Table 2 Fixed effect Omnibus tests (Type III Anova) on log RTs in correct word responses.

From: Mirror invariance dies hard during letter processing by dyslexic college students

Fixed-effects

LogRT, linear modela

Group

F(1, 36.0) = 15.316, p < .001, MS = 0.745 *

Letter

F(1, 372.6) < 1, p = .930

Prime

F(3, 13,161.1) = 45.365, p < .001. MS = 2.208 *

Group x letter

F(1, 13,055.7) = 2.128, p = .145, MS = 0.104

Group x prime

F(3, 13,295.0) < 1, p = .516

Letter x prime

F(3, 13,160.9) = 2.050, p = .105

Group x letter x prime

F(3, 13,294.9) = 3.206, p = .022, MS = 0.156 *

  1. Omnibus effects were extracted after centering each variable (contr.sum).
  2. a Log(RT) analysis: 13,484 datapoints; 384 items; 38 participants; p-values calculated via Satterthwaite’s approximation. AIC = − 1615.27; BIC = − 1472.594. Model equation (from lme4): lmer(log(RT) ~ ReadGroup * primeC * letterC + (1|Subject) + (1|Item), data = RT_word, control = lmerControl(optimizer = ‘bobyqa’)).
  3. * p ≤ .01 (significant effects are in bold).
Back to article page