Source: Image light, Public Domain, http://tinyurl.com/p4pfjr7
Welcome, I'm Trisha Fyfe. And in today's lesson we'll cover the lesson entitled, Student Growth Models. As we learn about this topic, we'll work toward several learning objectives. And together, we'll use the following questions to guide our learning in the video lesson.
First, how can we identify student growth when considering teacher evaluation? We'll also discuss the questions, what are different student growth models, and how can we identify which method is the best? As a teacher, student growth is something that is at the forefront of your thoughts at all times. This is, in fact, why you became a teacher. To help students learn and grow.
Student growth can be measured in many different ways. Teacher evaluation models vary from state to state and district to district on how they measure growth of students. Measuring teacher effectiveness requires teacher evaluation models to reflect on the levels of student growth.
The procedures for measuring that student growth are generally defined in teacher evaluation models that each state or district use. This growth can be based on improvement data for students that comes from state assessments, standardized assessments, or local assessments. When a school or district, or even state, is attempting to select a teacher evaluation model, it's essential to research how student growth is measured. This is an important consideration when aligning evaluation models to other elements and beliefs of an organization, like a school, state, or district.
In teacher evaluation models, there are three main methods of measuring student growth. First, student growth percentiles. The next is value added models that control for student backgrounds. And the third are value added models that correlate student growth with school and teacher growth. Let's take a look at each of these methods.
First we'll start with determining student growth using student growth percentiles, or SGPs. Combined student growth percentiles are used in this method to compare student standardized test performance, and standardized test performance of all students that score the same that that student did in the previous year. We'll look at an example in just a moment.
But first, let's look at some important things to understand about student growth percentiles. The data from all students in the group being assessed is included in the SGP of that particular teacher, school, or district. The median SGP, or student growth percentile, is found by the number of years of student assessment data has been used. Student demographics, or the school environment, are not considered here. Only assessment scores are used.
Let's look at an example. An SGP, or student growth percentile, of 50 for a fifth grade student, would indicate that the student performed better than half of the students with that same fourth grade score from the previous year. So for example, if that student scored a standard score of 480 in grade four, and then in grade five they scored a 570, what we would compare is the performance to all other fourth graders who scored a 480 in grade four, and their new score in grade five.
In this example, half of the students who scored a 480 in fourth grade, scored a 570 or lower in the fifth grade. And the other half scored higher than a 570. Let's look at the pros and cons of using this model.
The pros of using student growth percentiles are that the same expectations are used for all different students. This is due to the fact that demographics are not considered here. The cons for this model include the fact that marginalized groups of students might be at a disadvantage here. Schools with higher proportions of students in groups like English language learners students, or students of low socioeconomic status, or special education students, and even students of diverse races, may also be disadvantaged. This is due to the fact that this method does not control for other variables, like environmental factors.
The second method that we'll look at today is determining student growth using value added models that control for student backgrounds. This is a very commonly used evaluation model and part of the reason for this is because it is very simple, consisting of just one step. This method uses a calculation of students' past performance in math and language arts, and factors in student demographic characteristics. Characteristics like student backgrounds, socioeconomic statuses, and races, as well as school environment.
A school average for student test score growth is looked at here. The pros for this model are that student background is controlled. Another pro is that underlying effects of schools and teachers are also considered here, it looks at many different factors. While a common reason for use is the fact that it's one step, this can also be a con of this method. This model often favors schools with high levels of socioeconomic status students, when other subgroup accounts are insufficient or not diverse enough.
The third method that we'll explore here is determining student growth using value added models the correlate student growth with student and teacher growth. Step one of this method looks at student achievement and student in school characteristics, and then it measures the correlation between these two factors.
Step two consists of forming a growth measure for each school. This is accomplished by using assessment data that is adjusted for student and school characteristics. This method is much like the previous approach, using value added models that control for student backgrounds. But instead of one step, it has two. This leaves room for making comparisons between schools and teachers who teach students with like demographic characteristics.
The pros of this method include the fact that the particular school, teacher, and student growth measures, as well as student demographics, are connected and assessed. It takes a deep look at the diverse characteristics of the students, and essentially creates a level playing field. One con of this method is that it may tend to over adjust for the characteristics of the student and the school.
Let's talk for a moment about which model might be best. After much research, the two-step value added approach seems to be the most accurate and reliable interpretation of student growth. One of the reasons this is so, is that student achievement data is presented in two different ways here. The actual results with no adjustment for student or school characteristics is the first way to examine the data, and then the student growth levels are adjusted.
School and student characteristics are taken into account here. This is beneficial in that it helps us to catch important trends. In the actual scores, it can assist us in catching any lowering of expectations for improvement of student achievement. There is some worry, however, that the teacher evaluation ratings may not reflect their true effectiveness in increasing student achievement in all subgroups of students, without the actual pure scores. The selected models tend to rely heavily on student growth percentile.
There's also concern that without making any adjustments for demographics and school characteristics, teacher evaluations may not be a precise and fair representation of teacher effectiveness. We should think about effectiveness of teachers who teach comparable students, in comparable environments. Higher evaluation scores for teachers generally include greater gains of students. Part of the effectiveness rating is based on students' growth.
So let's talk a little about what we learned today. We looked at the following questions. How can we identify student growth when considering teacher evaluation? What are the different student growth models? And how can we identify which model is best?
In today's lesson we looked at the process for determining student growth as teachers. We examined three different methods for doing this. Student growth percentiles, value added models that control for student backgrounds, and value added models that correlate student growth with school and teacher growth. Not only did I define each of these methods for you, but we discussed the pros and the cons of each of these three.
Now that you're more familiar with these concepts, let's reflect. Which of these methods for determining student growth have you experienced or observed? What was your experience with this method?
Thanks for joining me today in discussing the lesson, Student Growth Models. I hope you found value in this video lesson and are able to apply these ideas about student growth models, to your very own teaching.
Now it's your turn to apply what you've learned in this video. The Additional Resources section will be super helpful. This section is designed to help you discover useful ways to apply what you've learned here. Each link includes a brief description so you can easily target the resources that you want.
(00:00- 00:28) Introduction/Objectives
(00:29- 01:10) How Can We Measure Student Growth?
(01:11- 01:49) Methods Available to Measure Growth
(01:50- 04:16) Method 1: Student Growth Percentile
(04:17- 05:20) Method 2: VAM- Controlling for Student Background
(05:21- 06:30) Method 3: VAM- Correlating Student Growth and School/Teacher Growth
(06:31- 08:01) What is the Best Method?
(08:02- 08:41) Recap
(08:42- 09:22) Reflection
Selecting Growth Measures for School and Teacher Evaluations: Should Proportionality Matter?
This research article examines the measures for student growth and the need to account for school environment and student demographics.
Choosing the Right Growth Measure
This Education Next article states that methods should compare similar schools and teachers.
Battelle for Kids: Student Growth Measures in Ohio
This portal from the Ohio Department of Education includes a comprehensive review of how student growth measures are calculated and used in the Ohio Teacher Evaluation Model.