Delphi Validation Process of a Tool to Assess STEAM Materials in Early Childhood Education

Dataset Overview This dataset contains the results of a two-round Delphi process aimed at developing and validating an assessment tool for STEAM educational materials in early childhood education (ages 3–6), grounded in Montessori pedagogy. The dataset includes both quantitative and qualitative data derived from expert evaluations, as well as a complementary systematic table of existing STEAM assessment tools and frameworks used to inform the development process. Context and Purpose The dataset supports a doctoral research project conducted at the Universitat Rovira i Virgili (URV), focused on designing a validated, research-based instrument to assess STEAM materials in early childhood education. Given the lack of standardized tools for evaluating STEAM implementation in Montessori settings, this dataset contributes to bridging that gap through expert consensus and alignment with existing frameworks. Methodological Design A Delphi methodology was employed to reach consensus among experts in: Early childhood education STEAM education Montessori pedagogy Digital and teacher competence The process consisted of: Round 1: Initial evaluation of proposed dimensions, indicators, and items Round 2: Refinement based on feedback and re-evaluation of items Dataset Contents The dataset is composed of three main components: 1. Delphi Round 1 Data Expert ratings of items (Likert scale) Qualitative feedback and suggestions 2. Delphi Round 2 Data Revised item evaluations Assessment of value system 3. STEAM Assessment Framework Table A structured review of relevant studies and frameworks, including: STEAM competence assessments (e.g., Chen & Ding, 2025) Teacher digital competence rubrics (e.g., Lázaro-Cantabrana & Gisbert, 2015) STEAM educator frameworks (e.g., Kim & Kim, 2016; Spyropoulou & Kameas, 2024) European project outputs (e.g., STEAMonEdu deliverables) This table includes: Type of contribution Educational level Key contributions to assessment and evaluation Variables and Measures Likert-scale ratings (e.g., relevance, clarity, coherence) Content Validity Index (CVI): Item-level CVI (I-CVI) Scale-level CVI (S-CVI) Qualitative comments from experts Item revisions across rounds Data Processing and Analysis Descriptive statistics (means, standard deviations) Calculation of Content Validity Index (CVI) following established procedures Iterative refinement of items based on: Statistical thresholds (e.g., agreement levels) Expert qualitative feedback Related Outputs Doctoral thesis (in progress) STEAM assessment tool (final validated version) Conference presentations and journal articles