
URACCAN
REVISTA UNIVERSITARIA DEL CARIBE No. 32
24
experience. In addition, most of the participants also possessed formal training in inclusive education,
Universal Design for Learning (UDL), or special education. Recruitment was facilitated by working in
collaboration with educational networks and professional development programs within the university.
Participants' informed consent was secured, and condentiality and anonymity were maintained in
accordance with ethical research standards.
For ensuring a rich description of teaching practices and system eects, four broad data collection
instruments were employed: (1) semi-structured interviews, (2) classroom observations, (3) analysis
of curriculum materials, and (4) instructional artifacts produced by the teachers. e semi-structured
interviews were conducted on an overall protocol with local context-specic modication. e interviews
were conducted in the mother tongue of the participants, i.e., English, Spanish, and audio-recorded after
obtaining the consent of the participants. All of the interviews lasted between 45 and 75 minutes and
dealt with the overall topics of the teacher's conceptualization of accessible geometry, specic strategies
and devices employed, experienced diculties, support from others, and their views in relation to policy
usefulness and teacher professional development. e interview guide consisted of 15 open-ended guiding
questions with optional probes and was piloted with two non-study sample teachers to gauge clarity and
consistency.
Classroom observation was carried out with 16 of 24 teachers, depending on availability and willingness.
e teachers were observed across two lessons of geometry each, and this was equivalent to a total of 32
observations. ey were implemented in person in Spain and virtually in the United States due to logistical
and institutional constraints, primarily regarding school district clearances and post-pandemic procedures.
An observation guide was used to document pedagogical routines, student participation, utilization
of materials, strategies of dierentiation, and physical and intellectual access to the classroom. Field
observations were also conducted by observers to record contextual variables, teacher-student interactions,
and salient events, which would be combined with qualitative data.
Analysis of curriculum documents included a review of national curriculum frameworks, inclusive
education policy, and contextualized school-level implementations of curricula. Material reviewed for the
United States was the Common Core State Standards for Mathematics, Individualized Education Program
(IEP) reports, and inclusive education manuals. For Spain, the materials included the national curriculum
in LOMLOE, curricular adaptation in the regions, and school-based educational response plans for SEN
students. All the documents were coded using a coding scheme from accessibility, inclusion principles,
and geometry content representation. e coding scheme was developed step by step via initial document
analysis and consisted of categories such as "visual representation," "manipulative use," "integration of
technology," "alignment with UDL," and "dierentiation in assessment."
Teachers' instructional artifacts, such as lesson plans, adapted materials, worksheets, student assignments,
and computer programs, were collected with their consent and analyzed to see how teachers incorporate
accessible geometry in the curriculum. Materials were coded for level of adaptation (low, medium, high),
type of accessibility feature (tactile, sound, visual simplication), and target student prole. Materials were
translated and anonymized where necessary for cross-country comparison, included follow-up interview
questions in artifact analysis to capture thought behind design decisions.
Data were coded and analyzed using NVivo software with thematic. Open coding was subsequently
followed by an initial coding phase in which two researchers coded separately a subset of observation
and interview transcripts. Axial coding was subsequently incorporated to establish relationships among
categories and thematic clustering to establish core themes. To assess reliability, intercoder agreement was
estimated and a Cohen's Kappa of 0.82 was obtained, indicating substantial agreement. Disagreements
were debated and resolved collectively. After setting up the coding scheme, this was applied to all of the
dataset, including interview transcripts, observation notes, documents, and artifacts.