Unit Analysis Of The 2016 General Social Survey

928 Words4 Pages
Data were extracted from the 2016 General Social Survey (GSS 2016) – “a repeated cross-sectional survey of a nationally representative sample of non-institutionalized adults who speak either English or Spanish” in the United States (National Science Foundation, 2007, P11). The unit analysis for GSS 2016 was individual. The GSS sample was drawn from an area probability design that randomly selects respondents in households across the country, which includes a mix of urban, suburban and rural geographic area (National Opinion Research Center, 2016, NP). Participation was strictly voluntary, but the response rate for the GSS was very high, which was over 70 percent (National Science Foundation, 2007, P11). The GSS data was collected from…show more content…
Key Independent variable
The key independent variable, marital status with four categories, provided the information about the respondent’s marital status. This variable was originally measured as a nominal variable labeled as “marital status” with five categories: 1. Married; 2. Widowed; 3. Divorced; 4. Separated; 5. Never married, as well as the missing value: 9. No answer. I recoded this variable into a nominal variable with four categories: 1. Married; 2. Widowed; 3. Divorced or separated and 4. Never Married, as well as the missing value: 9. No answer. As literature showed that referring to mental health, differences between the widowed and the divorced/separated mattered (J. K. Trivedi et al. 2009, NP). So, for this research, the widowed were left separately, while the divorced and separated were combined.
Risk and Prevention Factor
The first control variable, people the respondent knows suicide over lifetime, was measured as a nominal variable in the data set, but has a quantitative type of categories, ranging from 0 to 10, as well as 20, 21,50, 70, 95: 1 or more, number unknown, with missing values: 98 Don’t know, 99. No answer. I recoded this variable into a nominal variable with three categories: 0. None, 1. Few and 2. More.
Coding for all categorical variables was shown in Table
Get Access