Statistical analysis in Small-N Designs: using linear mixed-effects modeling for evaluating intervention effectiveness.

The clinical result comes from five adults, so it is a small demonstration rather than proof that distributed spelling training works for all people with acquired dysgraphia. The paper's main purpose is to teach linear mixed-effects modelling, and the dysgraphia analysis is used to demonstrate that method. The schedule comparison was within the same spell-study-spell training, so it cannot separate the effect of the training method from the effect of the schedule. With only five participants, estimates of variability across individuals are susceptible to inflated Type I error. The accuracy analysis excluded most within-session clustered trials, reducing available data by about 35%. Autocorrelation and low statistical power remain concerns for small repeated-measures training studies. Follow-up maintenance was reported but was not the focus of the main analyses.

No potential conflict of interest was reported by the authors.

Do not treat this as proof that distributed spelling training works for all people with acquired dysgraphia. It is a five-participant demonstration of statistical analysis, and the schedule comparison used the same spell-study-spell training with reduced clustered-session data.