2023 Iowa Child Care Workforce Study

Appendix D

Data Systems Inventory
and Recommendations for Use

Recommendations

Use Case 1: Compiling a comprehensive list of child care providers in Iowa
Use Case 2: Identifying workforce decline in relation to child care need
Use Case 3: Examining and quantifying the impact of participation in T.E.A.C.H. Early Childhood® Iowa (Iowa AEYC)
Use Case 4: Identifying and describing child care providers who leave the field

Conclusion: What Would it Take to Advance an Integrated Solution for Provider Data

Appendix D.1. Data Inventory Protocol

Appendix D.2. I-PoWeR: Iowa’s Early Childhood & School Age Professional Workforce Registry (HHS)

Appendix D.3. T.E.A.C.H. Early Childhood® Iowa (Iowa AEYC)

Appendix D.4. Child Care WAGE$® Iowa (Iowa AEYC)

Appendix D.5. Unemployment, Wages, and Workforce Needs (IWD)

Appendix D.6. Iowa Child Care Resource & Referral (CCR&R)

Appendix D.7. Kindertrack Child Care Provider Licensing Registry

INTRODUCTION

Iowa faces a significant challenge in recruiting and retaining a workforce to serve our early childhood programs. To better understand this challenge and foster evidence-based solutions, we need to understand more about who our workforce is. This requires access to comprehensive, longitudinal data about child care providers and the systems they come into contact with. Currently, the best way to do this is through surveys and focus groups that solicit information from providers about their training, experiences, wages, and challenges in the field. No comprehensive administrative data system captures this information in a format conducive to answering important policy relevant questions.

In 2023 the Iowa Association for the Education of Young Children (Iowa AEYC, an affiliate of the National Association for the Education of Young Children [NAEYC]) sponsored a statewide early childhood workforce study. One explicit component of the study aimed to identify and inventory current statewide administrative data systems and datasets that could be leveraged to continue advancing workforce efforts. Of particular interest were existing systems with statewide coverage that collect data related to workforce characteristics, professional development, training, licensure, wages, employment history and changes, use of other assistance programs, and child care provider registries.

The current report provides details about the steps involved in identifying and documenting existing data and systems, cataloging specific data elements relevant to the childcare workforce, evaluating the strengths and potential limitations of each system, and determining where data from different systems might be combined to provide richer sources of information to answer relevant questions across systems and between children and the workforce that serves them. The value of this approach to connecting and using administrative data is that it could provide comprehensive, population-level information about child care providers in the workforce, progression in child care careers (education), how long they stay in certain jobs and why (retention), the wages they make over time in different positions (compensation), as well as create the possibility for a return-on-investment study to investigate relationships between wages, benefits, and use of public assistance.

Beyond potential use for single investigations, the data discovery process includes envisioning a pipeline of administrative data that could allow for the longitudinal study of workforce changes over time to understand the impacts of current or future workforce investments and to change course more quickly and easily if programmatic efforts are not reaching intended targets or contributing to intended outcomes. Such a data pipeline could be used to regularly address questions about the Iowa child care and education workforce, specifically, it could investigate how training and education lead to higher wages and greater retention in the field, and the subsequent contributions to higher quality of care and better child outcomes. Such information can then be used to improve access to and quality of our child care systems.

APPROACH

During late 2021 and early 2022, the Workforce Study Advisory Committee meetings included discussion to identify potentially relevant data systems and sources to include in the administrative data inventory process. Once identified as a system of interest, the committee also helped identify relevant data owners and contacts with whom the inventory team could meet. Initial meetings with data system owners and teams started in February and were completed in April 2022. Simultaneous with this project, the Iowa Department of Health and Human Services (HHS) systems experienced a significant amount of change, including leadership changes, investments in new technology, and a very large child care shared services¹ effort that includes commissioning the development and building of an internal operational data store to share real-time data between child care businesses and HHS. These developments at the state level impacted our opportunities to interact with state data teams in both positive and negative ways. Some systems we intended to engage with for this inventory process simply could not dedicate the resources needed to do so. However, through our team’s involvement with the operational data store work, we have had more opportunities with some system teams to dig deeper and learn more about some of the systems than we would have through a single meeting as part of this project alone. The summary information presented below and the specific system descriptions provided reflect the information we learned from both our initial data discovery meetings and any additional meetings with specific system teams through March 2023.

¹ Early Childhood Iowa Phase I Shared Services & Child Care Task Force

TABLE 1. RELEVANT DATA SYSTEMS IDENTIFIED AND REVIEWED

I-PoWeR: Iowa’s Early Childhood and School Age Care Professional Workforce Registry
Iowa Department of Health and Human Services (HHS)

T.E.A.C.H. Early Childhood® Iowa
Iowa Association of the Education of Young Children (Iowa AEYC)

Child Care WAGE$® Iowa
Iowa Association for the Education of Young Children (Iowa AEYC)

Unemployment, Wages, and Workforce Needs
Iowa Workforce Development
Workforce Innovation and Opportunity Act (WIOA)
Labor Market Information Division (LMI)
Laborshed Survey
Workforce Needs Assessment
Occupational Employment and Wage Statistics (OEWS) Survey

Iowa Child Care Resource & Referral (CCR&R)
Iowa Department of Health and Human Services (HHS)

KinderTrack Child Care Provider Licensing Registry
Iowa Department of Health and Human Services (HHS)
Iowa Department of Health and Human Services (HHS) - Provider Search

Prior to the initial data discovery meetings, we developed a semi-standardized protocol to collect specific information consistently across each of the data discovery conversations (see Section D.1). The general structure of our protocol was standardized to assess the 1) general description of the system, 2) historical timeline and development of the current system, 3) process of data collection, 4) major shifts or changes in the data collection that would impact longitudinal study (e.g., policy changes, expansion, etc.), 5) current use and analyses of the data, 6) types of identifiers or linkage elements that exist in the data, 7) strengths and limitations of the data, and 8) potential benefit of including data in an integrated data system (IDS) for ongoing or future work within departments or with other partners across departments. Within each of these broad topics, our protocol remained semi-standardized to collect consistent important information from each data system but also allow flexibility in the conversations to uncover where each data system differed in purpose, content, or scope.

In many cases, our data discovery meetings included real-time demonstrations of the data system and relevant components so we could understand how the data are collected and used. These conversations also included exploring potential use cases for their data and how their work aligned with other state priority projects. We then conducted follow-up research to access systems, where possible through public-facing portals or dashboards, download system documentation from public websites, and view training videos for data collectors. Where possible, we also collected data dictionaries and system documentation beyond the materials generally available to the public.

Follow-up to the data discovery meetings was an outline of legal auspices for each data system. This included reviewing federal guidance documents, existing state contracts, and Iowa code (where it exists in relation to the data system). Potential pathways for data sharing were then identified. Many of the systems are housed within state departments where legal data sharing is already happening, so a reasonably clear path could be found. In other cases, legal agreements among state departments were identified as a potential pathway to explore.

SUMMARY OF FINDINGS

The data discovery process identified several strengths and limitations that span across data systems related to their use for longitudinal studies of Iowa’s child care workforce. The primary strength of the full catalog of data systems is the inclusion of unique characteristics and elements collected for specific purposes, that together, can provide a more complete picture of our workforce. Within all the data systems, many systems have staff working to check data, apply corrections when errors are detected, and dedicate effort to keeping data systems updated. Most of the systems are designed to collect data elements under restricted response options (e.g., pull down menus, click boxes, etc.) instead of relying extensively on the accuracy of open-ended text responding. System administrators and staff have also produced documentation for system data in multiple media formats to aid users to enter data correctly. Some of the data systems in our discovery offer simple solutions to linkage across data sources through use of a standardized unique identifier (e.g., state license number, social security number, etc.). Where such identifier keys are not included in systems we profiled, each system does collect and maintain a standard set of personally identifying variables (e.g., name, date of birth, etc.) that can be joined successfully using probabilistic linkage methods. Importantly, each system also includes specific individual-level information (e.g., education level, county, family size, etc.) that differs in availability across all systems, but that can be used as confirmatory evidence in linking two systems that contain the same element.

Although the data systems involved in our discovery process possess numerous strengths, there are two primary limitations across all the systems, regarding holistically studying the child care provider workforce in Iowa. The primary limitations are likely not surprising and certainly not limited to administrative data systems focused on child care. First, it remains unclear whether the set of systems involved in our discovery would capture the full population of child care providers in the workforce. Data systems that primarily contain provider registries do not contain data from unlicensed or unregulated providers. Obviously, this limitation omits all the child care providers in Iowa operating in an unregulated capacity. In Iowa, Child Care Homes can operate unregulated if care is provided in the home environment to five or fewer children. This presents a potentially severe limitation to conclusions or recommendations developed about the Iowa child care workforce from studying administrative data that omits providers in this group. Beyond not appearing in registries of regulated providers, these unregulated home care providers are also less likely to appear in other system data. For example, unregulated home care providers who do not draw salary or register for unemployment insurance would not be included in workforce data on wages. Although these are a few specific examples, the larger implication of such limitations is the potential omission of other care providers where their exclusion is not known.

Second, the frequency and timing of data updates or corrections for any particular system is a limitation. In several cases, updates are not systematic (i.e., they happen when an update is known but not at regular or predictable intervals) and sporadic (i.e., when/if a provider decides to update information but not related to a prompt or use of the data). Efforts to request updates have helped in some systems, but incentivization for systematic profile reviews or routine data collections is not apparent. However, when we consider linking all data across these multiple systems, the timeliness and systematic updating of profile information may be less concerning as the combination across systems would be likely to catch any updates made in one system but not the others. Currently, it is not entirely clear how accurately dissimilar information across systems could be discerned as a needed update in one or more systems or an error only in the system where the dissimilarity appears. Alternatively, timely and accurate updating of non-profile information (e.g., hours, rates, capacity, etc.) are simply not solvable by combining data and do require some capacity to incentivize/require updates.

RECOMMENDATIONS

Given the rich sources of data explored during this process, and using the discussions with data owners about potential needs or uses of their data if it were combined with other sources, the following sets of use cases were developed. These do not comprise the full gamut of possibilities but highlight opportunities that may be of priority relevance for Iowa to consider as a use of integrated data from these systems.

Use Case 1: Compiling a comprehensive deduplicated list of child care providers in Iowa

While many sources involved in our discovery process maintain lists or registries of providers, the degree of overlap across sources is not fully known. In addition, each of the data systems that collect provider information do so at different intervals raising the likelihood of inconsistency across lists where changes or modifications have occurred for a particular provider. For example, a provider that changes name or location might first be updated in systems that directly feed registration or licensure information, but such changes may take time to be incorporated across systems. Alternatively, systems that perform updates based on direct interaction with providers could reflect changes in the shorter term before those changes are officially submitted and processed by a licensing registry. Finally, it is possible that provider changes are not eventually updated across all systems, leading to multiplicative records for the same provider simply due to unconsolidated data. Given these issues with data entry, updates, and modifications, effort to combine lists across multiple sources, with the employment of deduplication procedures could be useful, initially, for yielding a primary list of providers that is most up-to-date and accurate by using information contained in each of the relevant data systems involved.

Once a single deduplicated list of child care providers is available (and maintained with updates), a number of seemingly simple, yet currently quite complicated questions could be addressed. Primarily, a consolidated provider list would be immediately useful for determining the unduplicated number of active providers in the state. This active provider list could also be used to determine both where active providers are and where comparatively large proportions of providers are, or have become, inactive. With the creation of a single unduplicated list of providers that is maintained over time, and possibly constructed backward in time with existing auxiliary data in specific systems, longitudinal patterns of provider stability could be examined. For example, historic trends could be examined to determine where child care providers have decreased (workforce shrinkage) and potentially, in combination with future trend analysis, where the child care workforce might be growing.

To obtain an unduplicated list of child care providers, data would be necessary from the HHS KinderTrack Provider Registry System, the HHS I-PoWeR Iowa Early Childhood and School Age Care Professional Workforce Registry, and the HHS Child Care Resource & Referral system. Importantly, additional potential sources of information about child care providers in Iowa that were not part of the current data discovery should also be considered for inclusion.

Use Case 2: Identifying workforce decline in relation to child care need

Once developed and maintained to include updated information, a single accurate unduplicated list of child care providers could be used, in combination with other system data, to determine where providers are concentrated geographically in the state and where providers might be severely scarce or nonexistent. Specifically, through incorporation of additional statewide data (like the data currently housed in Iowa’s Integrated Data System for Decision-Making; I2D2, including birth records or preschool enrollment), predictive analyses could help determine where future needs for child care are likely to be unmet before the shortage of available care directly impacts families in need of care options. That is, trend in child care provider availability could be contrasted with trend in presence (both births and migration) of young children to identify geographies where need is likely to outstrip availability in the next year or years.

To examine workforce trends and develop forecasting estimates of families who will need care and where those families might be, additional data contained in the IDPH Vital Statistics system and in the DOE PreK and Kindergarten data systems would provide near-population-level information about how many unduplicated children in each birth year could be expected to require care in specific geographical areas of the state. Importantly, the use of both birth record information and PreK/K enrollment information would yield more refined estimates of need in that both native-born children appearing in the birth record and migrating children, who do not appear in the birth record but do appear in Pre K/K would be represented.

While inclusion of birth records and Pre K/K enrollment information would be a minimum requirement, any effort to estimate the number of young children in need of care within the state would be wise to include additional sources of information as well. Specifically, systems currently indexing the number of children receiving care (e.g., HHS Child Care Assistance; CCA, Iowa Head Start, etc.) would lend valuable information about children in the state that might not be captured in either the birth or school enrollment years, but who still represent children in need of child care in Iowa.

Use Case 3: Examining and quantifying the impact of participation in T.E.A.C.H. Early Childhood® Iowa (Iowa AEYC)

With its focus on compensation, retention, and ongoing education for Iowa’s child care workforce, the T.E.A.C.H. data system provides quality data related to tracking which child care providers are using the system to further their educational attainment. However, the system does not systematically collect data about provider income over time. Such information would be useful to examine and quantify the potential impact of program participation on provider compensation/income increases over time. Specifically, combining data regarding T.E.A.C.H. participation over time with data specific to income as a child care provider over the same periods would allow for a longitudinal assessment of income trajectories for providers who participate in the T.E.A.C.H. program.

Combining T.E.A.C.H. participation data with lagged, but longitudinal, information about sector-specific wages from the Iowa Workforce Development (IWD) system could provide an interesting opportunity to answer fundamental questions about the impact of the T.E.A.C.H. program. For example, the combined data could easily be used to quantify the compensation benefit experienced by participants, but it could also be used to examine the lag between participation, credential completion, and corresponding wage increases.

Specifically, such combined longitudinal data could yield an estimate of how long it takes for a program participant to see the corresponding increase in compensation from attaining new levels of education. Importantly, use of IWD employment codes would also allow disentanglement among all-source income, that could rise for other reasons, and income specific to child care provision, where the impacts of T.E.A.C.H. are expected to occur.

A further nuanced possibility could combine both T.E.A.C.H. and IWD data as described above with additional information from child care provider registry systems (i.e., an unduplicated list; see #1 above) to identify child care providers participating in T.E.A.C.H. and providers who are not participating in the program but who match on important demographic characteristics (e.g., education, years in the field, current position, etc.). Identifying close matches where very similar providers differ only in T.E.A.C.H. participation would lend heightened rigor to an evaluation of compensation outcomes and increase the validity of conclusions about differential wage trajectories that differ across T.E.A.C.H. participants and non-participants.

Use Case 4: Identifying and describing child care providers who leave the field

Indexing unduplicated counts of providers, geographic density/scarcity, forecasting gaps in child care availability, and examining programming specifically targeted to increasing compensation through ongoing education are all important pieces necessary to understand, sustain, and grow the child care workforce. However, a large unknown among child care providers focuses on why they leave the field. Answering questions about who leaves, where leaving is more/less common, and particularly, where those who leave are vital to understanding how to target such mechanisms in efforts to keep child care providers in the field and grow the field as the need for care expands.

Combining data sources that include an accurate, unduplicated, maintained (up to date) list of providers, quarterly wage data and occupation codes from the IWD system over time, and educational attainment data contained in T.E.A.C.H., the question of who leaves could be addressed in an informative way. One possibility is that poor compensation drives child care providers to other opportunities that simply pay more. Another possibility is that increased educational attainment, while working in the child care field, opens doors that facilitate moving on to better-paying endeavors within the field of early care and education more broadly (e.g., school-based Pre-K programs). If combined, these data sources could be used to examine wage stagnation as a correlate of leaving the child care field. The data would also be useful for examining whether the loss of child care providers is, unexpectedly, happening at both the lower and upper wage ranges of the field (i.e., those with stagnant low wages and those who are increasing employability through ongoing education). Finally, these data would help inform the nature of outcomes for those who leave the field. That is, do those who leave child care positions quickly see wage increases in other fields? If so, what fields are the primary competitors to remaining in child care positions? Finally, are there a few specific fields that seem to draw child care providers at higher rates, suggesting potential avenues for sustaining the child care workforce through efforts targeted specifically to elevating child care work relative to specific competing occupations?

CONCLUSION: WHAT WOULD IT TAKE TO ADVANCE AN INTEGRATED
SOLUTION FOR PROVIDER DATA

This data discovery process identified multiple data systems with potential to inform a statewide approach to better understanding the needs of our child care workforce. It also prioritized a set of use cases that would demonstrate the value of this capacity by generating valuable information to inform state decision-making.

The following is a set of recommendations for steps necessary to bring these data together for this purpose.

Identify a data stewardship group that would advise the data integration team in the process of collecting, integrating, and using these data. A process similar to I2D2’s Community Advisory Group would be a good example for process and protocol, as it connects discussions to executive level decision-makers and includes relevant stakeholders at each phase of the work.

Determine how priorities would be set and establish a funding plan for the work. This could include alignment of the work with existing funded priorities that may be in place to meet federal reporting requirements or are tied to other statewide initiatives like Shared Services or the building of the realtime operational data store to identify child care vacancies. Funds would be needed both for the initial development and data ingestion process as well as for priority analytics to address relevant questions. If the desire were to have annual/semiannual updates of statewide workforce information, for example, then a sustainable funding source should be identified.
Extend current data sharing agreements with state departments to include the prioritized data systems, and develop new agreements where necessary. Most of the systems included in this review are part of a larger department that already has data agreements in place for integration and use within I2D2. This additional work would require a commitment from executive leadership to authorize agency legal teams to participate and amend those documents to include additional datasets.
Reconvene the data owners and users for a level-2 data discovery process. This will help fill in any missing pieces about how data are collected or currently used, so a streamlined process for sharing, inventorying, and using an integrated system could take place.
Commission departmental data teams to a short-term investment of time that will help build the data ingestion pipelines and routinize the work. The short-term investment will be beneficial for setting up processes that can then later be repeated without a lot of additional investments. This will require executive leadership authorization of time dedicated to work on data sharing, with estimates of time needed varied by data system.

APPENDIX D.1. DATA INVENTORY PROTOCOL

A1. General Description of the System

What is the system?
Why is the data in the system collected – for what purpose?
What type of system?

A2. System Timeline

When did the system/data collection begin?
What system developments have occurred over time?
What changes have occurred to what/how data are collected over key times?

A3. The Data Collection Process

How are the data collected? (Paper, electronic, portal)?
Who collects data from whom (self-entry, staff entry, etc.)?
When and how are data entered into a database?
Are data verified at entry – if so how?
Any known fidelity or quality issues?
How are data updated in the system over time (archived, deleted, overwritten)?

A4. Changes to Data Collection Over Time

Have policy changes altered the data elements collected or collection process?
Have enrollment or participation changes occurred historically – if so when? Why?

A5. How are the Data Currently Used?

Internal analyses and/or reporting?
External (government, public) reporting?
External users (public access, websites, dashboards, etc.)?
Data sharing with other entities?

A6. Types of Identifiers Collected?

Program and/or Site?
Provider, Teacher, Classroom?
Family, Parent/Guardian, Child?
Other (e.g., geography, existing within-system linkages, etc.)?

A7. Strengths and Limitations of the Data

A8. What Benefit Could an IDS Provide for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

APPENDIX D.2. I-POWER: IOWA’S EARLY CHILDHOOD AND SCHOOL AGE PROFESSIONAL WORKFORCE REGISTRY (HHS)

Meeting Date

February 23, 2022

Attendees

Christine Lippard, Cass Dorius, Todd Abraham, Ji Young Choi, Laura Betancur Cortés, Heather Rouse

Interviewees

Erin Clancy, Child Care Program Manager, Iowa Department of Health and Human Services

General Description of the System:

I-PoWeR is a registry system created by the Iowa Department of Health and Human Services to track provider professional accomplishments and new learning opportunities. The system enables professionals working with young children and students to update their qualifications as well as aiding current providers who are searching for further training and professional development opportunities. The I-PoWeR system aims to help providers track completion of their approved training, find opportunities and gain access to HHS-approved training, and serve Iowa regulators and systems by tracking completion of required training and certifications for the Iowa child care workforce within a centralized paperless system.

Timeline and Historical Development:

The I-PoWeR system launched as a pilot release in 2019. The current system full rollout started in 2020 with the expectation that the full child care workforce should be in system at present (early 2022). All child care providers who have been active in the previous two years (2020 – 2022; now 2020 – present) would be represented in the system.

The Data Collection Process:

How are the data collected?

Each care provider requests a role category that is then approved by the provider’s supervisor. After approval, a user profile populates wherein requestors can begin completing their profiles via an online entry portal.

Who collects data from whom?

Currently, all registered, licensed, and unregistered providers who accept Child Care Assistance (CCA) subsidy must create a login/profile in the I-PoWeR system.

When and how are data entered into a database?

In addition to self-entry of data via the portal, the I-PoWeR system also receives data from KinderTrack that is entered by HHS staff. Each user is required to enter education data and once entered, HHS staff review/approve the education information. Confirmation is indicated in the I-PoWeR system by a green check mark on each user’s education profile page.

Are data verified at entry?

Employers must approve each user’s role request. At this point, all data can be checked or verified but this step is not required. I-PoWeR staff do conduct regular checks for duplicated entries but checks are limited to exact matches on first/last name, gender, and DOB. Employers can verify/validate the accuracy of reported employee benefits but validation is not required. I-PoWeR staff have considered possibilities to incentivize employers to verify/validate employment benefit information.

Known fidelity or quality issues?

Because some data are not editable, the possibility of duplication via multiple registrations is possible. Although checks for duplicate records are conducted, the checks are non-systematic and require manual effort. In addition, duplicate checks are conducted semi-routinely (e.g., monthly) but not on a continuous basis. All checks in the system for duplicated records rely on deterministic methods using comparatively few demographic elements. Common names and likelihood of matches on birth dates could lead to erroneous deletion of valid records, or at minimum, extended effort to manually confirm a suspected duplicated identity.

How are data updated in the system over time?

Once a profile is created, the user’s name, DOB, and gender cannot be edited in the system. While the inability to update specific demographic information is helpful for tracking purposes, it does raise the possibility of a duplicated record due to an incorrect (or changed) entry on the original profile. User updates to their education records are self-determined such that there is no prompting to do so. However, I-PoWeR is an evolving single-record system in that when anything in a record is updated, the system overwrites the previous element(s) without an audit log of what was changed, or a process for restoration of the record
before the change was made.

Changes to Data Collection Over Time:

Though I-PoWeR is a relatively new data system, a system-wide record updating process will begin once the system expands to include the new Iowa Quality for Kids (IQ4K®) ratings as part of Iowa’s Quality Rating Improvement System managed by HHS.

How are the Data Currently Used?

Internal uses:

The manager of this system can get demographic summary reports (e.g. Beales code for urban vs. rural, roles, wages, demographics, benefits, etc.) and track educational advancement over multiple pulls. Cutting by login activity within date spans and time stamps on all separate pages are available. Also, I-PoWeR staff use system data to generate various workforce reports that can be tailored to specific time periods and/or categories of childcare providers.

External users:

Professional Development Organization staff can borrow training summaries and create their own, manage training schedule, enrollment and attendance.

Iowa regulatory agencies use the system to track professional development and training completions for childcare providers and educators across Iowa. Data are also shared with KinderTrack (HHS) that requires I-PoWeR registration by all licensed, registered, and unregistered providers who receive CCA subsidies on behalf of eligible children. Providers use the system to locate training opportunities and record all training completions and credentials that can be viewed publicly by child caregivers seeking provider services. Adult participants can use the system to search for and enroll in approved training as well as track training history.

What Types of Identifiers are Collected?

Profile creation generates a Contact ID that is unique numerical identifier that remains with each individual’s record throughout the system. In addition to a static unique ID, participant names (first/last), DOB, and gender, are also available as identifying elements. Beyond basic demographic elements, the system includes personal emails, phone, home address, race/ethnicity, and optional work email fields that can be used for identification, deduplication, or linkage purposes. Additional information includes educational and work history elements (i.e., awarding entity, awarding state, issuing state, license issuance date) that, when included with basic demographics, could assist in confirming/disconfirming suspected duplicate records.

System Strengths and Limitations

Many fields (e.g., employment/role, etc.) are driven by pull-down or expansion response capabilities limiting ability to respond with open-ended text. Such structured response formats ensure standardized data relative to open-ended fields reducing error and effort in coding of response information.

The system includes a chronological record of educational advancement that would allow for examination of attainment trajectories within system, as well as potential linkage of attainment with compensation and/or retention in other systems.

Each ‘page’ in the data system is time stamped allowing for data pulls based on login activity windows (e.g., in last calendar year). Such fine-grained control over informational subsets reduces the effort and resources necessary to work with data over circumscribed periods, as the entirety of the data is not necessary to provide the extraction of interest.

A primary limitation of the I-PoWeR system is that there are no current requirements to update profiles on the part of users, or to verify benefits information by employers. Incentivization might help with employer verifications but lack of profile updating can lead to record creep, where some or all of the existing record is outdated and inaccurate.

A second potential limitation of the I-PoWeR system involves lack of clarity about how the system handles name changes (e.g., marriage/divorce). Because name fields are not editable under a specific profile, motivation to create a new profile following a legal name change certainly exists. An obvious outcome is increased duplication of records within system that may or may not be caught via the manual deterministic process in place. Perhaps a larger concern is the decreased ability to link individuals probabilistically across systems where name changes are both possible and occur as overwrites to the previous name field.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

I-PoWeR staff indicated that they had limited capacity to utilize system data to the fullest potential and those limitations present opportunities for IDS involvement to the extent that static data can inform current and future priorities. For example, yearly data ingestion to an IDS could reduce resources necessary for current analytic/reporting activities that then could contribute to expansion of annual reporting priorities through reduced strains on current capacity.

Staff also indicated the need to create/improve current data checking, cleaning, and deduplication processes. Partnering with an IDS could also use existing I-PoWeR data to develop a protocol for checking, identifying, and cleaning duplicated records, given the expertise in these areas within the IDS. IDS staff could also build a translatable software solution that could be implemented by I-PoWeR staff (e.g., VBA programming in Excel, MS PowerShell development, etc.) to systematize duplication checking, improve the accuracy of checking, and increase the efficiency of checking for duplicated records. An alternative to developing a portable deduplication solution could involve a routine transmission of data (e.g., monthly, quarterly, etc.) to the IDS where a deduplication process is performed with the deduplicated results returned to I-PoWeR staff.

Other Notes:

Strategies need to prompt participants to update data and to incentivize participants to complete information that they are not required to complete (e.g., benefits variables).

APPENDIX D.3. T.E.A.C.H. EARLY CHILDHOOD® (IOWA AEYC)

Meeting Date

March 9, 2022

Attendees

Heather Rouse, Cass Dorius, Todd Abraham, Ji Young Choi, Carla Peterson, Jessica Bruning, Laura Betancur Cortés

Interviewees

Ashley Otte, Director of Workforce Initiatives, Iowa AEYC

General Description of the System:

In tandem with the Child Care WAGE$® Iowa program (see below), the T.E.A.C.H. Early Childhood® Program is intended to address compensation, retention, and ongoing education of the child care workforce in Iowa. T.E.A.C.H. (Teacher Education and Compensation Helps) offers scholarships to childcare professionals pursuing course credits toward specific early childhood credentials or degrees. Comprehensive scholarships support individuals through direct payments to colleges for tuition and reimbursements to learners for tuition, textbooks, and paid time away from work. Each T.E.A.C.H. awardee develops a contract that spans three semesters. Compensation models generally involve split costs between employers, Iowa AEYC, and the awardee but many combinations of support mechanisms exist.

Timeline and Historical Development:

Although connected in scope, the T.E.A.C.H. and WAGE$ systems are separate data streams. Participants do overlap (approximately 26% in FY 2022) but all data are housed in distinct database environments. T.E.A.C.H. began in 2003 in Iowa and has undergone multiple iterations (5 nationally and 3 in-state). Although the system has evolved, only small changes have occurred, and core data collection has remained reasonably consistent. The program sees approximately 400 participants each year, with total cumulative data for approximately 2,000 – 2,800 individuals.

The Data Collection Process:

How are the data collected?

Data collection begins at the application process and continues through each tuition payment and reimbursement issued throughout the awardees contract.

Who collects data from whom?

Iowa AEYC staff collect, enter, and manage all data collected from child care and early education workers who participate in the program.

When and how are data entered into a database?

Initial data are entered at application. All subsequent data for direct tuition payments are entered each semester. Data related to reimbursement payments to the awardee are entered yearly after award settlement at the end of each calendar year.

Are data verified at entry?

Demographic data elements are required at time of entry into the system through application via an online portal. Other data fields that are not required are monitored for missing entry by quality control staff who attempt to update or fill missing fields.

Known fidelity or quality issues?

The system does not currently use confirmation protocols to check/verify applicant demographic elements. In addition, user errors over multiple applications can result in possible duplication of records; quality control staff currently attempt to catch and clean such occurrences. Finally, staff indicated the potential for record duplication resulting from Head Start license numbers.

How are data updated in the system over time?

The T.E.A.C.H. system is cumulative single-record database that allows record updates. Updated records are addended in the system so that original/previous records and data are not deleted or overwritten.

Changes to Data Collection Over Time:

Although the T.E.A.C.H. system has continued to collect core data elements consistently across national and state iterations, data collection has expanded to include in-depth information about 1st generation students and information related to equity and diversity starting in 2014.

How are the Data Currently Used?

Internal uses:

T.E.A.C.H. staff use system data to generate bi-annual and annual reports. One focus of internal reporting efforts focuses on workforce turnover rates. Recent data indicates comparatively lower turnover rates of 5% among T.E.A.C.H. participants.

What Types of Identifiers are Collected?

Basic demographic elements including name, DOB, current education, gender, and race/ethnicity are collected for each provider awardee. In addition, awardees provide license numbers and SSN information. Other elements collected that could be used as confirming or disconfirming information include educational background, incoming credentials at enrollment, and household/family size.

System Strengths and Limitations

A strength of the system is that awardees are required to provide their SSN and the system does not allow duplicated SSN entries. This internal check ensures uniqueness of the awardee record that is trackable across multiple applications and award cycles.

The system allows data pulls by specific dates, date ranges, or by contract periods (3 semesters). This functionality provides flexibility in coverage the ability to easily track record updates made by users over time. One unique strength of the T.E.A.C.H. system is that it collects and maintains maiden names, where applicable. The maintenance of maiden names is exceptionally useful for longitudinal tracking and cross system linkages, particularly with systems that do not maintain name changes but are likely to span substantially different periods of time.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

Iowa AEYC staff expressed interest in continued effort to identify where workforce turnover is occurring and possible explanations for variability in turnover rates across the state. Partnership with an IDS could provide resources to investigate geographical and transhistorical trends in workforce turnover as well as connection with external data sources that might speak directly to correlates of workforce variability. Iowa AEYC staff also indicated that they have not yet directly integrated and compared WAGE$ and T.E.A.C.H. as verification efforts. Although overlap estimates between the two related programs are available, systematic linkage between the two systems to examine parallel progression in education and compensation have not occurred. Connection to an IDS would provide resources and expertise in system linkage that could then provide a connected data stream to directly examine the interrelations between the two sister programs. Finally, T.E.A.C.H. staff expressed some frustration over their ability to clearly assess what percentage of
the total workforce their program is reaching. The limiting factor in such a determination hinges on lack of a comprehensive (and accurate) list or registry of all childcare providers and early educators in the state. Partnership with an IDS would present an opportunity to combine current data with that from other registry systems to obtain an unduplicated workforce roster that retains the most recent record for each identified individual.

Other Notes:

Iowa AEYC staff indicated a strong desire for process development of a connection between state registry systems and the T.E.A.C.H. database to produce an automated flagging indicator when any provider registers for employment.

APPENDIX D.4. CHILD CARE WAGE$® (IOWA AEYC)

Meeting Date

March 9, 2022

Attendees

Heather Rouse, Cass Dorius, Todd Abraham, Ji Young Choi, Carla Peterson, Jessica Bruning, Laura Betancur Cortés

Interviewees

Ashley Otte, Director of Workforce Initiatives, Iowa AEYC

General Description of the System:

In tandem with the T.E.A.C.H. Early Childhood® Program (see above), the Child Care WAGE$® Iowa program is intended to address compensation, retention, and ongoing education of the child care workforce in Iowa. WAGE$ provides education-based salary supplements, or stipends, to low-paid early care and education providers working with children ages birth to five in regulated settings. WAGE$ supplements are tied to education level and can be increased by attaining higher levels of education, with intended educational
attainment support available through T.E.A.C.H.

Timeline and Historical Development:

WAGE$ started in 2014 as a pilot project that is now entering its eighth fiscal year. Initially, the participants enrolled from 33 of the 38 counties where the program was available. In January 2021, the Governor’s office directed funding to the program to support expansion to all 99 Iowa Counties. The program currently supports participants in 84 counties, as of FY 2022. As a result of increased funding and expansion efforts, the program was able to award 850 stipends in comparison to only 275 in the previous fiscal year. Continued funding through at least FY 2024 looks very promising. The WAGE$ system currently maintains approximately 1,100 – 1,300 active records (as of 3/2022) that is likely closer to 1000 records once back-payment reconciliations are completed.

The Data Collection Process:

How are the data collected?

Data collection begins with the stipend application process. To be eligible, applicants must be employed at a provider service that is QRS rated and that accepts CCA eligible children. Once enrolled, eligibility and employment data are collected with reevaluation every 6 months.

Who collects data from whom?

Iowa AEYC collects, enters, and manages all data collected from child care and early education workers.

When and how are data entered into a database?

Initial data about the individual are collected at application. Employment status and stipend eligibility are collected initially and then verified every 6 months. If a stipend recipient changes employers, a minimum of 6 months in the new position is required to reestablish stipend eligibility.

Are data verified at entry?

Known fidelity or quality issues?

How are data updated in the system over time?

The WAGE$ system is a cumulative single-record database that allows record updates. Updated records are added to the system so that original/previous records and data are not deleted or overwritten.

Changes to Data Collection Over Time:

Although the WAGE$ program has expanded over time, the data structure and types of information collected have not changed from the original version of the system.

How are the Data Currently Used?

Internal uses:

WAGE$ staff use system data to generate bi-annual and annual reports. As with T.E.A.C.H., one component of internal reporting efforts focuses on workforce turnover rates. In comparison to the overall workforce, recent data indicates comparatively lower turnover rates of 18% among WAGE$ recipients.

What Types of Identifiers are Collected?

Basic demographic elements including name, DOB, current education, gender, and race/ethnicity are collected for each provider awardee. In addition, awardees provide SSN and licensure information. Other elements collected that could be used as confirming or disconfirming information include educational background, incoming credentials at enrollment, and official transcripts. Due to program requirements, the WAGE$ system also includes both current employers and any former employers for recipients who have received stipends previously.

System Strengths and Limitations

As with the T.E.A.C.H. system, WAGE$ does not allow duplicated SSN entries. Data can also be pulled conditionally by dates and ranges affording similar benefits in terms of coverage and trackability. Finally, WAGE$ affords the same unique contribution to linkage and tracking efforts through collection and maintenance of maiden names, where applicable. One potential limitation in the WAGE$ system is the requirement of a 6-month employment period for eligibility. In such cases where a provider or early educator changes employers, the system data would demonstrate a 6-month gap that could mistakenly suggest that the individual has left the workforce. Such gaps, though potentially not overwhelmingly common, could lead to an underestimation of the workforce at point-in-time assessments or over relatively short periods.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

Given the shared goals of the WAGE$ and T.E.A.C.H. systems to improve education, compensation, and retention in the childcare provider and early educator workforce, the potential benefits or partnering with an IDS for the WAGE$ system exactly mirror those describe in relation to T.E.A.C.H. (see above). A particular benefit for both systems is the possibility of direct linkage and evaluation of how the two programs work together, and independently, to meet shared goals. Both programs also would gain valuable insights about program outcomes and further potential expansion with information specific to who both programs are reaching. The ability to compare program enrollments with a known definite workforce registry would address those questions.

APPENDIX D.5. UNEMPLOYMENT, WAGES, AND WORKFORCE NEEDS (IWD)

Meeting Date

March 24, 2022

Attendees

Heather Rouse, Todd Abraham, Carla Peterson, Laura Betancur Cortés

Interviewees

Donna Burkett, Bureau Chief Iowa Workforce Development
Wendy Greenman, Bureau Chief overseeing WIOA
Michaela Malloy Rotert, Executive Officer and Workforce Program Coordinator
Ryan Murphy, Director Labor Market Information Division

General Description of the System:

WIOA

The Workforce Innovation and Opportunity Act was established in 2014 under the U.S. Department of Labor and is implemented statewide by Iowa Workforce Development (IWD). IWD tracks participant level data for all individuals enrolled who received services from the agency or its partners. The WIOA is intended to link job seekers to occupational training, education and other support services that will eventually connect to employment opportunities and success in the labor market. In addition, the program aims to connect employers to the skilled workers they need.

The system currently contains weekly unemployment claims/amounts, and data related to employment after exit from WIOA program (2nd and 4th quarter median earnings), credential attainment, measurable skills gained, employment history, and employment goals. Barriers to employment are also tracked, including, displaced homemaker status, low-income status, disability status, ex-offender status, homelessness, foster status, English Language Learner status, low literacy, cultural barriers, migrant and seasonal farm worker status, within two years of exhausting TANF eligibility, single parent status, and long-term unemployment. Each unemployment claim requires a single application that includes linkage elements to connect multiple applications or other data sources.WIOA serves approximately 100,000 individuals a year (about 85,000 – 88,000 in 2022).

Labor Market Information Division

The Labor Market Information Division (LMI) was established by Iowa Workforce Development in the 1970s and has been providing labor market information to businesses, job seekers, and policymakers in the state ever since. Iowa Workforce Development, through the Labor Market Information Division (LMI), also maintains quarterly wage earnings for most employees in Iowa. Wage data are identified by social security numbers and include industry codes. Specific occupation within industry is not recorded. In addition, wage data are not collected from federal, student or religious employees and also not reported by those who are self-employed.

Laborshed Survey

In addition to the data sources above, IWD also collects survey data. The Laborshed survey is an anonymous response survey that assesses employment status, wages, benefits, industry, job searches, commute, happiness, and likelihood to switch employment. Area identification includes zip codes. The survey has been administered for 20-21 years, with only slight changes over time. Approximately 10,000 -12,000 individuals respond to the survey each year.

Workforce Needs Assessment Survey

Employers in Iowa are also surveyed annually to assess workforce needs. Specific content includes vacancies, retirements, challenges hiring, applicants, qualifications, and benefits. The employer survey is distributed to all employers on record, with five or more employees (approximately 45,000) and typically sees a 20%-25% response rate (9,000-11,000 respondents) each year.

Occupational Employment and Wage Statistics (OEWS) Survey

The Occupational Employment and Wages Statistics (OEWS) survey is distributed to two panels of employers (approximately 7,500 total) each year. Employer respondents provide the number of current employees in their organization/business, what roles/work their employees perform, and how much employees earn, both hourly and annually. Typical response rates have been good (70%-75%), though it is becoming more difficult to hit those rates in recent years. Data for worker titles from the OEWS are coded to occupation codes. Wage data from the OEWS are used to estimate hourly wages for the Iowa Wage Report.

Timeline and Historical Development:

The Occupational Employment Wages Statistics (OEWS) survey started collecting occupational wage data in 1996.

Since the LMI program was started by Iowa Workforce Development in the 1970s, it has given companies, job seekers, and state politicians in the state access to labor market data.

The Data Collection Process:

How are the data collected?

WIOA unemployment claims are submitted by individual workers. Wage data and specific survey data are submitted by employers.

When and how are data entered into a database?

Unemployment claims are reported and entered weekly. Wage data are collected quarterly. Survey data are collected annually.

Are data verified at entry?

Unemployment insurance wage records are manually checked by IWD staff and corrected where necessary. Business closures are verified by report of closure date to unemployment insurance. In some cases, businesses that are assumed closed due to no reported employees and no reported wages are investigated and verified as closed by IWD staff.

Known fidelity or quality issues?

Job posting boards are not perfectly deduplicated and could contain multiple postings for the same position. Accuracy is also impacted, though the extent is not fully known, by closed ads or reposted/reopened ads.

How are data updated in the system over time?

Changes to data records are tracked internally with change codes and change dates. Current records are identifiable, but previous records are also maintained allowing backtracking if necessary.

Changes to Data Collection Over Time:

The federal side of system started in the 1980s but transitioned to a new database in 2022. All data from the old system was archived and remains available for use.

How are the Data Currently Used?

Internal uses:

All data are used internally for periodic reporting, including annual reports that are made publicly available online. Various survey data are also used, in combination with data from other states, to update and define new occupational codes. Finally, survey data from employers are used to produce labor market forecast reports.

External users:

IWD maintains a job openings online dashboard for work seekers. In addition, the agency operates an industry projections data tool for public use with some filter capabilities. IWD has also entered into data sharing agreements with Iowa Department of Transportation to examine industry staffing patterns, the Iowa Department of Education to use student and wage records for program evaluation purposes, and other entities to examine U.S. Department of Labor program outcomes.

What Types of Identifiers are Collected?

Depending on the specific source of data and level of analysis, the IWD data streams include person-level identifiers such as name, DOB, race/ethnicity, gender, and SSN. The Laborshed survey also collects phone number and zip code. Employers also provide employee addresses. Employer-level data includes business ID numbers (EIN), location identifiers (parent and satellite locations of the same company), and industry codes.

System Strengths and Limitations

An obvious strength of the IWD data system is that much of the information is tracked with unique identifiers (SSN) and that wage data is reported directly by employers in real quarter-time. (Not self- or retroactively reported). Although the data only track wage earnings or unemployment claims in the state of Iowa, individuals are tracked over time, allowing for examination of work/unemployment trajectories, occupational changes, and career progression (as indexed by occupation and earned wages).

A second strength of the IWD data lies in the tracking of occupational codes and wage earnings for all instances each quarter for a given individual. Instead of collecting only the principle source of income or primary occupation, the IWD includes earnings for any paid work in any industry during a particular quarter period.

One potentially severe limitation, with regard to child care providers is the absence of wage earnings data for those who are self-employed. Wages earned by a home care provider running an independent business would not be reflected in the IWD data system, unless that provider was drawing a salary or had registered the business with unemployment insurance. Similarly, the specific definitions for student employment and religious employment could impact the availability of wage earnings data if individuals are working in the child care field under either condition (e.g., a student worker at a care center, a teacher at a religious-based preschool). While not a limitation of the IWD data specifically, a connected challenge is the inability to track individuals who work across the Iowa border in other states. This challenge is not unique to workforce participation. Many state data systems currently do not track use of other governmental, health and social services being accessed across state lines.

A second limitation of the data involves industry code information. Although industry codes are standardized and well-defined, they can be broad in some areas where multiple different specific occupations fall under the same umbrella industry code. For example, an individual working in a child care center, as identified by an industry code, could be working as a care provider, in an administrative role, in a support role, etc. This level of specificity is not available, as the industry code is tied to the employer, not the occupation.

Similarly, while wage data are reported cumulatively over quarters, the only time metric available is the quarter year split. As a result, assumptions are required to present wage earnings in expected annual salary (assuming quarter wages * 4), or in monthly earnings (assuming stable earnings and dividing quarter wages by 3). The data collected do not index whether employees are working full-time or fractional time, whether employees are temporary or seasonal, or whether employees are paid salary wages for a predetermined work period that differs from the traditional 40 hours/week model.

A final potential limitation is the ease with which businesses can be tracked over time. While business ID or Employer Identification Number could be used where available, tracking by business name presents potential challenges. In some cases, the actual business of interest is owned by a parent company (e.g., multi-site care providers) or uses a corporate/legal business name. In these cases, the business of interest to track would be identified by the parent company name. However, in the special case of multi-site satellite businesses owned by the same parent company, geographic identifiers are available to delineate each satellite location as a separate business entity.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

IWD data could be used to assess turnover and/or occupational volatility through examination of individual child care workers over time. Identifying those who leave a particular employer but remain employed in the field could shed light on who is moving within the industry and where the industry is stable/unstable.

Similarly, an IDS is well suited to examine employees who leave child care employment and resume employment in a different industry. Although this could be accomplished to some degree only using industry codes and employer information, the addition of wage data would speak directly to whether child care providers are leaving the field for higher earnings, and perhaps more importantly, identify patterns that indicate what specific industries are pulling providers away from child care.

Within-system examination of upskilling and whether participation results in higher wages is certainly possible using only IWD data. However, connecting IWD data to other systems like T.E.A.C.H. provides an opportunity to evaluate outcomes following the incentivization for child care providers to pursue credentials and extended education. The connection of these two systems would inform questions of whether upskilling/credentialling materializes in higher wages (how long does the increase take), whether higher wages following upskilling/educational progress keep providers in the field (in the same positions), and whether upskilling/education increases opportunities to transition industries thereby indirectly leading to child care shortages, particularly in rural markets.

Other Notes:

Although business closures are tracked (see above), the reasons for closure are rather broad indexing mergers, consolidations, and sale, but no other specific reasons for closure (e.g., funding, market, etc.) are recorded.

APPENDIX D.6. IOWA CHILD CARE RESOURCE & REFERRAL (CCR&R)

Department/Agency

Iowa Department of Health and Human Services (through Iowa Child Care Resource & Referral; CCR&R)

Meeting Date

April 21, 2022

Attendees

Heather Rouse, Todd Abraham, Jessica Bruning, Ji-Young Choi, Cass Dorius

Interviewees

Cassie Reuter, Program Specialist Region 1 (NW),
Emily Lamar, Regional Data Specialist Region 2 (NE),
Linda Heckman, Data Specialist Region 3 (SW),
Tiffany Ichelson, Regional Data Specialist Region 4 (Central),
Tami Holms, Program Services Supervisor Region 5 (SE),
Alisha Wiese, Data Analyst Region 5 (SE),
Becky White, Regional Director Region 5 (SE)

General Description of the System:

CCR&R supports quality child care in Iowa through on-site and virtual consultation with licensed preschools, Child Care Centers, nonregistered Child Care Home providers, and registered Development Home providers under regulatory authority of HHS. CCR&R staff assist providers with state regulation compliance and improving the quality of child care they provide. In addition, CCR&R maintains data and information on child care providers to assist families in locating and connecting with child care services within the community.

Timeline and Historical Development:

CCR&R started in specific areas of Iowa in 1989 and expanded statewide in 1992, with the mission of serving communities, child care providers, and parents. The agency split into five regional offices in 1997 due to welfare reform impacts on child care. The first database system, NACCRRAware (National Association of Child Care Resource and Referral Agencies – now Child Care Aware America), was established in 1998. In 2011, the Iowa Department of Health and Human Services implemented a new statewide initiative that required all Child Care Assistance (CCA) applications and Child Development Home applications be submitted through CCR&R. In 2019, all parent referrals provided by CCR&R were centralized in Region 5. In February 2022, the existing NACCRRAware data system was migrated and the current NDS 2.0 (National Data System developed and maintained by WorkLife Systems) database was released.

The current system consolidated separate, but very similar, databases across the five regions. Before NDS 2.0 transition, each region used two separate systems to track data. The change to NDS allowed consolidation of 10 regional databases. All historic data were fully migrated to the new system and remain available for use.

The Data Collection Process:

How are the data collected?

Data are collected from providers electronically through an online portal or via phone and then entered manually by CCR&R data specialists.

Who collects data from whom?

All registered and licensed provider information comes from the HHS (KinderTrack) provider registry. Preschool information is obtained from the Iowa Department of Education. Additional data are collected from nonregistered homes and unlicensed providers via word of mouth and regional specialists’ knowledge of the local care providers.

When and how are data entered into a database?

Data entry begins with a request from a community, from a provider through a consultant, or directly through HHS at licensure. Once initiated, a CCR&R data specialist creates the user profile and enters relevant minimum data. Once the profile is created, providers can access, update, and complete their information directly.

Providers can update their profile data directly in the NDS 2.0 system using the Provider Information Form (PIF). NDS uses an automated request for update to the PIF that is sent out at 90-day intervals (quarterly). Yearly requests are also sent to providers to update information. In addition to the automated PIF request, staff make direct contacts with providers to obtain information updates.

Programs that become inactive are retained in the system for five years. HHS verification that a program/provider is inactive is required for inactive status.

Monthly updates and new licensures from HHS (KinderTrack) are received as a .csv file by each of the five regional directors who manually enter any new data and update existing data as needed.

Parent searches can be conducted via phone, email, or directly through the referral website. Searchers can create an account that saves all relevant information and search history, or searches can be conducted under a general guest account. Guest account searches do not save specifics about the searcher but do record specific parameters of the search (e.g., location, age group, type of provider, etc.).

Are data verified at entry?

Cross validation of provider information obtained monthly from HHS is conducted manually. Updates by providers made directly to the NDS system are reviewed and approved by CCR&R staff. Inactivity status is validated by both CCR&R consultants and HHS list updates.

Known fidelity or quality issues?

In the previous system, data specialists across regions worked closely to develop data standards and definitions that were consistent for all regions. This consistency facilitated the transition to the new system leading to high fidelity of data elements across sites within the full system.

Providers are not tracked directly across facilities if they move. License numbers may stay the same for a home, but a new record ID will be produced. While detecting duplication and entity tracking remain possible, this process adds potential complication to those processes. Centers receive new numbers each time they enroll in the system. Programs with multiple sites do not receive a program level identifier that is attached to each site but linkage by program name or director could be possible. Region directors should be able to identify such clusters on a post hoc basis.

Openings for referral are not updated routinely and consistently, raising some concerns about variability and accuracy. In addition, response rates to automated PIF update requests are not necessarily known. Staff also recognize that nonresponse to the PIF cannot be confirmed as provider noncompliance or simply providers having nothing to update.

In the event that updated data are received independently by CCR&R consultants or staff but those data differ from monthly HHS KT update information, decisions about which source is correct are not standardized.

How are data updated in the system over time?

When data are changed or updated, change log entries are produced and can be viewed. Total number of historical updates to provider profiles is maintained in the NDS 2.0 system.

How are the Data Currently Used?

Internal uses:

Current reports (data sheets) are generated yearly in July. Current implementation of the NDS 2.0 system is moving toward point-in-time capacity to enable data sheet reporting capabilities and generation at any interval based on data currently in the system at that time. In addition, CCR&R staff generate yearly county data sheets that include county level child care data from the database. Summary information in each yearly report have also been used to produce 5- and 10-year trend reports.

External users:

Primary external use of CCR&R data involves families requesting referrals to child care providers. Those searching for provider services can contact CCR&R staff directly or perform their own searches of the provider database using the public search portal. Communities can also access summary data to assess current child care availability in their areas or to determine where expanded availability would be helpful. Finally, people working with providers, centers, and homes use information from the CCR&R system to determine average rates to charge for child care services.

What Types of Identifiers are Collected?

The CCR&R system collects and maintains provider-level identifiers including name, address, zip code, license ID, and current certificate, degree, and education credential.

System Strengths and Limitations

A clear strength of the CCR&R system is that data entry, maintenance, and management has a long history of consistency across regions. The effort to develop such consistency has not only translated to the current system data but also produced archival data before consolidation into NDS 2.0 that is still usable and comparable across regions.

The primary limiting issue with the CCR&R data centers on update frequency and provider compliance with update requests. Given uncertainty about whether nonresponse indicates noncompliance or simply lack of new information to report, nonresponse raises concerns about how up-to-date or accurate some of the information (e.g., vacancies, active providers, etc.) is within the system.

Because new records are produced even when a license number remains constant, and because multi-site centers are not directly identified as the same program, it is not clear how much deduplication and within-system linkage might be required for particular uses of the data.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

To ease staff burden and help redirect resources, an IDS could perform the monthly reconciliation between current data and updates provided by HHS KT. This would require monthly data uploads to an IDS system. Alternatively, like with I-PoWeR (see above), an IDS could use existing data to develop a solution to automate reconciliation and translate that solution into an accessible software environment for use directly by CCR&R staff. CCR&R is unique in that the system maintains provider availability and family need collected through the search portal and referral service. The combination of those data streams leads to multiple possible investigations into connections between care availability, demand for care services, and geographic alignment. For example, CCR&R provider information could be used to identify geographic locations where there simply is no available space (i.e., care deserts).

Similarly, use of provider and search data could be used in a density-disparity analysis aimed at identifying geographic areas where searches for care greatly exceed available care. Because CCR&R also tracks the results of referrals at the surface level, including which provider was chosen, these disparity analyses could be tied to the eventual outcome of referral (i.e., does the family eventually find a spot). From a supply standpoint, saved parameters of referral searches could identify specific features of the search that are connected to higher levels of disparity. Such findings could inform decisions by communities, existing providers, and potential new providers regarding what specific options for care are in most need.

Combining the elements of space availability, geographical area, and time allows for an examination to determine constant unavailability vs. sporadic unavailability vs. growing availability. The ability to identify volatility in available care leads to opportunities to examine outcomes of instability in terms of whether it reflects high turnover across employers, transition of providers out of the child care workforce, or geographic relocation of providers/businesses for other reasons. Indexing volatility in terms of active/inactive/active/inactive patterns of care availability also could connect directly to family (barriers to stable employment) and child outcomes (K readiness). Finally, volatility could be examined for potential impacts on providers that might connect occupational instability to barriers to career progression, corresponding wage increases, and as an eventual reason for leaving the child care and early education field.

APPENDIX D.7. KINDERTRACK CHILD CARE PROVIDER LICENSING REGISTRY

Department/Agency

Iowa Department of Health and Human Services

Meeting Date

February 16, 2018; June 2022; January – March 2023

Attendees

Heather Rouse, Cassandra Dorius, Maya Bartel, Allison Gress, Todd Abraham

Interviewees

Tammi Christ, Mark Adams, Ryan Page

General Description of the System:

The Child Care Assistance subsidy program (CCA) in Iowa is operated through the KinderTrack (KT) system under authority of HHS. The system stores information about registered or licensed childcare providers in the state, as well as families that receive funding for child care through the Child Care and Development Block Grant (CCDBG) mechanism. The KinderTrack system includes both a portal interface where data are entered and a separate data warehouse where data are stored. The KT system was developed and phased in during fiscal year 2010, with complete transition to the KT system beginning in fiscal year 2011.

Provider information includes data on all registered or licensed childcare providers, including their capacity, hours of operation, rates charged, contact information, and household composition (if home-based). Child and family data includes information for families applying for CCA, approvals and renewals, child care provider schedules, child attendance, and subsidy payments.

The Data Collection Process:

How are the data collected?

The KinderTrack system houses information for the state of Iowa’s child care provider registry, CCA subsidy applications and renewals, attendance data for children receiving the subsidy, and data related to payments and processing for subsidized care provided.

When and how are data entered into a database?

Data in the KT system reflects event data, meaning that there are multiple entries for each child reflecting family applications for assistance, determination of eligibility, selections of childcare, attendance in programs, subsidy payments for care, and re-authorization of eligibility. CCA family eligibility workers enter and process a family’s application through KinderTrack. Iowa Workforce Development staff can input information into KinderTrack, using the HHS state ID.

Are data verified at entry?

KT does apply verification rules to daily data related to attendance and payment determinations. Provider data are verified as part of the licensing process.

Known fidelity or quality issues?

System migration before 2010 required double verification efforts that were not entirely successful. Data in the system before 2010 are considered to be lower quality and less suitable for use.

How are data updated in the system over time?

Data entry to the system occurs daily (family applications, provider applications) and at specified service period intervals (attendance, payments). Any new or modified data entered to KT creates a delta record that passes to the data warehouse each day. Delta records received by the data warehouse create a new active record. The previous record is time/date stamped and becomes inactive. The KT system is cumulative in that inactive records are maintained with the current active records. Importantly, inactive records are purged from the system periodically.

How are the Data Currently Used?

Internal uses:

KinderTrack data is used for required federal reporting related to the Child Care Assistance subsidy program and fraud detection by families or providers. Data are also used between bureaus and departments for quality control purposes and internal projects.

External users:

KinderTrack data have been shared externally for research projects or grants where access is granted to parties who require broad information such as aggregated summary reports of frequencies or compiled lists of child care providers. Data are also shared across Iowa agencies for various purposes including program integrity assessments, creation of legislative service reports, and other Department of Management priorities. KinderTrack also feeds a publicly available child care provider search portal that allows individuals to tailor search criteria and obtain lists of providers and contact information that fit search needs.

What Types of Identifiers are Collected?

Each child, parent, and family receives a unique static ID at the time of application, which is used to link data tables in the system and track participants over time. However, new IDs are assigned in the event of new applications, changes in family structure, and/or lapses in services. Parent and child identifiers including names, DOB, gender, race/ethnicity, and SSN (in many cases) are collected. Family identifiers include member relationships and geographic elements. Provider identifiers include provider name, DOB, location, and license information.

System Strengths and Limitations

The primary strength of the KT data system is that it serves as the official child care provider registry for the state of Iowa. They system also includes expansive information for providers and families engaged with the Child Care Assistance subsidy program. Finally, the system structure designed to maintain inactive records allows for the possibility of tracking entities as demographic information changes over time across other systems. However, the KT system import of existing CCA data and the process of assigning new family, parent, and child IDs under specific circumstances does produce some complication with regard to entity deduplication.

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

Recent reports indicate that CCA has seen a 65% increase in denials and a 40% reduction in applications from FY19 to FY22. These changes suggest increases in income/wage earnings that put many families above the poverty cutoffs to either be eligible for CCA (denials) or leave families thinking they are not eligible for CCA (not applying). This raises an obvious question about whether families that were eligible for CCA before are simply no longer eligible due to income increases. Partnering with an IDS to connect CCA applicants over time with recent IWD wage data could shed light on multiple concerns. First, if denials are resulting from income requirements, are denied families just above the income/wage threshold (i.e., still in real need of assistance but just barely above the cutoff to receive it)? Second, if incomes are increasing for families that need child care, are incomes increasing for providers as well? Third, if providers’ income is increasing, is cost of care increasing and if so, are those cost increases being passed on to families? Finally, if cost of care is increasing and families are seeing income increases that put them just barely beyond the eligibility cutoffs for CCA subsidy eligibility, is the net gain in wage earnings eliminated completely by the associated increase in higher costs of child care?

ACKNOWLEDGEMENTS

We gratefully acknowledge the help and support of the following individuals:

Advisory Committee:

Ashley Otte, Iowa Association for the Education of Young Children (Iowa AEYC)

Ryan Page, Iowa Department of Health and Human Services (DHS)

Erin Clancy, Iowa Department of Health and Human Services (DHS)

Amanda Winslow, Iowa Department of Management (DOM); Early Childhood Iowa (ECI)

Wendy Greenman, Iowa Workforce Development (IWD)

Cassie Reuter, Mid-Sioux Opportunity, Inc.; Child Care Resource & Referral (CCR&R

Data Discovery Participants:

Erin Clancy, I-PoWeR (Iowa Department of Health and Human Services)

Ashley Otte, T.E.A.C.H. and WAGE$ (Iowa AEYC)

Donna Burkett, Bureau of Labor Statistics (Iowa Workforce Development)

Wendy Greenman, WIOA (Iowa Workforce Development)

Ryan Murphy, Labor Market Information (Iowa Workforce Development)

Michaela Malloy Rotert (Iowa Workforce Development)

Cassie Reuter, Iowa Child Care Resource & Referral (CCR&R)

Emily Lamar, Iowa Child Care Resource & Referral (CCR&R)

Linda Heckman, Iowa Child Care Resource & Referral (CCR&R)

Tiffany Ichelson, Iowa Child Care Resource & Referral (CCR&R)

Tami Holms, Iowa Child Care Resource & Referral (CCR&R)

Alisha Wiese, Iowa Child Care Resource & Referral (CCR&R)

Becky White, Iowa Child Care Resource & Referral (CCR&R)

Tammi Christ, KinderTrack/Child Care Assistance (Iowa Department of Health and Human Services)

Ryan Page, KinderTrack/Child Care Assistance (Iowa Department of Health and Human Services)

2023 Iowa Child Care Workforce Study

Appendix D

Data Systems Inventory and Recommendations for Use

TABLE OF CONTENTS

INTRODUCTION

APPROACH

TABLE 1. RELEVANT DATA SYSTEMS IDENTIFIED AND REVIEWED

SUMMARY OF FINDINGS

RECOMMENDATIONS

Use Case 1: Compiling a comprehensive deduplicated list of child care providers in Iowa

Use Case 2: Identifying workforce decline in relation to child care need

Use Case 3: Examining and quantifying the impact of participation in T.E.A.C.H. Early Childhood® Iowa (Iowa AEYC)

Use Case 4: Identifying and describing child care providers who leave the field

CONCLUSION: WHAT WOULD IT TAKE TO ADVANCE AN INTEGRATEDSOLUTION FOR PROVIDER DATA

APPENDIX D.1. DATA INVENTORY PROTOCOL

A1. General Description of the System

A2. System Timeline

A3. The Data Collection Process

A4. Changes to Data Collection Over Time

A5. How are the Data Currently Used?

A6. Types of Identifiers Collected?

A7. Strengths and Limitations of the Data

A8. What Benefit Could an IDS Provide for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

APPENDIX D.2. I-POWER: IOWA’S EARLY CHILDHOOD AND SCHOOL AGE PROFESSIONAL WORKFORCE REGISTRY (HHS)

General Description of the System:

Timeline and Historical Development:

The Data Collection Process:

How are the data collected?

Who collects data from whom?

When and how are data entered into a database?

Are data verified at entry?

Known fidelity or quality issues?

How are data updated in the system over time?

Changes to Data Collection Over Time:

How are the Data Currently Used?

Internal uses:

External users:

What Types of Identifiers are Collected?

System Strengths and Limitations

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

Other Notes:

APPENDIX D.3. T.E.A.C.H. EARLY CHILDHOOD® (IOWA AEYC)

General Description of the System:

Timeline and Historical Development:

The Data Collection Process:

How are the data collected?

Who collects data from whom?

When and how are data entered into a database?

Are data verified at entry?

Known fidelity or quality issues?

How are data updated in the system over time?

Changes to Data Collection Over Time:

How are the Data Currently Used?

Internal uses:

What Types of Identifiers are Collected?

System Strengths and Limitations

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

Other Notes:

APPENDIX D.4. CHILD CARE WAGE$® (IOWA AEYC)

General Description of the System:

Timeline and Historical Development:

The Data Collection Process:

How are the data collected?

Who collects data from whom?

When and how are data entered into a database?

Are data verified at entry?

Known fidelity or quality issues?

How are data updated in the system over time?

Changes to Data Collection Over Time:

How are the Data Currently Used?

Internal uses:

What Types of Identifiers are Collected?

System Strengths and Limitations

What Benefit Could an IDS Have for Ongoing or Future Envisioned Work in Your Department or With Other Partners You Work With?

APPENDIX D.5. UNEMPLOYMENT, WAGES, AND WORKFORCE NEEDS (IWD)

General Description of the System:

WIOA

Labor Market Information Division

Laborshed Survey

Workforce Needs Assessment Survey

Data Systems Inventory
and Recommendations for Use

CONCLUSION: WHAT WOULD IT TAKE TO ADVANCE AN INTEGRATED
SOLUTION FOR PROVIDER DATA