Introduction: Hedges and boosters are terms respectively used to decrease and increase the strength of statements. They are therefore essential lexical tools in scientific communication, in particular in fields conducting human-subjects experiments such as Human-Computer Interaction (HCI). Objectives: We present an analysis of the use of hedges and boosters in the proceedings of the main HCI conference, namely CHI, between 2010 and 2018 to better understand how CHI authors report empirical findings. Methods: We only considered papers reporting a user study and focused our analysis on the sentences in the abstract that describe empirical results. We used a program to detect boosters and hedges and manually adjusted their classifications. Results: We found that CHI studies reporting inferential statistics are more likely to boost than they hedge. Conclusion: Our work intends to raise awareness within the HCI community of the importance of hedging and boosting when reporting research results. Further, our results establish a baseline of the current use of these linguistic devices in HCI, adding to the current body of linguistic research. We finally contribute tools, procedures, and materials to facilitate future analyses of hedges and boosters in scientific communication. Reproducibility: All data and scripts are available on https://osf.io/kefmr/