|
Article Excerpt
R. J. Cruise and E. M. Wilkins's (1980) Statistical Anxiety Rating Scale (STARS), which measures levels of statistics anxiety, was administered to 169 technikon students (70.9% women) in the Eastern Cape of South Africa. Evidence of score reliability and validity was provided, suggesting that the STARS can be used with this population.
**********
An increasing number of college students worldwide, including those in the United States, are required to enroll in at least one statistics course as a necessary part of their degree programs (Onwuegbuzie & Wilson, 2003). Unfortunately, a large proportion of these students report experiencing high levels of statistics anxiety while enrolled in these classes. For example, between two thirds and four fifths of graduate students have been found to experience unmanageable levels of statistics anxiety (cf. Onwuegbuzie & Wilson, 2003). In fact, researchers have indicated that courses in statistics are among the most anxiety inducing among students taking non-mathematics-oriented disciplines (Blalock, 1987; Caine, Centa, Doroff, Horowitz, & Wisenbaker, 1978; Gaydosh, 1990; Lundgren & Fawcett, 1980; Schacht & Stewart, 1990, 1991; Zeidner, 1991). Unfortunately, high levels of statistics anxiety have been found to debilitate performance in statistics and quantitativebased courses (Onwuegbuzie, 1997). More specifically, statistics anxiety has been documented as affecting college students' ability to acquire the skills, knowledge, and strategies necessary to interpret and critique research reports. Also, this anxiety affects the ability to propose, to design, and to implement research studies (Onwuegbuzie, 1997). Furthermore, high-anxious college students often delay enrolling in statistics courses for as long as possible, with some students delaying enrollment until the final semester of their degree programs--which is clearly not the optimal time to take such courses (Onwuegbuzie, 1997; Roberts & Bilderback, 1980).
Statistics anxiety refers to feelings of anxiety that are experienced by those taking a statistics course or undertaking statistical analyses, in terms of gathering, processing, and interpreting data (Cruise, Cash, & Bolton, 1985). Onwuegbuzie, DaRos, and Ryan (1997) have defined statistics anxiety more broadly as worry and emotionality that occur when students encounter statistics in any form and at any level. Statistics anxiety is situation-specific because it comes to the fore when students are either learning statistical concepts, terminology, and formulae or applying statistics in a particular context (Benson & Bandalos, 1989). Statistics anxiety has been conceptualized as being multidimensional (Cruise et al., 1985; Cruise & Wilkins, 1980; Onwuegbuzie et al., 1997). In particular, Cruise et al. identified the following six components of statistics anxiety: (a) Worth of Statistics, (b) Interpretation Anxiety, (c) Test and Class Anxiety, (d) Computational Self-Concept, (e) Fear of Asking for Help, and (f) Fear of Statistics Teachers. According to these authors, Worth of Statistics refers to students' perceptions of the relevance and usefulness of statistics. Interpretation Anxiety concerns the anxiety experienced when students are faced with making decisions from or interpreting statistical data. Test and Class Anxiety represents anxiety that affects students when they are in a statistics class or are involved in taking a statistics test. Computational Self-Concept relates to the anxiety experienced when students attempt to solve mathematical problems. It also affects students' perceptions of their ability to do mathematics. Fear of Asking for Help represents the anxiety experienced by students when they have to ask others or their lecturers for help in understanding material covered in class or statistical data in articles, printouts, and the like. Finally, Fear of Statistics Teachers is anxiety that is triggered by students' perceptions that the statistics instructor is unapproachable.
Courses in basic statistics are important to nonstatistics majors because they represent the only formal exposure to research methods and analyses that students may receive before embarking on their chosen careers. Such courses are offered in order to afford students analytical skills, as well as research production skills. Analytical skills enhance students' ability to read, to interpret, to synthesize, and to use reported results. On the other hand, research production skills enable students to design and to initiate original research (Ravid & Leon, 1995). Students' experiences in basic statistics courses, however, often are a source of anxiety that may result in negative perceptions of the subject. This seems to be the case especially among students who have undergone 12 years of schooling without ever taking statistics as a subject.
In South Africa, statistics is currently not offered at school level, that is, from Grades 1 to 12. A few statistical topics such as handling of data, descriptive statistics, drawing graphs (histograms), and probability have recently been introduced in the mathematics syllabus. However, mathematics students in South Africa constitute a very small fraction of the student population. Another factor that may be a source of statistics anxiety in South Africa could be the inability of students taking courses such as taxation and civil engineering to appreciate why statistics is included in their programs of study. For example, in an investigation conducted at a technikon (i.e., a technological university) in South Africa, Mahiti (2002) reported that a student indicated, "I don't see why I have to know and fill my head with statistics and mathematics" (p. 9). When asked why she thought that way, the student revealed, "I have never performed well in mathematics, I don't like statistics and I just don't see how all this will play a role when I am employed as an IT specialist" (p. 9).
Because statistics anxiety is so prevalent among college students in the United States and likely elsewhere, clearly, early identification of statistics anxiety has important counseling implications. To identify statistics anxiety, measures are needed with adequate psychometric properties, that is, measures that consistently yield reliable and valid scores. Onwuegbuzie and Wilson (2003) identified five scales that directly measure statistics anxiety. These scales are the Statistics Anxiety Scale (Pretorius & Norman, 1992), the Multifactorial Scale of Attitudes Toward Statistics (Auzmendi, 1991), the Statistics Anxiety Inventory (Zeidner, 1991), an unnamed instrument measuring statistics anxiety and attitudes developed by Zanakis and Valenza (1997), and the Statistical Anxiety Rating Scale (STARS; Cruise et al., 1985). However, the first four measures of statistics anxiety have been used in very few studies, and little is known about their psychometric properties. Indeed, Cruise et al.'s STARS presently is by far the most utilized measure of statistics anxiety (Onwuegbuzie & Wilson, 2003). Moreover, of these five instruments, only the STARS has been subjected to validity studies. However, with the exception of the validity study conducted by Cruise et al., information about the psychometric properties of the STARS is very limited.
Reliability and validity are functions of scores and not of tests (Onwuegbuzie & Daniel, 2002, in press; Thompson & Vacha-Haase, 2000). Therefore, it should not be assumed that a scale that is appropriate for one population (e.g., U.S. college students) also is appropriate for another population (e.g., South African college students). Moreover, by documenting reliable and valid scores across different populations, researchers will be in...
|
