I believe that the best answer to this question is A. Savannah would need to use a descriptive research design. Although what she would be measuring is a preference, she needs to make sure that her own biases will not play a role in how she conducts the study.
By using the descriptive type of research design, she will deal with actual numbers and facts in order to get some results. She would be able to present the data that she will gather in a more methodological manner. This would allow her to gain as many details as possible so that she can present her findings properly.
This is definitely descriptive by design since Savannah would most likely be conducting a survey where results upon which would require statistics such as frequency and percentages. This type of research design would want to gain more information pertinent to the phenomenon understudy. The information is presented via quantitative means