(Table of means&SD’s)
The data were normally distributed, with skew and kurtosis within limits of +/-2. There were no extreme scores. However, although Levene’s test of homogeneity of variance was not significant for the texting data (p=.882), it was significant for the phone call data (p=.029). The data were therefore log transformed. Upon transformation the data fulfilled homogeneity of variance (Levene’s p=.316 for the text data and p=.298 for the phone call data). A 2 (phone: Nokia or Motorola) x 2 (task: text message or phone call) mixed design ANOVA was performed on the transformed data, with phone as the between subjects variable and task as the within subjects variable. There was an effect of task (F(1,22)=18.90, p<.001) indicating that text messages required a greater number of key presses than phone calls. There was no effect of phone (F(1,22)=.09, p=.771). However, there was an interaction between task and phone (F(1,22)=13.30, p=.001). Post-hoc t-tests, with p-values adjusted for familywise error rates, showed that for Motorola phones, phone calls could be made with significantly fewer keystrokes than text messages (t(11)=6.65, p=.004). Furthermore, text messaging could be completed in significantly fewer keystrokes using the Nokia phone over the Motorola phone (t(22)=3.25, p=.016). None of the other comparisons reached significance (p=.184 or greater).