Nonetheless, it is important that all findings be considered in light of the numerous limitations of this study. One of the main problems encountered with this study was the ceiling effect caused by numerous high scores on the pre-test. Over 75% of the original 61 participants scored a 17 or higher on the pre-test (out of a possible 20). This left little room for visible improvement. It may be possible that, with a more sensitive pre-test, differences could have been found.
Another limitation was the time on task for the training. Because the activities were repetitive, the length of each training recording was quite short, lasting approximately 13 minutes for each recording. By incorporating more training activities and more training time, it is possible that differences could have been found. Also, although this study began with 61 participants, data from 20 participants could not be used to analyze improvement from the pre-test to the two post-tests. With only 41 subjects, the generalizability of the results is somewhat questionable. Similar future research is necessary to confirm these findings.
Despite the limitations of this study, the results suggest that this matter of whether to use audio or video training is not conclusively decided. Future research should not only aim to replicate these findings, but also expand the scope of inquiry to include more pronunciation features. Thus far, only one English vowel pairing, /i/-/ɪ/, and one English consonant pairing, /l/-/r/, has been investigated [citation?]. Yet there are many other pairings that have clear differences in visual cues, such as /ʌ/ vs. /ɔ/ or /θ/ vs. /t/ that could add to the understanding of the impact of the visual cues.
Because there has previously been relatively little interest in this area of research, there are many possible directions to go with this type of research. Although researchers have recently started to turn their attention to the effects of audio vs. video training for teaching pronunciation, many questions are left unanswered. As teachers are already employing these modes of delivery in their classrooms and as homework, it is important that further research be conducted to determine the effects of these two methods of training delivery.