Abstract The Computer Anxiety Rating Scale and the Computer Thoughts Survey were administered to 661 students enrolled on undergraduate programmes in five different universities. A covariance matrix and an asymptotic weight matrix for each scale were computed from the sample data using PRELIS2. Two factor models were specified and estimated by maximum likelihood using LISREL8. The results revealed that the three-factor model of the Computer Anxiety Rating Scale was judged to be a poor explanation of the data. On this basis it may be suggested that the use of the sub-scale scores derived from the Computer Anxiety Rating Scale may lack validity. The three-factor model of the Computer Thoughts Survey was judged to be an acceptable description of the data. The factor analysis further revealed that the factor loadings were positive and statistically significant. The results suggest that the proposed factor structure of the Computer Thoughts Survey was consistent with the sample data.