487 Assignment 4 - Methods sample

.pdf

School

Ohio State University *

*We aren’t endorsed by this school

Course

3487

Subject

Sociology

Date

Apr 3, 2024

Type

pdf

Pages

7

Uploaded by taaliyahpalmer

Report
Olivia Li Professor Lynette Martin Sociology Research Methods 30 October 2023 Method and Variables In effort to examine the relationship between how rural/urban factors (independent variables) affect the development of eating disorders (dependent variable), this paper will follow through with a questionnaire survey, within Ohio specifically. The breakdown of the independent is as follows: distance to closest metropolitan city, distance to closest healthcare clinic, difficulty securing food, income level, and social media use. Therefore, a multivariate analysis will be used. Questionnaires will be sent through mail for better response rate and best accessibility for respondents (ie- email or online can pose differences in the rural and urban populations). The benefits of using this survey type includes saving money, questionnaires can be completed at convenience of the respondents, no interviewer bias, and greater anonymity and confidentiality. Because minors between ages 14-17 are eligible to complete the questionnaires, an IRB process may needs to be put in place to grant permission from parents and to make sure that the survey is done ethically. Question Descriptions The aim is to collect data on rural and urban demographics and whether one has traits of an eating disorder. Respondents will answer questions on their rural/urban residence as their distance from the closest metropolitan city from their home, food-security status, as well as their distance to the closest hospital. For the development of eating disorders, the Eating Disorder Examination – Questionnaire (EDE-Q) will be used for a composite score (Faiburn, Beglin, 2008). This questionnaire has 28-item self report questions that assess: range, frequency, and severity of an eating disorder. Questions 1-12 are based on an ordinal scale that asks respondents about specific behaviors. These questions are mapped for numerical responses; (NO DAYS, 1-5 DAYS, 6-12 DAYS, 13-15 DAYS, 16-22 DAYS, 23-27 DAYS, EVERY DAY) with ‘NO DAYS’=0. Questions 13-18 is open-ended and asks respondents to answer how many days they participated in certain behaviors. This is on an interval scale. Questions 19-21 also asks about the number of days they engaged in behaviors on an ordinal scale. Responses in this section are also
mapped; (NO DAYS, 1-5 DAYS, 6-12 DAYS, 13-15 DAYS, 16-22 DAYS, 23-27 DAYS, EVERY DAY) with ‘NO DAYS’=0. Finally, questions 22-28 asks respondents about the degree to which certain behaviors resonate with an individual (NOT AT ALL, SLIGHTLY, MODERATELY, MARKEDLY). This is also mapped with ‘NOT AT ALL’=0. Operationalization To measure the assessment of eating disorders within the study, a global score will be used for data analysis. Results are categorized into four main categories: Restraint, Eating Concern, Shape Concern and Weight Concern along with a global score. This is a valid measure because it has been proven to be consistent for both women and men within a study with a high test-retest reliability (Rose, et al., 2013). The EDE-Q was also convergently valid with a more recent measure of eating disorders: ED-15 (Accurso & Waller, 2020). To measure rural/urban status more specifically, distance (in miles) will be considered from a respondent’s home. This includes measuring the distance to both the closest metropolitan city and the nearest clinic. Most rural areas are far from metropolitan areas and most clinics are located within cities. Using distance to the city will allow us to determine a more face valid measure of rurality. The distance to the nearest clinic will allow us to also determine whether one has readily access to healthcare, specifically for seeking eating disorder help from professionals or referrals from primary care doctors. Then, we will also be using how often one felt food-insecure within the past 28 days to measure the access one has to a variety of food that is important to having a healthy relationship with food. Rural areas disproportionally face the challenges of food insecurity and accessibility (Sharkey, 2009). Because of limited access, diverse, and healthy foods, rural areas are at greater risk of nutrition related disease (Sharkey, 2009). This question was included to see if this factor also can affect the development of eating disorders. This question was also set to the past 28 days to match that of the EDE-Q. Control Variables Women of younger age are most likely to develop an eating disorder (Rohde, P., et al., 2017). Therefore, age and gender will be controlled for within the respondents. Sampling Method
To sample our final unit of analysis cases, we will use multi-stage cluster sampling. The first stage will include dividing Ohio into regions (N, S, E, W). This will be our primary sampling unit. Then within the regions, counties will be randomly selected. Next, the sample will be selected by city/town. Finally, a random sample of houses will be selected to mail surveys out to. Therefore the sampling process will follow: one primary, two secondary, and one final stage. Note however, that during the random selection of counties selected will be repeated twice, one from a list of rural counties within a given region and another from a list of urban counties within a given region. Thus, we will need a database of the rural/urban counties and cities/towns. The sampling process will also follow a probability sampling method. The regions, counties, and cities/towns selected will be randomly-generated. Because respondents are outside of Ohio State University, they will be contacted through mail and ask for consent within the mail to be part of the survey. The reason for this method of sampling was to ensure that every part of Ohio is represented and has an independent, random chance of being selected. This also will be easier than generating a non-probability sampling method. However, a potential disadvantage is that rural areas may represent minor areas of Ohio and vice versa for urban areas. Therefore, there could be a chance of having more of one type of area represented in the survey than the other. Hypothesis H1: Rural areas may experience have more preoccupation with food than urban areas (food insecurity) H2: People living in rural areas may experience greater severity of eating disorders than urban areas (lower access to healthcare services) H3: People in urban areas may experience more weight stigma by others (higher weight status in rural areas) H4: For those who use social media, the experience of body dissatisfaction is similar
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help