of bias rather than pointing out distinct instances of wrong information. Additionally, fewer reference interactions over the last 2 years may create fewer opportunities, so librarians might intend to bring up misinformation but have not yet done so.
One librarian commented that they "try to find a gentle way of reminding the patron to keep [the potential for political bias] in mind" when recommending government information in one-on-one interactions. Another respondent explained that discussing bias in government information is a way to "empower citizen engagement" and discuss how government information affects people. Another said they explain to patrons that "different government documents have more reliable information than others." Two respondents stated that they tell patrons government information is a reliable source for "certain kinds of information" and the "best source" for Census data and health information.
Sometimes data can be used and shared without an easy opening to discuss the information creation process or disinformation concerns regarding other areas of government. However, it is possible to rely on a person or agency without feeling trust for them (Kelsall, 2021, p. 290;Wilholt, 2013, p. 234). For example, if the federal government is a good source of certain kinds of information, we can rely on it to provide that information without ourselves having feelings of trust toward the government. Given that people can trust the likely accuracy of some government data while having less trust in the motives of other elements, it follows that different types of information will remain useful and necessary for academic research regardless of views toward the current administration or specific agencies (Bruno, 2017, p. 303).
Respondents noted that time limitations and the nature of the reference interaction or research consultation can make conversations about source evaluation, authority, bias, or credibility difficult: "[T]here is so much ground to cover in my one-on-one interactions, these unfortunately fall to the wayside. " Furthermore, when a student needs a specific kind of government information, "it would be difficult to complicate [Bureau of Economic Analysis] or [Bureau of Labor Statistics] or Census data, for example, because where else would we turn?" As academic librarians are less likely to offer detailed reference interviews in a virtual setting, they may not provide that context and nuance for patrons that simply want a quick answer (Logan & Lewis, 2011, p. 224). The pandemic has had a noticeable effect on how librarians provide reference and research assistance, which has altered some aspects of services. Most academic libraries moved to virtual reference and answered far more questions than before about technology troubleshooting and how or whether patrons could access the library's physical collections (Kathuria, 2021, p. 114;Yap & Manabat, 2021, p. 179).
Meeting an immediate information need can take priority over teaching political literacy more broadly in patron interactions. "Often, librarians are teaching to a particular format or resource and not always digging into how or why one should evaluate or trust government sources" (Kubas, 2019, p. 138). Patrons' government information needs are driven by assignments, research topics, or discipline, and students do not always know that government information will meet their needs. Respondents explained that they may recommend specific government entities, such as certain agencies, that would provide the relevant information, or they may recommend a particular kind of resource, such as data.
Another said that, in individual research consultations which allow for more time than a typical reference interview, there is "time to talk about some of the underlying creation processes and how that affects what information they may (or may not) be able to find. " Librarians also discuss authorship and the purpose of different kinds of government information. This can be especially important since students do not always readily identify government information when it is found online (Brunvand & Pashkova-Balkenhol, 2008, p. 202). Even in shorter reference interactions, a librarian can point out that a given resource comes from the government and indicate which agency or person is responsible for it. More recognizable publishers, such as elected officials or agencies in the news like the CDC, open an easier avenue to discuss elements such as bias and credibility.
COVID-19 continues to cause considerable upheaval in US academic libraries, along with the rest of the country. Consequently, the results of these surveys must be viewed through that lens, with acknowledgement that patron interactions between the Spring 2020 and Fall 2021 semesters are likely to be atypical compared to prior years. Further study on this topic once a so-called new normal is established would provide additional insight. The authors caution against generalization based on the results of the second survey due to the small population size. However, the results of both surveys raise thought-provoking questions about how librarians work with resources they themselves may not fully trust. A more nuanced look at librarians' trust in government information would be interesting, for example if and how librarians compartmentalize personal feelings about government actors when dealing with these resources professionally. Expanding to a larger audience, whether a greater number of academic librarians or a broader cross section of the library field, would also be worthwhile.

Conclusion
Greater overall political literacy can help students draw connections between resources such as Census data and presidential Twitter accounts that they might not have seen otherwise, but the reference interview does not always allow for that. Use of government information in reference services must be supported by a greater focus on political literacy in the classroom and elsewhere to ensure students develop a more rounded view.
The need for increased political literacy among student researchers using government information is clear, and many librarians try to address that need when possible. While the Trump administration's COVID response raised the visibility of this issue, most academic librarians made efforts to educate students on this topic prior to 2020 as well. Current challenges exist in the form of shortened virtual reference interviews, patrons who need or want a quick response rather than a teachable moment, the awkwardness of discussing bias and disinformation around deeply partisan political topics, and librarians who lack confidence with government resources. For the next few questions, you will be asked about your trust in government information. Trust in government information is defined as having a reasonable degree of confidence in its factual accuracy, and in the integrity of the information creation and publication process. Distrust is defined as viewing government information from a place of inherent suspicion and skepticism regarding its accuracy and integrity. Based on these definitions, how much trust did you have, in your professional capacity, in the reliability of federal government information prior to February 2020? a. Show the patron how to locate a known government information source to meet an information need b. Show the patron how to locate a variety of government information sources to meet an information need c. Discuss authority, bias, and/or credibility of government information with the patron d. Explain to the patron that authority varies among different kinds of government information e. Discuss aspects of government information creation with patrons f. Explain to the patron how the political climate affects the content, presentation, and distribution of government information g. Review source evaluation criteria with patrons seeking government information sources h. Recommend that patrons supplement government information with other sources such as peer reviewed scholarship, news articles, or gray literature 18. In one-on-one interactions with patrons, discussions about the reliability or accuracy of government information are usually initiated:

ORCID
a. By me b. By the patron c. Not applicable 19. Please tell us more about how you use or discuss government information in one-on-one interactions with patrons. 20. Please rate how strongly you agree or disagree with the following statements. (POSSIBLE CHOICES: Strongly agree, Agree, Neither agree nor disagree, Disagree, Strongly disagree) a. It is necessary to include government information examples in information literacy instruction b. I feel less trust in government information now due to the election of Donald Trump in 2016 c. I have less trust in government information due to some statements or actions of Trump prior to February 2020 d. I feel less trust in government information now due to the federal government's response to the COVID-19 pandemic e. I feel less trust in government information now due to the election of Joe Biden in 2020 f. My trust in one or more US federal agencies decreased during the Trump administration because of the handling of the COVID-19 pandemic by federal agencies g. The changes in COVID-19 data reporting ordered by the Trump White House in 2020 caused me to have less trust in the Centers for Disease Control (CDC) h. Trump's assertion that the FDA purposely waited to announce a successful COVID-19 vaccine until after the 2020 election caused me to have less trust in the FDA i. My trust in federal agencies regarding COVID-19 safety has increased since President Biden took office in 2021 j. I have less trust in government information because of Donald Trump's statements about hydroxychloroquine, mask use, and/or the COVID-19 vaccine k. I have less trust in information produced by federal agencies because of the way agency leaders responded to COVID-19 l. My trust in government-produced scientific or medical information (e.g., that produced by the CDC, FDA, EPA, NIH, or other federal agencies) decreased under the Trump administration