Overview
The use of student financial aid data is restricted based on the Higher Education Act of 1965 (HEA). These restrictions limit the purposes for using student-specific financial aid data, and the disclosure of the identity of financial aid recipients. These restrictions are in addition to FERPA guidelines that apply to student data more broadly and to critical data storage guidelines from the Gramm-Leach-Bliley Act (GLBA).
You can find more information on these financial aid-specific student privacy guidelines in these resources on the IR Data Guide website:
• IR FA Data User Agreement
• IR FA Data Sharing Decision Tree
• IR FA Data Use Guidelines (requires IU Staff Login - Learn about IU Login)
• IR FA Data Elements – Tables, Views, and Columns (requires IU Staff Login - Learn about IU Login)
Based on the documents above, best practices are to only publish and share descriptive statistics about students’ FAFSA data and related award information in a de-identified, aggregate format. De-identified data are presented in a way that readers cannot determine who the aid program applicants and participants are, including by a combination of attributes (race, gender, residency, course load, class section, academic program, etc.). Historical IR practice at IU is to mask headcounts of less than 11, and to not disclose percentages that reflect headcounts < 11. This Technical Guide provides student privacy tips for storing financial aid data, and for creating reports with financial aid data. IU report developers who store, use, display, or share student financial aid information need to modify their reports to match the guidelines in this document and in the linked documents above.
Storing Financial Aid Data
For the purpose of HEA student privacy, if important research to improve student outcomes needs to be done by IR, IR developers can store financial aid flags and categories in limited-access SQL-only tables for developers (e.g. tables ending in ST that only IR_DEV_R and FA_DEV_R users can access via SQL).
Please also abide by guidelines from the Gramm-Leach-Bliley Act (GLBA) Safeguards Rule, as described in the IU FERPA Tutorial and in the IU Critical Data Guide. Please also abide by these IU Data Management and Sharing Policies:
• https://policies.iu.edu/policies/usss-05-release-student-information/index.html
• https://policies.iu.edu/policies/dm-02-disclosing-institutional- information/index.html
• https://policies.iu.edu/policies/dm-01-management-institutional-data/index.html
Masking Headcounts < 11
If there is a report that is available to IU staff beyond Financial Aid and Institutional Analytics full-time staff, we need to mask small cell sizes where the headcount is less than 11. Below is some Tableau calculated field logic from an IA Retention report that changes headcounts less than 11 to -1 for University Internal, full-time staff who are not in the secure FA/IA user group iu-entlmt-uirr-sis-faosfa-uf. (This works if you add both of the following Tableau/ACM/Grouper security groups to the permissions for viewing in your Tableau report: iu-entlmt-uirr-sis-fa-osfa-uf and iu-entlmt-app-uirr-ua-nfstu-uf.)
if ISMEMBEROF('iu-entlmt-uirr-sis-fa-osfa-uf')
then ([Stu Use4 Base Pop Calc Nbr unmasked]
elseif ([Select Attribute 1]='21st Century Recipient' or [Select Attribute 1]='Pell'
or [Select Attribute 2]='21st Century Recipient' or [Select Attribute 2]='Pell')
and sum([Stu Use4 Base Pop Calc Nbr unmasked])< 11 then -1 else sum([Stu Use4 Base Pop Calc Nbr unmasked])
end
Once the calculated field above is created in your Tableau workbook, right-click on the calculated field in the measure values window on the left-hand side of the Tableau developer worksheet. Next select Default Properties > Number Format > Custom. Then in the Custom Format box, type something like the following to
indicate a masking symbol for values < 11: #,##0;<11 .
Note that the logic above still enables IA staff with IR_DEV_R access to see small cell size values. Here are two ways for IA Staff to validate that the masking logic above works for IU staff outside of IA and FA:
1. In Tableau Desktop, navigate to a dashboard, and in the lower right corner, change the user filter to your office group name such as UIRR Tableau Publisher. Then modify the filters to display rows that are expected to have cell sizes < 11.
2. Log into the Tableau Server (https://tableau.bi.iu.edu/) using the username and passphrase for your office publishing group (e.g. UIRR Tableau Publisher). Then in Tableau Server, view a dashboard with the masking logic above and modify the filters to display rows that are expected to have cell sizes < 11.
Dashboard Metadata for End Users
In reports that mask headcounts < 11, consider how to add notes to the Tableau dashboards to clarify what is being masked and for whom. Here is an example of this in an IA Retention dashboard:
Note: On some pages, if you select the Student Attribute “Pell” or “21st Century Recipient” and the cohort headcount for that row is between 1 and 10, you may see a double-quote (") displayed instead of actual headcounts on that row. This is to comply with Federal Higher Education Act guidelines for guarding personally identifiable information (PII) for students who filed a Free Application for Federal
Student Aid (FAFSA). Both Institutional Analytics/Compliance & Reporting staff and your campus Institutional Analytics/Compliance & Reporting Last Update:
institutional reporting office staff have access to see actual headcounts less than 11, so you may contact us with questions about those headcounts.
Dashboard Design Adaptations In some cases, a report or combination of available reports will show total headcounts that enable users to deduce the values behind asterisks. In these cases, please modify the dashboard design in one or more of the ways listed below to ensure student privacy for headcounts < 11. To view the examples below with masked values, be sure to first log into the Tableau Server (https://tableau.bi.iu.edu/) using the username and passphrase for your office publishing group (e.g. UIRR Tableau Publisher).
* Combine Masked Headcounts and Displayed Percentages
1. In some cases, headcounts need to be masked (e.g. with an asterisk, double-quote, or literal “<11”) while percentages can still be displayed in a way that protects student privacy. For example, use the following Undergraduate Beginner Cohort Retention Report tab labeled “Retention Characteristics,” and select the following filters:
2. Dashboard:
a. https://uirr.iu.edu/facts-figures/outcomes/retention/index.html
b. Direct link to Tableau workbook:
https://tableau.bi.iu.edu/t/prd/views/ir_retn_ofcl/RetentionCharacteristics?:showAppBanner=false&:display_count=n&:showVizHome=n&:origin=viz_share_link
3. Filters:
a. Campus = IUPUC only
b. Attribute 1 = Ethnicity
c. Attribute 2 = Pell
* Display “Not Applicable” or N/A instead of disaggregating by student attributes
1. In some cases, a dashboard can be modified to hide data when a financial aid attribute like Pell is selected. In the following example from the Time to Degree report on the Population tab, there is no breakdown by whether students have or don’t have a financial aid attribute like Pell. Instead, where headcounts and percentages are shown, the attribute label appears as N/A (not applicable).
2. Dashboard:
a. https://uirr.iu.edu/facts-figures/outcomes/degrees/bachelor-degree-recipient/index.html
b. Direct link to Tableau workbook:
https://tableau.bi.iu.edu/t/prd/views/uirr_sr_time_to_degree/Population?:showAppBanner=f alse&:display_count=n&:showVizHome=n&:origin=viz_share_link
3. Filter:
a. Pell Recipients or 21st Cent Schlr Recip
4. Note that on the same dashboard, on the graph in the bottom right corner, we can display the breakdown of average GPA or test scores by the Pell or 21st Century attribute because we are showing an aggregate measure with no way to determine any headcount. Institutional Analytics/Compliance & Reporting Last
* Create a Separate Dashboard with Less Granular Attributes
1. When totals are required, a separate dashboard for financial aid reporting can be created with less granular attributes. In the following example from the
Official/First Day Headcounts Report
(Enrollment Summary Tool – EST), the population is limited to full-time, degree-seeking undergraduates (a large population by definition). In addition, the attributes are limited to those that will not reveal small cell sizes.
2. Dashboard:
a. https://uirr.iu.edu/facts-figures/enrollment/index.html
b. Direct link to Tableau workbook:
https://tableau.bi.iu.edu/t/prd/views/uirr_sr_est_official_enrollment_UI/HeadsandHoursPell2 1stCentury?:showAppBanner=false&:display_count=n&:showVizHome=n&:origin=viz_share_link
