The Assessment, Insights, & Measurement (AIM) Team of the Office of Data Analytics (ODA) administers a number of surveys on a regular or cyclical schedule (e.g., surveys of alumni, graduate students, and seniors; campus climate; administrator appraisal) and client-requested surveys on special topics or for special populations on an ad hoc basis (e.g., faculty housing, instructional technology use, School of Education doctoral students).
Survey respondents sometimes contact ODA with questions and concerns they have about taking part in a survey. Answers to questions and concerns that arise are presented here.
People who fill out our surveys are concerned about the confidentiality of the information they give us, and they want to know who will have access to that information. Survey participants' privacy is of utmost importance to us, and strict internal safeguards are in place to ensure that privacy.
The data we collect from people are confidential. No one outside ODA staff responsible for the survey administration—no students or students’ parents or faculty members or administrators—has access to survey respondents’ identities or to files that may connect names or email addresses with answers to survey questions.
One person here manages each survey database, wherein is recorded who has and has not completed the survey questionnaire. Members of our staff carry out all of the analyses of the data, and we organize and write up the results. Questionnaire responses are aggregated for statistical analysis and reporting; for example, it may be reported that X% of seniors were satisfied with their academic experience at CU Boulder and that Y% were dissatisfied.
In cases where the data are particularly sensitive (e.g., faculty members’ evaluations of a dean), we take extra precautions to safeguard the data. When data collection has been completed, we remove from the final data file the identifying information that permitted the database manager to keep track of who did and did not respond to the questionnaire. In other words, each case in the final data file is an anonymous case. In addition, the data file is encrypted and password-protected.
Data from some surveys include written comments made in response to open-ended questions, such as this question from the 2009 Graduate Student Survey: "If you think it is unlikely or you areuncertain that you will complete your degree, please tell us why." Here is how we treat such data:
- We collect these comments into a text file and generate data tables that summarize the aggregated responses. Such a table may report, for example, that X% of those answering the question reported financial reasons, Y% reported family issues, and Z% reported academic difficulties.
- We provide these tables and the text file of the verbatim comments to the appropriate audience, e.g., the Graduate School, the Admissions Office, the chair of the Administrator Appraisal Program committee. Prior to release, the text files of comments are stripped of any identifying information used by our office, e.g., a randomly assigned survey identification number. It is possible, however, that in the written comment(s), a person may unintentionally include information that reveals his or her identity, e.g., "As the only minority female doctoral student in department X, I …." We do not remove such comments.
- In some cases, we also post written comments. The tendency in recent years has been to password-protect the posted comments, restricting access to CU Boulder employees with internal, administrative needs for viewing such data. For example, note the password-protected comments from the Spring 2008 Senior Survey (click on the "open-ended items" link in the sidebar to the left).
ODA safeguards any sensitive employment and student data it accesses by using state-of-the-art data security protocols. The survey platform, Qualtrics, meets Vendor Security Assessment Questionnaire(VSAQ) data standards. Data are encrypted in transit and at rest (Read more about Qualtrics security). Once the survey has closed, data will also be encrypted in transit to ODA. The full database of anonymized responses will then be held in encrypted form on the secure ODA server. The de-identified data will be stored to allow investigation of trends by comparison with future surveys.
Some surveys we administer include financial incentives, typically a random drawing for one or more cash awards. As part of our procedures to ensure confidentiality, we do not ask for any identifying information in the survey. Because there is no place to enter their contact information (name, email) in the survey, participants sometimes wonder how we determine what names to include in the drawing.
When a person submits a survey, he or she is automatically entered in the pool of people eligible for the awards. Here’s how it works. When someone accesses and fills in the online survey, the survey management system uses a random survey-specific identification (ID) number to keep track of that response. For example, when Hermione Granger submits her survey, the survey management system will register that survey 5555555 (sent to Hermione’s email) has been received. The random ID numbers of individuals who have responded to the survey can then be linked with email addresses in the database to distinguish those who have responded from those who have not. We can then send reminders only to those people who have not yet responded to the survey (no further reminders will be sent to Hermione, for example). We can also keep track of the people who are eligible for the awards drawing (5555555 will be included in the pool for the awards drawing). At the close of the survey, we randomly select the award winners from that pool of ID numbers, and we use the linked email addresses to notify the owners of each that they have won one of the awards.
Different surveys have different purposes and are administered to different groups of people, e.g., CU Boulder alumni, grad students, seniors, or incoming freshman students. Such a group is called a "survey population."
For some surveys the population is sufficiently small, and all members of the population are invited to take part in the survey. This was the case, for example, in the CU Boulder Senior Survey in spring 2008. All CU Boulder seniors were emailed and asked to complete a survey to share their opinions and feedback on their experiences at CU Boulder to help the university improve the undergraduate experience for others.
For other surveys, the population is too large for us to efficiently survey all of its members, or it may be unnecessary to survey all members of the population. For example, the population for the CU Boulder Campus Climate Survey in 2006 included all undergraduate and graduate students. In this case, only some of the people in the population were invited to complete a survey. These people were randomly sampled from the population and assumed to represent that entire group.
Sometimes people are excluded from a survey population. For example, if a student has placed a privacy flag on her or his records, we won’t include that person in the survey population. Some surveys will not include students who are not full-time or who are not degree-seeking students.
ODA is transitioning from providing static reports to presenting survey results within Tableau so that clients can find what they need (e.g., results by school/college). We are in the process of updating the website to provide the most recent report available for our major surveys, in addition to Tableau reports if available. If you are interested in the results of a specific survey, please contact us.
Various members of the CU Boulder campus community use data generated by different surveys, and some data may also be useful to a broader audience, including people at other colleges and universities, and the general public.
- Data from some surveys are of primary interest to a single CU Boulder office or unit. For example, findings from the Admissions Survey help Admissions Office staff improve the information and services they provide to prospective students and their parents.
- Data from other surveys may be used by a broader array of campus units.
- Findings from the Senior Survey may be used by department chairs and faculty to modify teaching practices or to improve student advising services.
- Findings from the Campus Climate survey may be used by (among others) departments, classroom instructors, Disabilities Services, the Office of International Education, and the Office of the Vice Chancellor for Diversity, Equity, & Community Engagement—to monitor and improve policies and practices that make CU Boulder a supportive environment for all members of the campus community.
- Staff members in ODA incorporate some survey findings in reports to various constituencies, including the Board of Regents, accrediting agencies, and college guides such as the Princeton Review Complete Book of Colleges and the Fiske Guide to Colleges.
- Relevant findings from several major surveys (e.g., National Survey of Student Engagement, Senior Survey, Graduate Student Survey) are routinely reported as part of the CU Boulder Academic Review and Planning process, the regular review of all academic programs required by the University of Colorado Board of Regents and the Colorado Commission on Higher Education.
- As noted above, findings from surveys are posted on the ODA website and available to the public. This information may be used, for example, by other colleges or universities to compare themselves with CU Boulder.