This is the third analysis of results from the Academic Library Response to COVID19 survey, which we deployed on March 11 in order to gather as-it-happens data from and for the academic library community. Libraries were encouraged to not only log their current status but to also come back to retake the survey as circumstances evolved. The first update was received that same day at 11 pm, illustrating just how quickly things have been developing on campuses.  

In past posts, we have presented an analysis of the first 24 hours of responses and a comparison of those findings with an analysis of the responses in the subsequent 48 hours. Today we present an analysis of changes in responses from first submission to the most recent update for the 253 institutions that have returned to the survey to log an update.  In contrast to the first two analyses, which presented aggregated data across institutions, this analysis is longitudinal and reports on changes within institutions. 

Please have your library fill out the survey if it has not done so. If your library has filled out the survey and policies or service offerings evolve, please submit an update. We structured the survey to accommodate libraries submitting updates as often as they need to. (In fact, there is one library that has already returned to log five updates!) The link to the survey with all of the questions on a single page is: https://surveys.ithaka.org/jfe/form/SV_8qN8F2274hMBBBz. The link is the same for taking the survey the first time and for submitting an update. You may also view a live results report as surveys are submitted at: https://tinyurl.com/covidlibrary. Because the results are updated live, they will not match the data reported here. 

The data here were extracted at the end of the day on March 21 and thus represent data from the first 10 days of data collection. In that time, of the 721 that had responded to the survey, 253 institutions submitted at least one update and, of those, 60 submitted more than one. Please note that we have not yet analyzed the open-ended text entered with “other” responses.

Summary of findings

Institutions continue to move classes online. There is a clear relationship between online class delivery and libraries closing or limiting their hours; however, these changes in library hours and services are often time-delayed. Libraries have pivoted reference services to online/phone delivery. Access to print collections, whether onsite or via delivery, has declined significantly. Given these shifts, it is understandable that libraries are expanding who is allowed to work remotely and, in some cases, have mandated remote work. 

Institutional updates

With regard to the status of classes, the majority of updates report pivoting from classes continuing as usual to in-person classes moved to online/remote instruction (n=30, 12 percent). Given we started data collection at a point when the majority of institutions had already shifted online, it is not surprising that the trend continues in that direction. In fact, of the 253 institutions that have logged updates, 178 (70 percent) noted in their first response that classes had already moved online. We see a similar shift from students being in residence to students being required to vacate (with some exceptions permitted) (n=38, 15 percent). Of the 253 that have provided updates, 67 (26 percent) had already required students to vacate (with exceptions). Forty-four (17 percent) started out with and still have students in residence, and 45 (18 percent) report having no residential facilities.

Library building status updates

Overall, there is a clear relationship between classes moving to online delivery and libraries closing or limiting their hours. Nonetheless, changes in library hours and services are often time-delayed. For example, while 182  institutions had moved classes online at the time of the first survey submission, less than 10 had closed physical library locations at that point.

The most frequent shift that emerges in the analysis of first to most recent reporting is from the library/libraries being open usual hours to the library/all libraries being closed (n=49, 19 percent). Though the data do not indicate when the decision to close was implemented, only when it was reported, on average it takes 5.4 days for libraries to close after being fully open, according to the submitted data.

A number of other relatively more frequent patterns of changes emerged as well. The first is that a single location library that was open with limited hours shifted to being closed (n=31, 12 percent) with an average of 4.6 days from first to final report. The second is that moving from having a multiple location library open but with some hours limited and/or some locations closed to all libraries closed (n=25, 10 percent) averages 5.3 days from first to final report. And, finally, moving from a single location library being open usual hours to being open but with limited hours occurred in 23 (nine percent) libraries. 

Access to services and collections updates

Reference services are continuing to shift to online/phone delivery (n=128, 51 percent), as we would expect given libraries limiting open hours and closing. Specifically, for those previously reporting that reference services continue as usual (in person and online/phone), 96 (38 percent) shifted to reference services are limited to online/phone only. And, for those that previously reported reference services continue as usual but hours limited, 32 (13 percent) now report that reference services are limited to online/phone only. Fifty-seven institutions (23 percent) had limited reference services to online/phone formats at the outset of data collection.

Ninety-seven institutions report that access to print collections has been discontinued. Of those, 43 (17 percent) had originally reported that access to print materials continues as usual and is only onsite and 39 (15 percent) had originally reported that access to print materials continues as usual and is onsite and/or via delivery. It is particularly notable that not only has onsite access been discontinued for the latter group but that delivery has also ceased. 

Library staff impact updates

With respect to policy on remote work, there is a shift not only to expanding who is allowed to work remotely (n=69, 27 percent) but also to mandating remote work (n=50, 20 percent). For many institutions logging their first response, remote work had already been expanded (n=84, 33 percent), but only three (one percent) had already moved to mandating it.

Specifically, of those reporting a change in remote work policy from first to most recent update, the following patterns emerged as the strongest: 

  • From remote work policy has not changed and remote work is not allowed to remote work policy has changed and expanded who may work remotely (n=21, eight percent)
  • From remote work policy has not changed and remote work is allowed for some employee groups to remote work policy has changed and expanded who may work remotely (n=42, 17 percent)
  • From remote work policy has not changed and remote work is not allowed to all employees are required to work remotely (n=13, five percent)
  • From remote work policy has changed and expanded who may work remotely to all employees are required to work remotely (n=25, 10 percent)

As time continued, we had expected to see a pivot towards more regular communication to staff with updates and guidelines on safety measures as a safety measure. We were surprised to see that not only was our expectation not borne out by the data, but that there are reports that this strategy is no longer in place at many institutions where it once was occurring. Specifically, of the 71 libraries reporting a change on this measure, only 30 pivoted from not having regular communication to having regular communication, while 51 pivoted from having regular communication to not. What this may reflect, however, is that what is perceived as “regular communications” when changes were minimal now feel inadequate when change becomes more significant or perhaps is happening more quickly. This will be something to investigate in the future, perhaps through interviews or other methods that are better suited to examine perceptions and lived experiences.