小學中、高年級學生一般詞彙知識與數學詞彙知識對數學成就的預測力

Abstract

一般詞彙知識與數學詞彙知識在數學成就中所扮演的角色日益受到關注。本研究探討在控制非語文智力和語文智力之下，一般詞彙知識與數學詞彙知識對數學成就之預測。研究對象為342位四年級學生與395位六年級學生，共737位學生。在研究工具上，測量智力採用國小中高年級學校能力測驗，測量一般詞彙知識採用詞彙成長測驗，測量數學詞彙知識採用國民中小學數學詞彙知識測驗，測量數學成就的採用自編的數學成就測驗。本研究在107年6月進行數學詞彙知識與一般詞彙知識測驗之施測，並於108年5月進行智力測驗、與數學成就測驗之施測。本研究採Mplus 8.3統計軟體進行分析，透過建立結構方程模式進行路徑分析，亦採SPSS 23.0統計軟體進行相關分析。研究結果顯示：第一，在四年級與六年級中，非語文智力和語文智力對數學成就皆具有預測力。四年級非語文智力的預測力為 .50，語文智力的預測力為 .22；六年級非語文智力的預測力為 .32，語文智力的預測力為 .42。顯示年級較低時，非語文智力的預測力較大；而年級較高時，語文智力的預測力較大。第二，在控制非語文智力和語文智力之下，四年級與六年級在一般詞彙知識對數學成就有獨特的預測力。四年級一般詞彙知識的預測力為 .14；六年級一般詞彙知識的預測力為 .25。第三，在控制非語文智力、語文智力與一般詞彙知識的預測之下，四年級與六年級在數學詞彙知識對數學成就仍有獨特的預測力。四年級數學詞彙知識的預測力為 .11；六年級數學詞彙知識的預測力為 .20。第四，四年級與六年級的數學詞彙知識在一般詞彙知識對數學成就的關係皆具有部分中介效果。四年級的部分中介效果值為 .04；六年級的部分中介效果值為 .09。此結果代表非語文智力、語文智力、一般詞彙知識、數學詞彙知識對數學成就分別都具有獨特的預測力，且隨著年級的增長而變化。一般詞彙知識會有部分透過數學詞彙知識間接預測數學成就，且高年級學生之預測力更高。最後，本研究建議未來相關研究可以針對施測時間有更好的設計，並擴展研究對象與變項多樣性的搜集，期望對一般詞彙知識與數學詞彙知識對數學成就之關係有更深入地了解。

The roles of general vocabulary knowledge and mathematics vocabulary knowledge are gradually gaining recognition in mathematics achievement. This study aims to determine the predictive power of the knowledge of general and mathematics vocabulary for mathematics achievement after controlling for non-verbal and verbal intelligence. A total of 737 students was recruited to participate in this study, including 342 fourth-graders and 395 sixth-graders. OLSAT-8 was administered to measure the subjects’ intelligence quotient, while general vocabulary was measured using the vocabulary growth test. In addition, the national elementary and middle school mathematics vocabulary knowledge test was used to measure the participants’ mathematics vocabulary, and a self-edited mathematics achievement test was adopted to measure the subjects’ mathematics achievement. Measurements of mathematics vocabulary and general vocabulary were conducted in June 2018, while intelligence tests and self-edited mathematics achievement tests were administered in May 2019. Mplus 8.3 was used for statistical analysis, structural equation modelling was performed for path analysis, and SPSS 23.0 was employed for correlation analysis. The results demonstrate that (1) non-verbal intelligence and verbal intelligence are both predictive of mathematics achievement in the two observed groups. The predictive powers o f non-verbal intelligence and verbal intelligence, respectively, were .50 and .22 in the fourth-graders and .32 and .42 in the sixth-graders. This result implies that the predictive power of non-verbal intelligence is greater in lower grades, while that of verbal intelligence is higher in higher grades. (2) After controlling for non-verbal and verbal intelligence, general vocabulary showed a particular predictive power for the mathematics achievement of fourth-graders and sixth-graders of .14 and .25, respectively. (3) Controlling for non-verbal intelligence, verbal intelligence, and general vocabulary, mathematics vocabulary showed a predictive power for mathematics achievement of .11 for fourth-graders and .20 for the sixth-graders. (4) Mediators exist in the relationship between mathematics vocabulary knowledge and general vocabulary knowledge with mathematics achievement, with mediation effects of .04 for fourth-graders and .09 for sixth-graders. This result indicates that non-verbal intelligence, verbal intelligence, general vocabulary knowledge, and mathematics vocabulary knowledge show specific predictabilities for mathematics achievement that vary by grade. Furthermore, some general vocabulary knowledge may indirectly influence mathematics achievement via mathematics vocabulary knowledge, with higher grades experiencing greater predictability. Finally, in the hope of acquiring a deeper insight into how general vocabulary knowledge and mathematics vocabulary knowledge would affect mathematics achievement, this study suggests lines of future research to optimize the timing of measurements on the aforementioned variables, expand the investigation to more subjects, and broaden the variables collected.