Mathematics anxiety

Mathematics anxiety describes feelings of apprehension, tension or discomfort experienced by many
individuals when performing mathematics or in a mathematical context (Richardson & Suinn, 1972).
It has been associated with cognitive difficulties performing mathematical tasks, potentially because
anxiety interferes with our ability to hold and manipulate information in mind (working memory),
but is predominantly an emotional problem (Ashcraft & Krause, 2007). Developmental dyscalculia,
on the other hand, is a cognitive difficulty in acquiring mathematical skills (Devine, Hill, Carey, &
Szűcs, 2017).
Aims
This project had multiple goals. Firstly, we wished to provide estimates of mathematics anxiety
prevalence amongst UK primary and secondary school students. Secondly, we wished to validate an
instrument for measuring mathematics anxiety in this group. Thirdly, we wanted to see how
mathematics anxiety is related to other individual factors, such as maths performance, gender and
individual cognitive differences. Additionally, we used interviews to gain a richer understanding of
students’ experiences with and feelings towards mathematics.
Methodology
Our study had two main phases. In the first of these, we worked with more than 1700 primary and
secondary students to screen for mathematics anxiety, test anxiety and general anxiety, and gain a
measure of mathematics and reading performance. In the second phase, we worked one-to-one
with the children to gain a deeper understanding of their cognitive abilities and feelings towards
mathematics, using a series of sessions administering cognitive tasks, questionnaires and interviews
(Carey, Devine, et al., 2017; Carey, Hill, Devine, & Szűcs, 2017; Devine et al., 2017). Throughout, we 3
worked on two further sub-projects looking at the relationship between maths anxiety and
performance and cross-cultural comparisons (Carey, Hill, Devine, & Szucs, 2015; Hill et al., 2016).
Summary of Key Findings
o We have validated, in our large sample of British children, that the modified Abbreviated Math
Anxiety Scale is reliable (it appears to measure one construct) and valid (this construct seems to
really be maths anxiety, rather than another form of anxiety or other feelings towards maths
(Carey, Hill, et al., 2017). See Validating the modified Abbreviated Math Anxiety Scale (mAMAS)
for more details.
o We conducted a literature review into the long-established relationship between maths anxiety
and performance (those with higher maths anxiety tend to have poorer maths performance).
We conclude that this is likely because anxiety interferes with performance and poorer
performance increases anxiety, acting as a vicious circle (Carey et al., 2015). See Exploring the
relationship between mathematics anxiety and performance for more details.
o In our large sample of British children, we investigated the relationship between maths anxiety
and developmental dyscalculia. We found that whilst more dyscalculics than typical children
met criteria for maths anxiety, the majority of those with maths anxiety had normal
performance (Devine et al., 2017). See The relationship between mathematics anxiety and
developmental dyscalculia for more details.
o In a separate group of Italian children, we participated in research looking at developmental
change, gender differences and specificity of maths anxiety. We found that unlike general
anxiety, maths anxiety increases with age. The relationship between maths anxiety and
performance becomes more specific with age – in younger, but not older, children, this
relationship disappears after accounting for general anxiety. See Maths anxiety: Gender
differences, developmental change and anxiety specificity for more details.
o We have identified, in our large British sample, anxiety subgroups. These may increase in
complexity with age. In our secondary school students, we found that those with anxiety specific
to academia (high maths and test anxiety) had poorer performance than those with higher, but
less specific, anxiety. We conclude that this may reflect a dual path in anxiety development and
maintenance (Carey, Devine, et al., 2017). See Anxiety profiles and their relationship with
performance for more details.
o In our smaller subsample of British students, with whom we conducted further testing, we
looked at the relationship between various cognitive variables and maths performance. It seems
that a myriad of factors are associated with maths performance, but that basic numerical
processing is not (unpublished data). See Cognitive factors in mathematics performance for
more details.
o In another Italian sample, we investigated specific memory subtypes and their relationship with
maths anxiety and dyscalculia. Whereas maths anxiety appears to be associated with a deficit in
verbal working memory and perhaps also visuospatial working memory, dyscalculia is
associated with deficits in visuospatial memory; both short-term and working memory are
affected (Mammarella, Hill, Devine, Caviola, & Szűcs, 2015). See Working memory, dyscalculia
and maths anxiety in Italian students for more details.
o Our qualitative research has shown that children of 9-10 years are able to discuss their
experiences and origins of mathematics anxiety. Mathematically anxious children seemed to
describe negative events with less contextualisation. They were also more likely to discuss 4
physical sensations in their maths classes and clearly articulated some of the negative
consequences of maths anxiety. See Qualitative research: Experiences and origins of
mathematics anxiety for more details.
Conclusions
Each of the completed projects within o