Researchers have recently started evaluating whether stimulating the brain noninvasively with a weak and painless electrical current (transcranial Electrical Stimulation, tES) enhances physiological and cognitive processes. Some studies found that tES has weak but positive effects on brain physiology, cognition, or assessment performance, which has attracted massive public interest. We present the first meta-analytic test of the hypothesis that tES in a learning phase is more effective than tES in an assessment phase. The meta-analysis included 246 effect sizes from studies on language or mathematical competence. The effect of tES was stronger when stimulation was administered during a learning phase (d = 0.712) as compared to stimulation administered during test performance (d = 0.207). The overall effect was stimulation-dosage specific and, as found in a previous meta-analysis, significant only for anodal stimulation and not for cathodal. The results provide evidence for the modulation of long-term synaptic plasticity by tES in the context of practically relevant learning tasks and highlight the need for more systematic evaluations of tES in educational settings.