Video Interaction Guidance (VIG) is an intervention that aims to improve communication and relationships. It is a video-based approach that involves a trained VIG professional filming, editing and reviewing film clips with a client, of their interactions with a significant other person. This thesis explores whether VIG can be used to improve pupils' participation in group work lessons, following a novel application of VIG in natural classroom settings. Study 1 aimed to determine whether there were improvements in video samples of observable classroom behaviour following VIG and to investigate participant experiences of the intervention. Participant evaluation of VIG, via focus groups, interviews and questionnaires, was positive. However, while Percentage of Data Points Exceeding the Mean (PEM) calculations of video data using a multiple baseline across participant small-N experimental design showed some degree of post-intervention changes, these were not statistically significant (p>.05) as indicated by Dugard and Todman's bootstrapped exact probability test (2011). Study 2 investigated whether the improvements reported by participants from Study 1 could be objectively observed by experienced professionals. A sample of 4 educational psychologists blind rated a random sample of pre- and post-intervention videos and identified post-intervention videos as evidencing significantly more effective examples of group work (p=.003). The criteria used in their clinical decision-making were used to inform a new video coding schedule. Re-coded observations from the sample of videos used in Study 1 were then analysed. Target pupils were found to be significantly more attentive and attuned to their peers after VIG (p=.05). While PEM scores of video data again indicated other post-intervention improvements, these were not statistically significant (p>.05). This study is unique in reporting objective, observable pupil behaviour change over the relatively short period of VIG intervention, with high levels of client satisfaction and acceptability. Methodological limitations and recommendations for future research are discussed together with key implications for practicing EPs.