Program-Level Assessment and Using Data for Program Improvement

Why Program-Level Assessment Matters

When teachers assess individual children, they gather information about one child's growth. Program-level assessment goes further: it pulls together data from every child in the program to answer a bigger question — how well is this program serving all children? Directors, lead teachers, licensing inspectors, and quality rating reviewers all use aggregated (combined) data to make decisions that affect every family enrolled.

As an early childhood professional, you will be expected to understand these reports, contribute to them, and help translate findings into better classroom practice. This lesson walks you through the skills you need.

What Is Aggregated Child Outcome Data?

Aggregated data combines individual child assessment results — such as scores from an observational tool like Teaching Strategies GOLD or the Desired Results Developmental Profile (DRDP) — across an entire class, age group, or program. Instead of showing one child's score, the report shows, for example, what percentage of three-year-olds met a benchmark in language and literacy by spring.

Common formats include:

• Bar charts showing the percentage of children at each developmental level per domain
• Tables comparing fall and spring results to show growth over the year
• Disaggregated breakdowns by age group, home language, or special need status

Reading a Data Report Carefully

Before drawing conclusions, ask four questions:

1. What does the data show? Read axis labels, legends, and column headers carefully. Note which domains are included (cognitive, language, social-emotional, physical, etc.) and what measurement scale is used.

2. What is the trend? Is there growth from one time point to the next? Are certain domains stronger than others?

3. What is unknown or missing? Data reports rarely tell the whole story. A chart may show that 60 percent of children met a literacy benchmark but give no information about children who were absent frequently, enrolled late, or received special education services. Missing data is a safety signal: never recommend major program changes based on incomplete information.

4. Is the sample size meaningful? A program with eight children enrolled will show very different percentage swings than one with eighty. Small sample sizes make percentages unreliable.

Comparing Data to QRIS Benchmarks

Most states operate a Quality Rating and Improvement System (QRIS) — a tiered framework that sets standards for program quality across areas such as learning environment, staff qualifications, family engagement, and child outcomes. Programs earn a rating (often one to five stars) based on how well they meet each standard.

When you compare a program's outcome data to QRIS benchmarks, you are asking: Does this program's performance meet, exceed, or fall short of the quality level it is claiming or working toward?

To do this comparison:

1. Locate the specific QRIS indicator related to child outcomes in your state's benchmark document.

2. Find the threshold required for each star level (for example, a certain percentage of children demonstrating expected growth).

3. Match the program's actual data to the threshold. Note whether the evidence supports meeting the indicator or not, and write down exactly which data point you used.

Because QRIS documents are updated periodically, always use the current version from your state agency's official website.

Identifying Strengths and Areas for Improvement

A strength is an area where data consistently shows children meeting or exceeding expectations. An area for improvement is where data shows a pattern of children falling below benchmarks — not just one child, but a meaningful portion of the group, across more than one data point if possible.

Be precise. "Children need to do better in language" is too vague. A useful statement sounds like: "In spring, 42 percent of four-year-olds scored at the 'building' level in expressive language, below the program's goal of 65 percent meeting 'meeting' level by end of year."

Writing a Program Improvement Memo

A Program Improvement Memo is a professional document that turns data analysis into action. A strong memo has four parts:

1. Data evidence: State clearly which data point shows the problem and how large the gap is.

2. Proposed change: Name one specific, evidence-based practice change. "Evidence-based" means the strategy is supported by published research or a recognized professional organization (such as NAEYC or the National Center for Pyramid Model Innovations). Cite your source by author or organization and year at minimum.

3. Measurement plan: Describe exactly how you will know whether the change worked. Include what you will measure, how you will collect that information, and a realistic 90-day timeline.

4. Professional tone: Address the memo to the director by role, use clear paragraphs, and avoid casual language.

A Note on Data Privacy and Ethics

Even when working with fictional data in a classroom exercise, practice the habits you will use in real programs. Child assessment data — even aggregated — must be kept confidential and shared only with staff who have a legitimate need. Never leave printed reports unattended, and follow your program's data security policy. These habits protect children and families and are expected by licensing boards.

Connecting to Your Credentials

The CDA credential asks candidates to demonstrate understanding of how to use assessment information to improve program quality. State childcare licensing reviewers may ask programs to show how they use child outcome data as part of compliance monitoring. Learning to read, discuss, and act on program-level data now prepares you for both.