Claim Missing Document
Check
Articles

Found 1 Documents
Search

From Data Reading to Statistical Literacy: Examining Grade-Level and Gender Differences in Indonesian Elementary Students’ Data-Description Skills Uswatun Hasanah; Susilahudin Putrawangsa
Jurnal Kependidikan : Jurnal Hasil Penelitian dan Kajian Kepustakaan di Bidang Pendidikan, Pengajaran, dan Pembelajaran Vol. 12 No. 3 (2026): September (IN PRESS)
Publisher : LPPM Universitas Pendidikan Mandalika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33394/jk.v12i3.21114

Abstract

This study aims to examine Indonesian elementary students’ data-description skills as a foundational dimension of statistical literacy and to investigate differences across grade levels and gender. A quantitative cross-sectional design was employed with 2,106 students from Grades 2, 4, and 6 across 16 urban and rural schools. Four open-response tasks assessed students’ abilities to identify information in tables, recognize visual codes, explain data, and make predictions or draw conclusions. Confirmatory Factor Analysis (CFA) and multi-group CFA were used to examine the measurement structure, while Welch’s t-test, Welch’s ANOVA with Games–Howell post hoc comparisons, and Cohen’s d were used to assess group differences and effect sizes. All four indicators significantly contributed to the data-description construct, with drawing conclusions showing the strongest association. Most students (78.3%) were classified at the Idiosyncratic level, indicating predominantly literal or surface-level interpretations of data. Mean scores increased from Grade 2 (M = 1.15) to Grade 4 (M = 1.33) and Grade 6 (M = 1.92), with significant differences across all grade-level pairs (p < .001; d = 0.47–1.39). Female students scored slightly higher than male students, although the effect size was small (d = 0.27). Instruction should therefore move beyond literal data reading by engaging students more explicitly and consistently in comparing data, explaining relationships, and justifying predictions or conclusions using evidence from tables and graphs.