Fig 1.
Performance score distributions.
Distributions of (A) standardized reading performance scores and (B) standardized math performance scores.
Table 1.
Performance and anxiety scores and correlations.
Fig 2.
Distributions of (A) scaled general anxiety (GA) scores, (B) scaled test anxiety (TA) scores and (C) scaled MA scores.
Fig 3.
The Relationship between academic performance and math anxiety.
(A) Density scatter plot showing probability of each standardized math performance score at each scaled MA level. (B) Density scatter plot showing probability of each standardized reading performance score at each scaled MA level. (C) Conditional probability of standardized math performance being equal to or above the specified threshold at each scaled MA level. (D) Conditional probability of standardized reading performance being at or above the specified threshold at each scaled MA level.
Table 2.
Measures of LPA model fit.
Fig 4.
Anxiety, gender and performance in each LPA profile.
(A) Line graph showing mean levels of general anxiety (GA), test anxiety (TA) and math anxiety (MA) in each LPA profile for year 4 children. (B) Line graph showing mean levels of GA, TA and MA in each LPA profile for year 7/8 children. (C) Stacked bar graph showing the number of girls and boys in each LPA profile for year 4 children. (D) Stacked bar graph showing the number of girls and boys in each LPA profile for year 7/8 children. (E) Line graph showing mean standardized reading and math performance score in each LPA profile for year 4 children. (F) Line graph showing mean standardized math and reading performance in each LPA profile for year 7/8 children. All error bars show standard error of the mean.
Table 3.
Regression model fit statistics for secondary students.
Table 4.
Optimal regression model statistics for secondary students.
Fig 5.
Hypothesized model of LPA profile determination.
Simplified binary-choice diagram suggesting factors which might influence a child’s LPA profile via their levels of general anxiety (GA), test anxiety (TA) and MA. Note that children in the “High anxiety” profile are likely to have developed higher MA levels as a result of a general predisposition to anxiety, whereas those in the “Academic anxiety” profile are more likely to have developed high MA as a result of poor academic performance. This explains why those in the “High anxiety” profile have higher academic performance relative to their absolute MA levels than those in the other profiles.