HB 221-WORKFORCE & ED RELATED STATISTICS PROGRAM  8:04:04 AM CHAIR DRUMMOND announced that the only order of business would be HOUSE BILL NO. 221, "An Act relating to the duties of the Alaska Commission on Postsecondary Education; relating to a statewide workforce and education-related statistics program; relating to information obtained by the Department of Labor and Workforce Development; and providing for an effective date." 8:04:27 AM GEORGE ASCOTT, Staff, Representative Harriet Drummond, Alaska State Legislature, presented HB 221 on behalf of Representative Drummond, prime sponsor. He read the first paragraph of the sponsor statement [included in the committee packet], which read as follows [original punctuation provided]: House Bill 221 will help Alaska prepare to best allocate and utilize increasingly scarce resources available for education and workforce training by clarifying the authority of the Alaska Commission on Postsecondary Education (ACPE) to receive and analyze existing data from state entities through a statewide workforce and education related statistics program. This system, currently known as the Outcomes Database, will be known as the Statewide Longitudinal Data System (SLDS). MR. ASCOTT next offered a sectional analysis [included in the committee packet]. Regarding Section 1, he said currently ACPE is allowed to adopt regulations to administer financial aid programs for institutional authorization functions and interstate compacts, including the collection and confidentiality of data. Under HB 221, new language in Section 1 would create additional authority for ACPE to adopt regulations relating to the collection and analysis of K-12 data, as approved by the Department of Education & Early Development (EED), under its current authority to collect that information. Further, Section 1 would add regulations for a new purpose, which is the administration of a statewide workforce and education-related statistics program. MR. ASCOTT addressed Section 2. He said currently law allows ACPE to collect data and share it with the governor, the legislature, and other state and federal agencies, but ACPE does not have clear authority to maintain and analyze the data. Under HB 221, Section 2 would allow ACPE to maintain a database for the purpose of administering a statewide workforce and education-related statistics program and enter into cooperative agreements regarding education and employment with other agencies. The commission would also be charged with removing personally identifiable information (PII) before putting such information in the database. Section 2 would also specify that PII may not be provided to the federal government. MR. ASCOTT addressed Section 3. He said currently the Department of Labor & Workforce Development (DLWD) is not allowed to share data regarding unemployment insurance with other agencies. Under HB 221, Section 3 would change the law to permit DLWD to share such data for the purpose of the statewide workforce and education-related statistics program, as permitted in federal regulation and contingent upon a written agreement with ACPE. Section 4 would allow DLWD to disclose data to ACPE [regardless of when it was obtained]. Section 5, he said, would provide for an immediate effective date. 8:09:21 AM STEPHANIE BUTLER, Executive Director, Alaska Commission on Postsecondary Education (ACPE), Department of Education and Early Development (EED), offered background information regarding ACPE. She said the mission of ACPE is "to provide Alaska's citizens with tools and resources to gain access to and to experience success in higher education." To meet that mission, ACPE manages the Alaska Performance Scholarship (APS) and the Alaska Education Grant (AEG) programs, offering low-cost student loans and providing a number of planning programs. She continued as follows: But are these the most effective ways to increase access and success? In this time of severely constrained resources, are we using those increasingly limited resources in ways that produce the best return for our students and the Alaska public, to whom we are responsible? What is the return on investment in programs like the APS and AEG? Or how 'bout that old one: loan forgiveness? The things that we can easily measure, like the number of program participants, or even the number of program completers, cannot answer that return on public investment question. To answer that [return on investment] (ROI) question, we need to know the long-term outcomes across different public sectors. For example, Alaska spends approximately $5 million annually on Alaska Grant Program awards. Do recipients graduate at higher rates than non-recipients? Does the program help recipients enter the workforce sooner than similarly situated non-recipients? Do they earn more once they reach the workforce? MS. BUTLER returned to the subject of forgivable loans. She said if there had been a cross-sector database back [when these loans were given], then it would be possible to figure out whether the students who took the forgivable loans completed their degrees and returned to Alaska at rates higher than other students and whether those students worked in Alaska. She said the question is whether money is being spent in smart ways that make the most difference for students and for the Alaska economy. She related that the university and EED indicate the desire to answer similar questions about their programs. 8:11:52 AM MS. BUTLER continued: The challenge to answering questions like this is not that we don't have the data. We do have it. The challenge is that the data is currently housed in separate transactional databases. Each time we have a question like this, about long-term outcomes, we have to put together a data sharing [memorandum of understanding] (MOU), extract the relevant data from our various systems, match it, link it, and then get answers to our questions. And in accordance with federal law, once we have those answers to our questions, we have to destroy the linked data to protect individual privacy, which means if we have a follow-up question, we have to start the whole process again. MS. BUTLER said HB 221 would streamline the process by allowing DLWD to share unit level employment and wage records with the outcomes database, which is currently housed at ACPE. She said the database houses linked information from ACPE, the University of Alaska (UA), EED, and DLWD; however, currently it "is only training in [general educational development] (GED) data" from DLWD. Under HB 221, the database would also house employment and wage record data "to know workforce outcomes." She said once the date is linked, PII would be stripped off and never stored with the linked data. She said the outcomes database offers a faster, more cost-efficient to do longitudinal outcomes assessments by minimizing the proliferation of PII. Further, the database allows the retention of data that can then be used to respond to follow-up questions or to "efficiently perform future analyses." MS. BUTLER said Alaska currently spends more than $2 billion annually on K-12, postsecondary, and workforce training, but it does not currently have a database that can fully provide efficient, cost-effective, and secure feedback to the legislature and administration regarding the value received from those public investments. She said HB 221 would enhance the state's ability to provide the feedback in several ways. It would allow DLWD to provide unit-level employment and wage data to the outcomes database for longitudinal analysis, which is something 28 states already do; it would prohibit the sharing of any unit-level data from the outcomes database with the federal government; it codifies in law that the database must be deidentified and clarifies what that means; and it clarifies ACPE's authority to maintain longitudinal data. 8:14:22 AM MS. BUTLER stated that the proposed legislation would not "commit any funding for these purposes" but "simply allows for the inclusion of this important additional data in the outcomes longitudinal database," which was created a few years ago via federal grant. She continued as follows [original punctuation provided]: Specific to funding, it's our intent to seek resources potentially through another federal grant to provide outcomes reports once this additional data is available in the outcomes database. In this time of such severely constrained resources, we believe there's a significant opportunity to attract grant dollars to the project, recognizing that tough budget times makes it even more important than ever that policy makers be able to access the cross-sector data needed to look at what happens to students in our education and in our workforce pipeline and to what extent they either enter or advance in the Alaska workforce and contribute to the Alaska economy over the long term. MS. BUTLER said she doesn't want to give the committee the wrong impression that ACPE would be able to immediately produce lots of reports and outcomes analyses upon passage of HB 221; funding would be required to use the data and do the analyses. Nevertheless, she emphasized that HB 221 would "get us a step closer" - an important step at no additional cost toward providing the aforementioned outcomes and analyses. One example of an outcome would be the annual APS outcomes report, which was required under law when APS was created. It is a time-consuming and costly report, but is part of the budget each year. She said the long-term goal is to be able to have the ability to quickly, easily, and inexpensively produce reports like the APS outcomes report to look at other programs, including K-12, collegiate, and career education and training - not just the cost of the program, but also the return of the cost investment. Ms. Butler noted that included in the information provided by Mr. Ascott is a list of links to reports from other states that give examples of the types of outcomes reports that can be created using cross-sector information. 8:16:38 AM REPRESENTATIVE JOHNSTON pointed out that there is a zero fiscal note attached to HB 221. She asked if there would be a cost to redacting individuals. MS. BUTLER answered that there is a cost to doing that. She said under the federal grant received a few years ago, the database was created, which included the process for redacting the information and the process for accepting the unit-level DLWD information; however, state law prohibited DLWD from sharing that data with other state agencies. 8:17:24 AM REPRESENTATIVE JOHNSTON, in terms of further federal grants, asked Ms. Butler if she is considering the savings in reports [predicted as a result of HB 221]. MS. BUTLER answered, "What we're hoping to get is a federal grant from the Institute of Educational Sciences that does not require matching funds. But in the event that we were to secure a grant that did require matching funds, some of the existing reporting that we do in the old-fashioned way, such as the APS report - those costs could be used for match." REPRESENTATIVE JOHNSTON asked what Ms. Butler would be looking for from the federal grant. MS. BUTLER replied that the Institute of Educational Sciences, which provided the first grant, encouraged ACPE to apply for a second grant to use the data created in the databases under the first grant. She said, "Our goal would be to produce a number of key, ... valuable reports for legislators and for the administration and then to seek further funding beyond that, based on ... the value of those reports." REPRESENTATIVE JOHNSTON asked for confirmation that what ACPE is seeking in the grant is how best to use the data. MS. BUTLER clarified, "The ability to demonstrate how valuable the data is when it's applied in analyzing outcomes." 8:19:22 AM CHAIR DRUMMOND asked how the unemployment insurance data would be utilized under HB 221, considering that such insurance data is confidential. MS. BUTLER offered an example wherein ACPE would be able to see who participated in the education grant and then determine whether those in the workforce that completed the program were earning wages at a rate different from those who did not complete the program. She said this would always be at the aggregate level; therefore, ACPE would not be tracking individuals, but would be tracking the outcomes of the programs. CHAIR DRUMMOND asked for confirmation that currently ACPE is not allowed to get that wage information from DLWD. MS. BUTLER confirmed that is correct. 8:20:41 AM REPRESENTATIVE PARISH indicated that a handout in the committee packet entitled, "Snapshot: Statewide Longitudinal Data Systems," which shows that 50 states have the ability to connect between systems, while 47 states and Washington, D.C., receive federal statewide longitudinal data systems grants. He remarked, "I like our odds of ... getting a grant." He asked what a "P20W" system is. MS. BUTLER answered that "P20" means preschool through grade 20 or college completion. In response to Chair Drummond, she said the "W" stands for workforce. 8:21:55 AM CHAIR DRUMMOND directed attention to a [questions & answers (Q&A)] document in the committee packet entitled, "Alaska Education and Workforce Outcomes Data System (Outcomes Database) February 2018." She asked what Administrative Order 261 meant. MS. BUTLER answered that Administrative Order 261 (AO 261) was issued by former Governor Sean Parnell directing the ACPE, EED, DLWD, and UA to share data to the extent allowed under law for the purpose of performing outcomes analyses. She said AO 261 also created a governance body consisting of the president of the university, commissioners of EED and DLWD, and the executive director of ACPE, to ensure that the data in the database was used appropriately and cost efficiently. CHAIR DRUMMOND concluded, "And so, the data that currently exists in the database, that ACPE is keeping warm, was collected and first put in the database under Administrative Order 261." MS. BUTLER answered that is correct. CHAIR DRUMMOND stated her assumption that "according to ... the snapshot, ... this system was created by executive order, and ... now we're seeking to modify it legislatively." MS. BUTLER answered that is correct. 8:23:53 AM REPRESENTATIVE PARISH referred to an ensuing sentence in the aforementioned Q&A handout, which read as follows [original punctuation provided]: However, a 2016 amendment to the budget bill prevented use of federal receipts for ANSWERS in FY17, which resulted in a lack of funding to develop the proposed ANSWERS program and use the database created under the original grant. REPRESENTATIVE PARISH asked what the reason was for "turning down federal money." MS. BUTLER explained that at that time, ACPE was going to seek a federal grant in order to use the data, and concerns were expressed by the legislator who introduced the amendment that the protection of individual student data was in question. REPRESENTATIVE PARISH asked Ms. Butler to speak to that concern. MS. BUTLER responded, "We believe that the data in the outcomes database is extraordinarily secure." She explained that the PII is stripped before the linked data is loaded; it is also encrypted once it is loaded and stored in the state's secure data center, where most of the transactional data on which it is based is already stored. 8:25:22 AM CHAIR DRUMMOND concluded that the PII, even though stored in the same vicinity as the transactional data, cannot be "relinked," because the individual identifiers have been stripped away. MS. BUTLER answered that is correct. She added that care is taken to ensure that "they are stored on separate servers." She said, "The design of the process was specifically to inhibit the ability to relink." CHAIR DRUMMOND asked how the outcomes database would remain relevant in the future if it is not currently collecting new data. She asked if someday it would collect new data to add to that which had already been gathered. MS. BUTLER answered that ACPE is able to "do lookback information" using the existing data. She said school districts have expressed interest in looking at graduates from a specific district four to six years later to see where those graduates are in terms of higher education and career preparation. To the second part of Chair Drummond's question, she said the goal is for ACPE to provide grant resources to the partners that provide data to the outcomes database, so they are able to update their data on a regular basis. CHAIR DRUMMOND surmised, "You still have district-level data. So, you can say ... the Fairbanks North Star Borough School District data is identifiable, but not to the individual or school level." MS. BUTLER responded that is correct. She added that a good example of reporting at the district level would be the APS outcomes report. She stated, "If you looked at that and looked at the reports by district, you probably noticed, and may have experienced some frustration, that [for] some of the smaller districts, all of the information is asterisked out. So, if we're reporting at a level where the number of individuals is so small ... that an individual in that district may be able to figure out who it is, [then] that information is not reportable." CHAIR DRUMMOND offered the same problem existed in relation to No Child Left Behind data. She thanked the presenters and noted that she would likely have more questions during future public testimony. 8:28:39 AM CHAIR DRUMMOND announced the committee would hear invited testimony. 8:29:04 AM DOUG WALRATH, Director, Northwestern Alaska Career and Technical Center (NACTEC), relayed that NACTEC is a regional training center, with the primary student population served from the 15 villages of the Bering Strait School District surrounding Nome, as well as Nome public schools. He said NACTEC delivers over 25 unique vocational training programs annually in a variable term format for communities that would otherwise be without such short-duration, intensive format programs. Mr. Walrath said his testimony would cover the administration of a statewide workforce and education-related statistics program. He continued as follows [original punctuation provided]: As a regional training center, it's not uncommon to report on placement into employment, with outcome measures 7 to 12 months following training. For programs serving primarily high school students, such measures neglect to account for the impact of training and keeping students engaged and in school. 75 percent to 80 percent of our high school training population annually is enrolled in the ninth, tenth, and eleventh grades. Employment measures seven to twelve months after training can unintentionally disregard three-quarters of a training population and present a skewed view of the success of training. Across rural Alaska, dropout rates tend to track disproportionately higher, just as graduation rates tend to track disproportionately lower. A measure of persistence would shed light on the value of programs that are keeping students engaged and on a path to graduation. HB 221 presents a tool for measurement purposes through the K-12 years, tracking to both secondary and thereafter employment. Now, this would be of value to determine how a coherent sequence of vocational courses can begin, as we do here, at a junior high core exploration level, leading to certificated career immersion courses by the twelfth grade leading either directly into employment or as a linkage to postsecondary continuation. Here at NACTEC we would find value in longitudinal measures that HB 221 would present the opportunity to capture. For the purpose of student protection, a tool that protects personally identifiable information while using aggregate comparisons would certainly be of value. HB 221 proposes to do this and could provide outcome measures of enhanced value over some of the currently applied outcome measures. MR. WALRATH related that NACTEC has served students for 15 years and would find it valuable to know the impact of training on the students it trains in the region. He further suggested it would be valuable for the state to know the impact of training centers in other regions. 8:32:34 AM CHAIR DRUMMOND recalled that Mr. Walrath had, a year ago, talked about the importance of longitudinal data systems to track what students are doing. She asked how NACTEC would be impacted by the size of student populations, for example when data for groups of five students appears as asterisks. She asked, "Would your students be able to be grouped ... across the villages in groups of more than five, so that you could actually track the data for that particular program?" MR. WALRATH answered, "I believe it could be used in many different ways." He said he often gets asked questions about a select training program. He related that the Nurse Assistant Program population is probably the smallest of any, with the entire class of twelfth graders and adults totaling six, and the student population of just three. He said a one-year snapshot is not as accurate as looking at the trend over several years. He said the data he tracks "could pull the amount of trainees per village." He added, "But I guess it would all depend on what that measure is, if it's looking at a select training program or a track or just an aggregate of students per community. But ... for the most part, I would think that most measures would be above that number of five." CHAIR DRUMMOND surmised that tracking data must take up a significant portion of Mr. Walrath's time. MR. WALRATH responded that it is part of NACTEC's process to use a FileMaker Pro database to track students. The information begins with the entrance of junior high students into the program, and to that is added training programs completed and certificates earned. For example, the database can track when a student gets his/her state identification, followed by a permit, followed by a license. CHAIR DRUMMOND thanked Mr. Walrath for his testimony. 8:37:02 AM CHAIR DRUMMOND announced that HB 221 was held over.