Hypotheses: Teenagers who spend more time per day on social media sites such as Facebook are more likely to have lower self esteem than their peers who spend less time per day on social media
The research method I would use for this study is a correlational design, which would allow us to see to see if there is a positive or negative correlation between social media use and teenagers’ self esteem (Lilienfeld, Lynn, Namy, & Woolf, 2011.) A case study would not be of much use in this situation, because it would be difficult to apply the experiences of one or two teenagers to teens as a whole. An experimental design would allow us to infer causation, (Lilienfeld, et al. 2011) but would be difficult to control, assuming the teenagers are all using social media for different amounts of time each day. It would also be difficult to find a control group of teenagers who are not using social media at all- there may be other factors, such as socioeconomic status, that may effect both their self esteem and their access or lack of access to social media. It would also be difficult to use naturalistic observation for this study because it is difficult to measure self esteem through observation, so some self-reporting or surveying would be required.
I would survey a randomly selected group of students from different high schools across the United States, and ask them how much time per day they spent on social media, what they typically posted about on social media sites, and which social media platforms they preferred. The second part of the survey would have them rate themselves in a few categories using a scale from 1-10 to estimate their self esteem and confidence levels. The data would then be placed on a scatterplot to see whether the correlation is positive, negative, or zero.
It would likely be challenging to find subjects willing to self-report honestly about their self esteem levels. The subjects may engage in response sets in order to make themselves appear differently, depending on what results they think the researcher is interested in. (Lilienfeld, et al. 2011) Another challenge of this study would be the lack of ability to infer causation. We could find a correlation between high amounts of time spent on social media and low self esteem in teenagers, but that does not necessarily mean that one causes the other. There could be a third variable causing one or both variables to occur. (Lilienfeld, et al. 2011)