This OSF project hosts all relevant files for *Science's* surveys of scientist harassment and intimidation, which accompany the March 25, 2022 story "[In the Line of Fire][1]". We conducted two surveys: 1. The "Web of Science" (WoS) survey pulled corresponding authors' email addresses from COVID-19 papers from the Web of Science. Materials related to this survey have the keyword "wos". 2. The AAAS survey contacted members of the American Association for the Advancement of Science, the publisher of *Science*. Materials related to this survey have the keyword "aaas". A brief guide to the files: 1. The [IRB protocol][2] describes our recruitment, debrief, and data management procedures. 2. Consent forms for both the [AAAS][3] and [WoS][4] surveys. 3. Survey materials for both the [AAAS][5] and [WoS][6] surveys. The surveys were conducted in Alchemer, which allows a .doc print-out of the questions, provided here as PDFs. However, these print-outs do not show the survey logic (for example, questions shown only to participants who responded "yes" to a previous question). This logic is described in the survey write-ups. 4. R and Rmd scripts. "[Deidentification][7]" exports any email addresses supplied by respondents, removes the details, and exports deidentified datasets (both surveys). "[Wrangling][8]" wrangles these deidentified datasets, and exports clean datasets (both surveys). The two analysis scripts (separate R Markdown files for [AAAS][9] and [WoS][10] surveys) analyse the clean datasets; generate summary data tables and plots for all questions; export 'processed' versions of the datasets with all composite scores and other measures; and export anonymised datasets for publication. "[Identification][11]" is a short script that takes the processed datasets and examines individual responses where there were questions about their veracity or reliability. 5. Survey write-ups for [AAAS][12] and [WoS][13] surveys. These are HTML files (the outputs of the Markdown analysis files mentioned above), which can be downloaded and viewed in your browser. 6. Anonymised datasets for [AAAS][14] and [WoS][15] surveys. Given the sensitive subject, our data management procedure required the removal of all potentially identifying details before publishing the data. These details have been redacted, rather than columns removed, to improve reproducibility with the scripts supplied above. It will still be possible to generate many of the results we report, and to conduct new analyses, with these reduced datasets. [1]: http://www.science.org/content/article/overwhelmed-hate-covid-19-scientists-face-avalanche-abuse-survey-shows [2]: https://osf.io/rzp9t/ [3]: https://osf.io/tq9sv/ [4]: https://osf.io/huvcy/ [5]: https://osf.io/cbh7d/ [6]: https://osf.io/3tb7q/ [7]: https://osf.io/vwfh9/ [8]: https://osf.io/946tm/ [9]: https://osf.io/bdhws/ [10]: https://osf.io/54pfb/ [11]: https://osf.io/njav2/ [12]: https://osf.io/eg3xs/ [13]: https://osf.io/xatkg/ [14]: https://osf.io/y84f2/ [15]: https://osf.io/8upft/