6.1.4. Interim discussion
Contrary to our expectations, the manipulation checks of the two aesthetics conditions and the usability condition revealed that our manipulations were unsuccessful. This was an unexpected result as a thorough pre-study was conducted in order to find color designs that would be perceived as high vs. low in either classical or expressive aesthetics. This pre-study compared 32 designs that were tested with a large sample size (N = 173). The result for the usability manipulation was even more surprising. In the low usability condition, each of the 60 pages of the learning material needed an extra 8.5 s to load compared to the high usability condi- tion. However, the perceived usability of the learning material did not differ between those two experimental conditions. While a detailed discussion of these results will be given in the general dis- cussion section, the next steps of our analyses need to be discussed in light of the failed manipulation checks. Without successful manipulation checks, we lack the basis for the between-group comparisons that would have been the first intended method of analysis. Our manipulation checks were based on perceived aes- thetics and usability. One could therefore argue that the differ- ences – at least for the usability manipulation – are nevertheless objectively given. We therefore decided to report the results of the between-group comparisons in order to thoroughly present and discuss the data. However, the results need to be interpreted with the failed manipulation checks in mind.