Spatial ability contributes to math word problem solving beyond general cognitive ability: evidence from structural equations modelling and network analysis
Maxim Likhanov, Fang Wang, Xinlin Zhou, Jianing Lyu
Abstract
Maxim Likhanov, Fang Wang, Xinlin Zhou, Jianing Lyu
Abstract
There is a growing body of research into the factors contributing to math word problem solving. However, these studies usually use limited number of potential predictors (precluding assessing of their contribution in comparison with other factors or “g” general intelligence) and some predictors (such as analogical and hypothetical reasoning) are largely omitted. Thus, the aim of the current study was to explore contributions of different types of reasoning to math word problem solving and whether these contributions have added value compared with general cognitive ability. Chinese schoolchildren in Grades 3 (N = 199; Mage = 102.4 months), 4 (N = 162; Mage = 114.6), 5 (N = 174; Mage = 126.1) and 6 (N = 180; Mage = 138.6) completed 8 tasks tapping into spatial, mechanical, verbal, mathematic, hypothetical and analogical reasoning. Our data showed that when 6 general cognitive factors load onto General cognitive ability factor in a Structural Equation Model (SEM), only a measure of spatial visualization (Paper folding) has additional contribution to Word problem solving factor. Gaussian Graphical models (GGMs) showed that 2 verbal tasks and Paper folding showed stable (present in at least 3 out of 4 grades) contributions to both word problem solving tasks. Analogical reasoning showed contribution to Process of word problem solving only. To sum up, both SEM and GGMs converged on the importance of spatial ability for mathematical word problems solving. Our results call for verbal and spatial ability to be routinely assessed and targeted by educational interventions within maths curriculum.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
There is a growing body of research into the factors contributing to math word problem solving. However, these studies usually use limited number of potential predictors (precluding assessing of their contribution in comparison with other factors or “g” general intelligence) and some predictors (such as analogical and hypothetical reasoning) are largely omitted. Thus, the aim of the current study was to explore contributions of different types of reasoning to math word problem solving and whether these contributions have added value compared with general cognitive ability. Chinese schoolchildren in Grades 3 (N = 199; Mage = 102.4 months), 4 (N = 162; Mage = 114.6), 5 (N = 174; Mage = 126.1) and 6 (N = 180; Mage = 138.6) completed 8 tasks tapping into spatial, mechanical, verbal, mathematic, hypothetical and analogical reasoning. Our data showed that when 6 general cognitive factors load onto General cognitive ability factor in a Structural Equation Model (SEM), only a measure of spatial visualization (Paper folding) has additional contribution to Word problem solving factor. Gaussian Graphical models (GGMs) showed that 2 verbal tasks and Paper folding showed stable (present in at least 3 out of 4 grades) contributions to both word problem solving tasks. Analogical reasoning showed contribution to Process of word problem solving only. To sum up, both SEM and GGMs converged on the importance of spatial ability for mathematical word problems solving. Our results call for verbal and spatial ability to be routinely assessed and targeted by educational interventions within maths curriculum.
Key concepts: Analogical reasoning, Verbal reasoning, Spatial intelligence, Psychology, Abstract reasoning, Cognitive psychology, Computer science, Linguistics