The Relationship of Trait and Test Anxiety with Mathematics Anxiety

Abstract

Correlational and comorbidity analyses of math anxiety's relationship with trait and test anxiety among college students were conducted. Significant correlations were obtained among all three measures, with women reporting higher levels of math and test anxiety. A majority of students considered themselves math anxious, with most indicating interest in services for it. Comorbidity analyses revealed heightened trait and/or test anxiety for both math anxious men (71.4%) and women (67.2%). Equal percentages of each reported comorbid levels of both trait and test anxiety. Elevated levels of only one other type of anxiety varied as a function of gender. Women were more likely to exhibit comorbid test anxiety and men trait anxiety. Implications for treatment of math anxiety are discussed. ********** Mathematics anxiety often has been cited as a factor limiting educational and career choices of college students, particularly women (Betz, 1978; Chipman, Krantz, & Silver, 1992; Resnick, Viehe, & Segla, 1982; Singer & Stake, 1986). Students who are severely math anxious may opt for academic majors within the humanities, fine arts, or social sciences that require minimal coursework in mathematics (Chipman et al., 1992). However, even students who select majors in which mathematics is not emphasized may be expected to successfully demonstrate at least minimal competency in mathematics as part of their general degree requirements. For example, all students graduating from the university with which we are affiliated are required to complete a course in college algebra or its equivalent. In so far as such a requirement would be expected to be especially stressful for students high in math anxiety, it is not surprising that treatment programs for math anxiety typically have targeted students majoring in nonscientific areas (e.g., Wadlinglon, Austin, & Bitner, 1992). A salient issue in designing interventions to help college students cope with math anxiety is the relationship between math anxiety and other forms of anxiety, including trait and especially test anxiety. For instance, students seeking treatment for math anxiety who experience significant comorbid levels of generalized or trait anxiety might be expected to benefit more from an intervention designed to reduce anxiety more generally (e.g., stress inoculation training) than one that more narrowly targets math anxiety, such as systematic desensitization. While math anxiety commonly has been regarded as a subtype or form of test anxiety, there appear to both conceptual and empirical reasons for not viewing the two as equivalent. Richardson and Woolfolk (1980), for example, have argued that math anxiety is most meaningfully conceptualized as a reaction to both mathematical content per se (numbers) and to evaluative situations, such as testing, in which mathematical skills are assessed. In particular, as it relates to mathematical content, math anxiety may be associated with feelings of perfectionism and inferiority and concerns about gender roles and identity. Empirically, math anxiety measures have been found to be more closely related to each other than to test anxiety and its components, especially among women college students (Dew, Galassi, & Galassi, 1983). However, the development of interventions for math anxiety appears to have substantially been influenced by the formulation of treatment approaches for test anxiety (Schneider & Nevid, 1993), and students who are both math and test anxious might be expected to respond optimally to an intervention that primarily focuses upon test anxiety. Unfortunately, research to date examining the relationship of trait and test anxiety with math anxiety has generally been limited to correlational analyses with college students not preselected to be at risk for math anxiety. For example, significant correlations between math anxiety and trait anxiety have been reported for a group of introductory psychology students (Betz, 1978) and students enrolled in an upper level statistics course (Adams & Holcomb, 1986). …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 88 citations exist upstream.