Bubbles bursting: Investigating and measuring the personalisation of social media searches

•This study presents an open, extensible, and reproducible framework for investigating the implicit factors that affect personalisation in social media searches, which are not explicitly stated or visible to users, and measuring the extent of personalisation in the search results affected by these factors.•The results of a comprehensive set of experiments on Twitter carried out in this study suggest that the hypothesised factors of followees, cookies, and carry-over effect have limited impact on personalisation, with the exception of polarised topics, where the experiment revealed a notable one-sided bias in the search results.•The article presents a range of mitigation strategies to address the negative consequences of personalisation in social media search results, in line with existing research works. These include promoting greater transparency of algorithmic personlisation mechanisms for social media search results, empowering users to control the personalisation settings of their search results, encouraging healthy online search behaviours such as deliberately seeking diverse perspectives and regularly clearing internet cookies and social media search histories, and advocating for policymakers to develop policies to promote algorithmic awareness and protect user data privacy.