Social Sciences Proposal
Field(s) of study or investigation: social science, social attitudes and views
Describe in a few sentences what you want to do in your independent study or project:
I want to explore public understanding of weight and report on any dissonance between public expectation and actual weight. My hypothesis is that in media nowadays the prevalence of Photoshop has created a warped understanding of what people of a certain weight actually look like, and I will test this by collecting many images of people of various verified weights and asking a large body of individuals to guess their weights. I will then report on these findings in a formal lab report, complete with appropriate graphs and margins of error.
How did you become interested in this topic?
I have had many friends in high school who complain that they're too fat and need to lose weight. They always use numbers when arguing their case -- "I need to lose five pounds", "I need to be 120lbs by summer", etc. The same goes for many on the internet, e.g advertisements promising to let you drop 10lbs a week and the ulzzang community, whose members aim to keep their weight below ~110lbs (50kg). People put such importance in these numbers, but I feel like they are just repeating some "ideal numbers" they heard from somewhere else. If someone came up to you and said they were 115lbs, would you know if they were lying? Given a photo of a stranger wearing non-deceptive clothing, would you be able to correctly guess their weight?
What past experience do you have in this subject area?
Since realizing that regardless of height or body type, my friends all seem to quote the same "ideal" weight ranges, I've been reading about health and weight and the use of Photoshop in popular media. I've also tried finding real, unaltered images of people online to train myself to see weight properly. The internet has been somewhat helpful but in most cases the weight and height data is unverified, hence me wanting to do this project -- both to personally verify weight and visual correlations and see for myself what the dissonance between perception of weight and reality is.
What steps might you take to help you reach your goal?
I will first need to collect images of people of various heights and weights wearing non-deceptive clothing, i.e no shapewear and/or clothing designed to make them appear taller, shorter, thinner, curvier, etc. For each image I will keep data regarding their height and true weight, then label them by number and randomly select a subset of them to show each individual person taking the survey. Each person taking the survey will have an empty text box to try and guess what the weight of the person in the corresponding image is. I will allow for some margin of error, e.g if someone puts down 141 and the person's actual weight is 140 it will count as "correct". Afterwards I will calculate the percentage of correct to total guesses and write up a lab report on my findings.
What kind of guidance do you seek from a mentor?
Since I haven't taken any statistics class yet, I would need a mentor who knows how to decide whether or not something is statistically significant, i.e if a certain percentage of people are guessing wrong, is it enough for me to claim that people don't understand weight? I would probably also need help finding enough volunteers for both the image data and to take the surveys so that the report won't be a huge generalization on a tiny pool of fifteen or so responses.
What are some challenges that may arise as you carry out this independent study or project? How might you mitigate these challenges?
It will likely be difficult to find volunteers for the image data. I can try reaching out to some of my friends and or Facebook but will probably need my mentor's help with this issue. Finding volunteers to take the survey may be a bit easier, since more people may be willing to "test their skill" at weight guessing, and I can also presumably ask my fellow Junction students and co. to take the survey.
Last modified on April 15, 2015 at 09:38 p.m.