Analysis Of Variance

1822 Words 8 Pages
Repeated Measures Analysis of Variance (RM-ANOVA) was conducted to compare the reaction time of visual and acoustic conditions (previously shown in study phase) with new brands (not presented before)(Graph 1). Mauchly’s test showed that the assumption of sphericity had not been violated, χ2 (2) = 0.68, p = .711. Degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity ( = .98). There was a significant effect of modality type, F (2, 60) = 20.09, p < .001. Bonferroni Post-hoc tests displayed significant effect for visual versus acoustic; the visual brands were faster than acoustic (p < .001). On the other hand, Mixed ANOVA used to compare the performance of gender and it showed that there was not a significant effect …show more content…
The old/new with modality effect had maximal positivity during the 300-500ms time range and FN400 was negative-going waveform (Figure 2-3). Paired samples t-tests were conducted to see if modality has an effect on the retrieval at Fz, FCz, F1 and FC2 (300-500ms). After analysis of the data found that there visual and acoustic brands did not differ according to voltage at Fz amplitude t(30) = 1.29, p = .206 and at FCz t(30) = 1.26, p = 216 (Figure 2, Graph 2). Additionally, the left and right lateralization was examined with F1 and FC2 between 300-500ms. Interestingly, the mean amplitude of Graph 2 shows slightly a significant change between modality, the paired samples t-test provided no significant effect except FC2. The source memory over FC2 was the most positive waveform of FN400 than others (Graph 2). Thus, it has been found that acoustic brands had a larger negative amplitude than visual brands at FC2, even though the amplitudes not seen in this Graph 2 due to more positivity, paired samples t-test found a significant difference t(30) = 2.31, p = .028, but not at F1 t(30) =1.48, p = .153 (Figure 2, Graph 2). As a conclusion, the modality effect could not determine according to peak amplitude except FC2 which visual brands had larger positive peak than …show more content…
Therefore, all the chosen electrodes reported significant latency difference based on old/new brands. The result suggested that old brands elicited significantly shorter latency than new brands at F1 t(200) = 73.39, p < .001, FCz t(200) = 69.92 p < .001, FC2 t(200) = 58.40, p < .001 and at Fz t(200) = 66.84, p < .001. Another paired samples t-tests provided significant effect of modality type, but the latency changes were small magnitude over the Graph 3. In the FN400 component, acoustic brands had slower latency than visual at F1 t(200) = -7.25, p < .001 and at FC2 t(200) = -6.24, p < .001 but at FCz acoustic brands were slightly faster than visual t(200) = 3.26, p = .001 and there was no modality difference in latency at Fz t(200) = -.81, p =.418 (Graph 3). In general, new brands were slower than old brands, also in two electrodes (FC2 & F1) had faster latency for visual than acoustic

Related Documents