FESE HandBook v03c 15112023 MEV- COMPLETO - Flipbook - Página 400
400
LIDERAR LA EDUCACIÓN INTEGRAL (LEI) COMO CAMINO PARA LA INNOVACIÓN PEDAGÓGICA / TERCERA PARTE
of students within schools. Classroom and school level outcomes were
analyzed by employing propensity weighting in the impact estimation, given some imbalance on teacher and student level characteristics
at baseline. The qualitative data was coded and analyzed by employing
multiple steps and techniques such as performing a thematic analysis,
searching for and examining both conorming and disconorming evidence and summarizing key thematic ondings in a matrix that facilitated cross-site analysis, and identifying common or divergent ondings
across the eight focal schools.
3. RESULTS
3.1 QUANTITATIVE RESULTS
The researchers orst investigated the impact on student foundational
skills in literacy and mathematics. With regard to literacy, students in
classrooms taught by TFN fellows scored approximately .07 standard
deviations higher than students in comparison classrooms. With regard
to mathematics, students in classrooms taught by TFN fellows scored
approximately .11 standard deviations higher than students in comparison classrooms. In terms of expected impacts on educational outcomes, these can be interpreted as medium-sized effects. These results
show that TFN classrooms have up to half a year of advantage in terms
of math achievement and up to a quarter of an academic year for literacy
achievement compared to comparison classrooms based on translating
standard deviation effect sizes into months of schooling exercises1. The
study did not ond signiocant subgroup differences driven by gender
and grade level.
In regards to the prioritized holistic student outcomes of TFN, the
quantitative analysis found that there were few differences in social
and emotional learning between TFN students and non-TFN students.
Most impact estimates were positive, non-signiocant and no larger than
0.06 SD in absolute magnitude. The study did ond some signiocant positive differences among subgroup analyses by grade level, speciocally
that students improved in their social awareness and growth mindset
among students in grades 6. However, subgroup analyses of the impacts
on student self-efocacy in grades 4 and 6 are statistically signiocant and
1. We convert the Standard Deviation growth into years of schooling for ease of interpretation by following
recommendations and assumptions from Baird and Pane (2019). The ranges of months of schooling
represent differing assumptions of typical standardized growth including conservative estimates from
Bloom et al. 2008 and a universal average of standardized growth of 0.25 SD per year across subjects/grades
from Hanushek et al. 2012. We recognize the limitations of using years of schooling as an accurate measure
of learning gains given assumptions are based on studies outside of the Nigeria context.
