The article “Is It Really Good to Talk? Testing the Impact of Providing Concurrent Verbal Protocols on Driving Performance” (Ergonomics 2017: 770–779) reported an experiment in which 20 drivers drove four laps on a fixed course. During two of the laps, drivers remained silent; on the other two, drivers were instructed to “think aloud” about their driving as they proceeded along the course. Lap order was randomized for each driver, and each driver’s average speed throughout the lap was recorded.
With A = protocol (silent or thinking aloud) and B = driver, sums of squares consistent with information in the article include SSA = 6.272, SSB = 343.975, SSAB = 46.733, and SSE = 138.571. Protocol is a fixed factor, while driver is a random factor. Perform the appropriate two-factor ANOVA (complete with ANOVA table), testing each effect at the .05 significance level. Explain what each F test tells you.