Papers and Journal Articles

Books

Treatment Protocol

Biofeedback: Spectral Analysis

Current Research Project

Contact Information

Statistical techniques

To explore these questions a variety of statistical techniques were used.  Simple univariate descriptive statistics were calculated to summarize each variable; t-tests were used to look at differences in the variable means between the subgroup of people who attained seizure control and the subgroup who did not; and associations between the variables and seizure control were examined by the Chi-square(X²) test.  Multivariate analyses were used to consider the relationships among several variables at the same time.  Regression equations were estimated using the number of sessions (NUMSESS) as the dependent variable and the remaining variables as independent variables.  In addition, difficulty in achieving seizure control was predicted using discriminant analysis with a dependent variable constructed from both SEIZCON and NUMSESS.