Many respondents complete web surveys using smartphones, and this number increases every year. There has been a great deal of research on the impact of device and screen size on response quality. Many surveys now use a responsive web design to provide a user-friendly interface to respondents, regardless of their device. There are many web frameworks that have been incorporated into surveys to implement responsive web design. Importantly, the choice of framework has large implications on the experience of respondents, from layout and design to functionality of input controls. Each framework has strengths and weaknesses on each device, and these lead can lead to varying respondent behavior and response quality. Two of the most popular frameworks are Twitter Bootstrap and jQuery Mobile. To our knowledge, previous research has not compared different frameworks.In this study we investigate the differences between Bootstrap and jQuery Mobile. To do this we use the RAND American Life Panel, a nationally representative probability based panel that participates in regular internet surveys. For one survey, we randomly assigned respondents to a version of the survey implemented using either Bootstrap or jQuery Mobile. The survey has varying question formats and controls, such as radio buttons, checkboxes, and text areas. These are rendered differently using the two frameworks. We analyze the impact of these two frameworks on response quality and respondent behavior across devices.