The popularity of social network sites has given rise to concerns about the amount and nature of information that people disclose on them. While the media has publicized several high-profile instances of people losing their jobs because they had disclosed inappropriate job-related, empirical research about the use of social network sites by job recruiters is almost non-existing. Social network sites are nowadays tools for recruiters to screen job applicants during the job application procedure. This study examined what specific information on social network sites recruiters use to form an impression of job applicants and this impression affects applicants’ career opportunities. Based on the warranting principle pertaining to impression formation in Internet communication, judgments about a target rely more heavily on information derived from others than information provided by the target self. The warranting principle has been tested in many situations, but not yet on the impact on recruiters when using social network sites for hiring job applicants. Therefore, two studies were conducted to test if the warranting principle also applies to use of social network sites by recruiters. As the warranting principle had not yet tested in relation to recruiters’ use of social network sites, the first study (N = 11) had an explorative, qualitative nature to gain insight into how job recruiters use social network sites in the job application process. Eleven recruiters from different companies of different sizes were interviewed in-depth about their use of social network sites and about how the specific content of the profile pages of applicants contributed to the end decision of hiring a job applicant. The results of the first study revealed that about half of the recruiters use social network sites during the application procedure.. According to this first study, recruiters who do use social network sites do so in order to form an impression of the job applications or out of curiosity about them. Recruiters who do not use social network sites indicate that fear of misjudgement and unreliable information are reasons not to use them. The results also revealed that information generated by others (i.e., not by the profile owner) on social network sites, such as comments by others, were not perceived as important information. On the contrary, other people’s comments about the job applicant were perceived as information a job applicant could not control, and therefore, irrelevant and not more trustworthy. Rather, information from job applicants themselves was perceived as much more important such as information about hobbies, interests and posted photographs of job applicants. The second study aimed to better understand the way recruiters used the profile pages of job applicants and to test whether other-generated information did play a role in the job applicant process. It was quantitative in nature, involving an online survey and consisting of a 2 x 2 factorial design. One variable was self-generated information (positive or negative) and the other was other-generated information (positive or negative). Four different Hyves profiles were created and these differed in the way information was given on these profiles. The particular social network site, Hyves, was chosen because it is the most popular one in the Netherlands and was revealed in the first study as the most popular one for recruiters. The Hyves profiles consisted of self-generated information (information from the profile owner self) and other-generated information (information derived from others). The profiles were manipulated and turned into four different versions of the questionnaire in order to test the warranting principle. A total of 159 recruiters filled in the questionnaire of whom 74 were men and 85 were women. Chi-square tests showed that the respondents were equally distributed among the experimental conditions both for gender, age, education and income. The respondents worked for a diverse range of branches of industry such as service organizations, consultancy firms and educational institutes. The results of the survey showed that information derived from others was not perceived as more trustworthy than information provide by job applicants themselves. Thus, the results of both the first study (interviews) and the second study (survey) indicate that the warranting principle was not supported. Recruiters themselves do not indicate that other-generated information is more important and trustworthy than self-reported information, but when tested experimentally, the results show that recruiters do rely more on information derived from others than on self-reported information. This research confirms the popular suspicion that social network sites influence job applicants’ career opportunities, but also goes further to demonstrate what kind of information on these sites influences recruiters. The findings have theoretical implications for the warranting principle and practical implications for both recruiters and job applicants which are described in more detail in the paper.