There are many types of research methods in social sciences. The key in determining which design is best depends on the specific question the research hopes to answer. Social networks and personality traits are developing areas of interest and would probably work best in a non-experimental design.
An experimental study is the traditional scientific model with a control group and treatment group. This design works well when testing a particular variable. The typical question is, "Are people affected by x?" A true experimental design will test a random sample. In a quasi-experimental design the sample is controlled by the researcher. This may be done to test a specific reaction at a target location rather than the generalized public.
A field survey is a non-experimental form of research gaining information through statistical modeling. This model is best suited to attempting to answer a complex relationship between two variables. For example, are people with personality type A more likely to be involved in social networking? Surveys can be sent to a random sample of people questioning their personality type and social networking habits.
Another research method similar to a survey is the focus group. A focus group is a small sample of people led through a discussion by the researcher. The information gleaned can be compared against other focus groups to complete the research project.
Secondary data analysis is a form of research where data already collected by a third party is reviewed for trends. A researcher might use this method to discover trends, patterns or anomalies in behavior such as if life events encourage online purchasing.
Action research and ethnography are two other research methods, but they focused on directed action within a narrow spectrum of people or culture. For research involving relationships between two or more variables, focus groups and surveys remain two of the best options.