Hawaii Growth Model Talking Points

Directing and FocusingGrowth Conversations

04/19/20232

Position and Messaging•What questions are important to

answer?

•What values and information are important to communicate?

04/19/20233

A. Using Growth in Accountability SystemsWe use growth in our accountability systems because it provides us with an additional perspective of how well we are serving our students. School median student growth percentiles are used to inform school Strive HI Index scores, CESSA principal evaluations, and EES teacher evaluations for school teachers that can’t have individual medians calculated due to their job assignment. When possible, EES teacher evaluations incorporate weighted median student growth percentiles calculated from student rosters that teachers help verify.Achievement or proficiency information is important because it helps us evaluate how students are performing relative to the standards that we know are linked with college and career success. This is why achievement will always play an important role in how we evaluate our schools and is almost always presented alongside growth metrics. Adding growth measures to our accountability systems deepens our understanding by helping identify which schools and educators are moving students forward the fastest given each student’s prior performance.We use growth when evaluating teachers because we know that student achievement levels are heavily impacted by prior achievement levels. Relying on growth in teacher evaluations helps acknowledge teacher contributions while ensuring we don’t penalize teachers for serving struggling students.Our growth model will be a strong asset as we transition to the Common Core State Standards and the corresponding Smarter Balance Assessment Consortium tests. While new assessments or proficiency standards could cause reported proficiency levels to fluctuate during transitions, the effectiveness and interpretation of our growth measures will not change. Student growth scores will be helpful during this transition as they'll help us measure how well students are performing on the new assessments compared to peers with similar past achievement.

What accountability systems use growth?

Why achievement and growth are both essential for school accountability?

Why growth only for teachers?

Why does growth help given the content standards and test transition?

04/19/20234

B. Growth Model Comparison GroupsThe Hawaii Growth Model is rooted in the idea that growth is best understood as a comparison between similar students. This is because the interpretation of any growth metric requires an understanding of what is normal or expected. Whether measuring physical development, athletic performance, financial gain, or academic achievement, we only know whether an amount of growth is notable or cause for concern by comparing it against a norm. Each year, our growth model analyzes all available HSA data to determine how each student’s new test results compare to other students throughout the state with similar past performance on the HSA. This analysis only relies on prior HSA performance and doesn’t factor in any other demographic information or student enrollment in support programs like those for disabled students or English language learners. By comparing students with similar past performance, we ensure that we account for where each student began the year. By only factoring in previous test scores, we honor the diverse range of ability levels between different students with shared characteristics and prevent the creation of comparison groups that unjustly propagate lower expectations for high-needs students.This grouping and calculation process is rerun every year so that our understanding of what is relatively low or high growth given each student’s starting point is always informed by the latest student performance data available. If you want more information about the math and technical details, you should reference the technical overview available on the EES website.

Why is the meaning of growth rooted in comparison?

What is being compared in our growth model?

Where to get more information?

04/19/20235

C. Interpreting a Median Growth PercentileMedian student growth percentiles, sometimes just called median growth percentiles, are summary measures that aggregate individual student growth percentiles. Medians are simply the score of the middle student or the average of the middle two students when all the scores in a group are sorted from least to greatest. To understand these simply, the median gives the student growth percentile that approximately half the group did better or worse than. For example, in a school with a median student growth percentile of 60, about half of the students had individual student growth percentiles greater than 60 and about half of the students had individual student growth percentiles less than 60. For an individual student, a student growth percentile of 60 signifies that the student scored higher than 60 percent of other students throughout the state with similar prior HSA performance. School growth medians include all students that have been enrolled for the full school year and individual teacher growth medians weigh students based on a roster verification process that teachers complete at the end of each school year. For more information about school medians see the Strive HI section of the DOE website and for more information about teacher evaluation see the teacher EES Manual.

What is a median?

How can medians and SGP be interpreted?

Where to find more information?

04/19/20236

D. Where to Find Important DataThe Department has invested a great deal of resources into ensuring that educators have access to the data they need to make the best decisions for our students. It’s important to note that multiple sources of data are often necessary to help decision makers fully understand a given situation. It is also true that qualitative data educators collect from their direct interaction with students and schools can change the way they interpret the data they see. This is precisely why the Department has made intentional efforts to promote data literacy and access at the school level. While not necessary for every educator to become an expert in every data system or study every dataset, all educators and school leaders should be familiar with what data is available and how to access the data that is most relevant to the decisions they have to make.When trying to find the data you need, it helps to keep the purpose of each system in mind: • The Longitudinal Data System (LDS) is intended to provide educators with operational school data and

historical information about student academic performance. • The Hawaii Growth Visualization Tool created by (SchoolView) and the ARCH Data Center supplement

the customized reports available on the LDS with dynamic visualizations of summative growth and achievement data that can help make information easier to understand and communicate.

• Strive HI reports and other school reports available through ARCH and the Department website help school leaders evaluate how well the school is meeting its summative goals and progressing on various state initiatives.

• PDE3 is a professional development management system that is intended to capture and report information about professional development opportunities, trainings, and performance evaluations.

How should the task of data analysis be approached?

What is the purpose of each major data systems?

04/19/20237

Frequently Asked Questions1. What is an academic peer? How are

academic peer groups created?2. How can we compare test scores when

tests and standards are changing to Smarter Balance (SBAC) and Common Core? How do transitions between different types of tests impact the model?

3. Can high performing students and schools still show growth? Will it be easier for low performing students and schools to show growth?

4. Does the Hawaii Growth Model account for factors that may impact individual student performance? Are differences in student characteristics or the impact of one-time events considered?

5. Does the Hawaii Growth Model account for

different school characteristics? Are differences in school demographics or conditions directly factored in?

6. Is the Hawaii State Assessment (HSA) the only test used to calculate student growth percentiles? What about other assessments used in our school, complex area, or district? What about alternate assessments?

7. How do we account for students who don’t have historical HSA data? What about students who missed HSA administrations or come from other states?

8. How are students that skip or repeat grades handled by the Hawaii Growth Model?

04/19/20238

Frequently Asked Questions1. Does successfully moving from one

proficiency level to a higher one necessarily produce a higher growth percentile than moving up within a proficiency level?

2. Won’t some students always have low growth?

3. Must the assessment used for calculating growth be vertically scaled in order for the results to be accurate?

4. How come we aren’t using beginning and end of year data to show growth? How can growth be measured without a pre-test? Why are we using growth percentiles rather than looking at percentage growth?

5. Which students will receive a student growth percentile (SGP) score? What about non-tested grades and subject areas?

6. Which student growth percentile (SGP)

scores are included in a school, complex, and complex area median student growth percentile (MSGP)?

7. How will data from very small schools and groups be reported? Is there an established minimum group size for creating a median growth percentile (MSGP) for a school, disaggregated group, classroom, etc.?

8. When will SGP data become available and how can it be accessed?

9. Which school-based staff has access to SGP data? What SGP data will teachers have access to?

04/19/20239

Frequently Asked Questions1. Do parents see the growth data for their

child/children?2. What information does the public have

access to?3. What growth data is available at the high

school level?4. What additional information will the

Department provide to teachers, administrators, and other education stakeholders on how growth data is calculated and how to use it effectively? Where can I find this information?

5. If the HSA is administered multiple times in a year, which administration is used for growth calculations?

6. If schools are only having students take the HSA until they meet or exceed proficiency, wouldn’t that impact student growth results?

7. What is growth? What do student growth percentiles (SGP) in the Hawaii Growth Model measure?

8. What is considered low, typical, or high growth?

9. What questions can the Hawaii Growth Model help answer?

10. Does the Hawaii Growth Model tell us why certain students or groups of students are growing more or less than others?

11. How can classroom teachers use growth data?

04/19/202310

Frequently Asked Questions1. What is a median growth percentile (MGP)

or median student growth percentile (MSGP)?

2. Are student growth percentiles a better measure of student performance than HSA proficiency levels?

3. Can the student growth percentile be interpreted the same way regardless of grade level or test content area?

4. How is growth data being used other than as part of a teacher evaluation system? Is growth data going to impact school accountability or AYP?

5. How will growth data be used in teacher performance evaluations? How much weight will student growth percentile (SGP) data have in determining a final rating?

6. When will growth data be used in teacher

performance evaluations?7. How can I help my school get a higher

median growth percentile?8. How can I be sure that I will only be held

responsible for student growth that can fairly be attributed to me?

9. Will tested grades/subject teachers receive extra pay for high growth?

10. Some principals group advanced or challenging groups of students under one teacher. Is there some way to account for this?

04/19/202311

Actual Question/Comment Samples1. Do you know if there are percentile values we can refer to as

we try to project numbers for our current fourth and fifth graders who have some HSA scores already on record?

2. If a teacher is a non-classroom teacher, what SGP will count toward their evaluation? When I read the manual, it says for a non-tested classroom teacher, they will be looking at the school's ELA MSGP. Is that correct?

3. How do we meet expectations with a very large class size like 33 for 4th grade?

4. Using HSA as an example, how do you compare a 350 kid and a 360 kid if the band is +/- 18?

5. SGP is grading on the curve; moving away from the standards-based grading.