The findings of this study demonstrated that the childcare students in the CAL package group had a significantly higher retention performance score than those in the leaflet group (p = 0.02, 95% CI = 0.83–12.19, effect size = 0.8). A possible explanation for this finding is that the presentation of more complex information about autism via the visual (images, diagrams, video and animation) and auditory (sounds and narration) channels in the CAL package may increase the efficiency of the participants' working memory. Information entering through both channels simultaneously; for example with the CAL package, can reduce the cognitive load, and resulted in improved learning efficiency [27
]. Moreover, animations, narration, pictures and sound, if appropriately used in the CAL package, can effectively establish mood, persuasion and explication to the learners [30
Furthermore, students who watched the CAL package performed better in the last two sections of the questionnaire, compared with the leaflet group. This may be due to the ability of the CAL package to hold learners' attention longer compared with reading the leaflet [32
]. The use of various presentation styles in the CAL package, e.g. different colours, many text sizes, changing pictures or background, animations and video, may enable learners to maintain their attention better than reading from the printed text [33
]. Although, the study found small difference in the retention performance between the CAL package and leaflet groups (p value = 0.02), effect size was relatively high (an effect size of 0.2 is described as "small", 0.5 as "medium" and 0.8 as "large" [34
]). However, due to relatively small numbers, caution needs to be taken in interpreting the results. In addition, randomization by a sequence of random numbers may result in selection bias and may lead to imbalance in groups.
In this study, the CAL package group demonstrated a greater level of enjoyment than the leaflet group (p = 0.04, 95% CI = 0.03–2.75, effect size = 0.7). It has face validity that this increased level of enjoyment improved the participants' motivation, which may enhance working memory. In a previous study where level of enjoyment and information recall ability in standard HTML-driven Web sites (predominantly text) users and animated Flash-enhanced sites (integrated animations, moving texts, graphics, rollover effects and sounds; provided the user with a potentially more engaging, immersive, and entertaining experience) users were compared. The Flash presentations were more enjoyable than the text presentations, and the majority preferred using the Flash presentations. In addition, participants who used Flash-enhanced sites were significantly more able to recall information than participants who used HTML sites [35
The effectiveness of the CAL package was also evaluated by measuring the level of confidence to identify a child as autism. Participants who read the information leaflet had higher confidence in identifying a child with autism than students who watched the CAL package, but the difference did not reach statistical significance (p = 0.39, effect size = -0.3). More than 80% of participants (88% in the CAL package group and 81% in the leaflet group) had seen 5 or less children with autism. However, 6.25% of students in the leaflet group had seen more than 10 children with autism, whereas none of the students in the CAL package group had seen more than 10 children with autism. This finding is a possible explanation of why the CAL package group demonstrated an improved knowledge and enjoyment, but did not show confidence in identifying a child as autism. The findings from the study suggested that, apart from knowledge, experience in working with children with autism is a key factor in determining the level of confidence to identify a child as autism. This is supported by the findings of a recent study in which teachers who had experience with children with autism demonstrated significant greater confidence than teachers who had no or little experience in identifying children as autism [36
Although, this study suggested a superior effect of the CAL package over the leaflet, there were some limitations, which may affect the validity of the study. Firstly, there was no validity test of the questionnaire. Validation of the questionnaire would have increased the validity of the study. Secondly, in this study, the participants' knowledge about autism prior to interventions was not evaluated. Pre-interventional knowledge would be useful in establishing a baseline level of knowledge of the participants prior to the intervention, thus, an improvement observed in the post-intervention scores could be due to the intervention, rather than due to baseline variations between the groups. Lastly, the study evaluated only enjoyment and acquired knowledge (level 1 and 2 of the Kirkpatrick's hierarchy model), we did not evaluate transfer of knowledge to the workplace and benefits to children with autism as a direct result of the CAL package.