Jeri Benson

Structural Components of Statistical Test Anxiety in Adults: An Exploratory Model

AbstractThe purpose of this study was to explore the basis of test anxiety expressed when taking a statistics course using a structural modeling approach. The study involved 219 university students. The data indicated that statistical test anxiety was different from general test anxiety. The females expressed more general and statistical test anxiety than males, and students who had taken more prior math courses had higher math self-concept scores. Math self-concept and achievement in statistics were negatively related to statistical test anxiety, and the students who reported high levels of general test anxiety also reported high levels of statistical test anxiety. The structural model revealed variables not studied previously to be important in understanding statistical test anxiety.

Measuring test anxiety in children: Scale development and internal construct validation

Given the increased testing of school-aged children in the United States there is a need for a current and valid scale to measure the effects of test anxiety in children. The domain of childrens test anxiety was theorized to be comprised of three dimensions: thoughts, autonomic reactions, and off-task behaviors. Four stages are described in the evolution of the Childrens Test Anxiety Scale (CTAS): planning, construction, quantitative evaluation, and validation. A 50-item scale was administered to a development sample (N=230) of children in grades 3–6 to obtain item analysis and reliability estimates which resulted in a refined 30-item scale. The reduced scale was administered to a validation sample (N=261) to obtain construct validity evidence. A three-factor structure fit the data reasonably well. Recommendations for future research with the scale are described.

Testing the Factor Structure Invariance of a Computer Attitude Scale Over Two Grouping Conditions

A computer attitude scale is tested for invariance over the grouping conditions of males/females and graduate/undergraduate. Through both exploratory and confirmatory factor analysis, the original scale, developed by Loyd and Gressard (1984a), was revised to 23 items representing three factors. The revised scale was found to be invariant over both conditions in terms of number of factors and intercorrelations among the factors. Although there were slight differences in the item loadings and the item errors, the differences were not practically significant. Thus, the construct of computer anxiety appears to be multidimensional with highly correlated factors.

Using Lisrel to Evaluate Measurement Models and Scale Reliability

The present paper demonstrates how LISREL can be used to examine measurement model assumptions and to assess the reliability of a scale. LISREL was used to investigate (a) the nature of the basic measurement model for a scale, (b) scale invariance across time, and (c) scale invariance across groups. Information obtained in such analyses helps to assess accurately the estimate of the reliability of the scale. As an illustration, responses of 722 elementary school students to the Coopersmith Self-Esteem Inventory (Form B) were analyzed. Results indicated that a congeneric measurement model with correlated errors was most appropriate. Internal-consistency estimates were not affected by correlated errors, within each of the two time points (pretest and posttest). However, correlated errors across time produced an overestimate of the stability coefficient. Further, the factor structure of the scale was invariant across three ethnic groups, an invariance that implied similar reliability estimates across groups.

The multidimensionality of test anxiety: A test of different models

Confirmatory factor analysis was employed to test the multidimensional structure of test anxiety. Starting with the traditional two-factor model, it was of interest whether additional dimensions could still be represented by a single higher-order factor of test anxiety or whether the additional dimensions represented correlates of test anxiety. The Revised Test Anxiety (RTA) scale and the German Test Anxiety Inventory (TAI-G) were combined collecting data from a binational sample of 218 American and 218 German university students. Considering the statistical and theoretical aspects of model fit, they indicated that a model consisting of three primary factors (worry, emotionality, and lack of confidence) fit the data best. While distraction and self-efficacy may be regarded as correlates of test anxiety, lack of confidence was substantiated as a component of test anxiety. Implications for the conceptualization of the content domain and hence area of test anxiety were discussed.

The motive to avoid failure and test anxiety responses : Empirical support for integration of two research traditions

Prior research and theoretical considerations suggested that motive to avoid failure would serve as a basic and general factor of test anxiety. This suggestion led to the expectation that motive to avoid failure should account for a considerable amount of variation and covariation among the four factors of test anxiety responses suggested by Sarason (1984). A covariance structural modeling approach applied to a sample of 260 college students provided support for the expectation. Furthermore, the four test anxiety response factors correlated with motive to avoid failure to the extent they measured evaluative achievement themes. The findings were considered promising support for integrating the research traditions of achievement motivation and test anxiety.

Further refinement and validation of the revised test anxiety scale

The purpose of this study was to provide additional validity data concerning the four dimensions and precision of the Revised Test Anxiety Scale (RTA) with a new sample of American and Egyptian examinees. The RTA was developed based on combining nonredundant items from the Test Anxiety Inventory (Spielberger, Gonzalez, Taylor, Algaze, & Anton, 1978) and the Reactions To Tests scale (I. Sarason, 1984) in an effort to produce a psychometrically efficient and valid measure of test anxiety for multinational samples. The dimensionality of the RTA was evaluated by confirmatory factor analysis and cross‐validated. In addition, several new items were tested to improve the precision of each of the four dimensions. A 20‐item RTA resulted, which upon cross‐validation showed stability of the dimensionality and parameter estimates with adequate precision.

Second-Order Confirmatory Factor Analysis of the "Reactions to Tests" Scale with Cross-Validation.

The factor structure of the RTT was evaluated by testing a series of confirmatory factor models including a second-order structure. The results indicated that a reduced 20-item RTT scale fit the data better than the 40-item original RTT and that a second-order structure explained the data very well. The models were cross-validated using several strategies. The reduced RTT was found not to cross-validate in terms of overall fit. However, when the parameter estimates were tested across samples, all parameters were found to be invariant except the item residuals. Because the cross-validation results were not conclusive, the need for additional theoretical and empirical research on the cross-validation of covariance models was discussed.

Detecting Item BIAS in Affective Scales

The purpose of the study was to demonstrate the use of confirmatory factor analysis to investigate the existence of item bias in an affective scale measuring self-concept. The data were collected from three ethnic groups (Black, Hispanic, and White) in grade eight (N = 1264). The factor structure and item uniqueness were simultaneously compared across ethnic groups by using the LISREL program. A differential response pattern found across the three groups indicated potential bias in score interpretation. The study illustrated that structural analysis may be a useful methodology in the investigation of item bias in affective scales.

The emotional regulation during test-taking scale

The article describes the development and initial validity efforts for a context-specific measure of Emotional Regulation during Test-taking (ERT). Presented are results from both exploratory and confirmatory factor analysis on independent half samples, one used for development and one used for validation, which supported a 39-item, four-dimensional structure (i.e., Task-Focusing, Emotion-Focusing, Regaining Task-Focus, and Cognitive-Appraising Processes). In addition, results examining the relationship between the ERT and relevant measures of test anxiety, mood regulation, and coping showed support for the construct validity of the ERT scale.