There are many examples of context effects in survey measurement. Context can be shaped by the order of questions, the format of response options, manipulation of the survey environment, and so on. For Web surveys, the inclusion of visual images is a trivial design issue, but may have consequences for the responses obtained. Understanding the role that images may play in shaping respondents' answers is an important step in designing effective online surveys. We have conducted a series of experiments examining how context may be shaped by the use of images in Web surveys. For example, we examined the effect that pictures of a healthy woman exercising versus a sick woman in a hospital bed may have on responses to a question on self-rated health. In an effort to understand how pictures may influence respondent's answers, we have replicated the experiments in three different surveys, varying such factors as the size of the image (large versus small), the placement of the image (in the header versus alongside the question versus on a prior screen), and the position of the question (first question versus later in the survey). We have also conducted experiments on questions of mood or depression, and on behavioral questions such as travel. We report on the results of these experiments, and discuss the implications for Web survey design.