of 1

Current View
Research Brief | California Science Test
California
California Schools That Use Twig Science
Elementary Have Significantly More Students
Who Meet California Science Test Standards
OVERVIEW
Twig Science Elementary is a phenomena-based core science curriculum that places the power of discovery into students’
hands and encourages them to inquire deeply and think critically about science and engineering in the real world. It combines
engaging science content with a flexible digital platform to support in-person, hybrid, and virtual-only models.
Twig Science Elementary was utilized by almost 100 school districts in California during the 2021-2022 school year. To investigate
the overall efficacy of the Twig Science Elementary curriculum, we evaluated the average performance on the Grade 5 California
Science Test (CAST) for schools in districts that used Twig Science Elementary compared to the average performance of schools
that did not use Twig Science Elementary.
Table 1. Demographics of Non-User and User Schools
Non-User User
Number of Schools 5,257 638
Average Number of Grade 5
Students Per School 73 80
Economically Disadvantaged
Students 60.8% 57.0%
English Learner Students 20.0% 21.4%
Black/American Students 4.4% 3.7%
Hispanic/Latino Students 56.6% 52.9%
Students Who Identify as
2+ Races 4.4% 4.7%
0%
10%
20%
30%
40%
Non-User School User School
30.8%34.3%*

Note: *p < 0.001
RESULTS
Figure 1 presents the average school-level percentage of students who
met or exceeded Grade 5 CAST standards based on Twig Science
Elementary user status. A t-test found that California schools that
used Twig Science Elementary had a statistically significantly
higher proportion of students who met or exceeded Grade 5 CAST
standards than California schools that did not use Twig Science
Elementary (p-value <0.001). Statistical significance indicates that
it is unlikely that the difference in performance between user and
non-user districts is due to chance alone. The computed Cohen’s d
effect size for this comparison was found to be 0.17 which can be
considered a medium effect.1
An ordinary least squares regression model was used to determine
whether a difference in performance could be further demonstrated
after controlling for school-level demographics between user and
non-user districts. The regression controlled for student demographics
including the percent of economically disadvantaged students,
percent of English Learner (EL) students, percent of Black/African
American students, percent of Hispanic/Latino students, and percent
of students that identify as two or more races. See Table 1 for a
summary of the demographics of user and non-user schools.
Results indicate that Twig Science Elementary schools still had
significantly higher proportions of students who met or exceeded
the Grade 5 CAST standard after controlling for these school-level
demographics (coefficient on Twig user indicator = 1.5%, p-value <0.001).
1P-value is calculated using a weighted t-test and effect size calculated using Cohen’s
d. An effect size less than 0.05 is considered small, 0.05 to less than 0.2 is considered
medium, and 0.2 or greater is considered large according to Kraft, M.A. (2020). Interpreting
effect sizes of education interventions. Educational Research, 49(4), 241-253.
Figure 1. Average School-Level Percentage of
Students Who Met or Exceeded CAST Standards
by Twig Science Elementary User Status
imaginelearning.com
877-725-4257 • solutions@imaginelearning.com
1020312037 2302