preview

National Retirement Survey

Decent Essays

Methodology:
The data were extracted from the national longitudinal survey, Health and Retirement Study (HRS), which was started from 1992 by studying the population of age range 51-61 and their spouses through re-interviews every two years. Later, the study population was extended to the population above 51 years, resulting to 11 waves at present which was collected in RAND HRS file. The purpose of the health and retirement study was to serve as an affluent resource for the information related health, social well-being and economics of the middle aged population who were followed over the years. Furthermore, HRS included stratified multistage area probability which incorporates stratification, clustering and differential selection probabilities. HRS used mixed-mode for data collection: telephone and in-person interviews. The sample size of HRS for …show more content…

Data were recoded and standard errors were adjusted for the complex survey design and data. List-wise deletion was done to adjust missing respondent data which resulted that the valid number of respondents in this study was 4190. The sample descriptions were obtained by using a univariate statistics; gender (M=1646, F=2553), with at least one disability in one of the activities mentioned above (n=2582, M=896, F=1686), having at least one of the disease condition (n=693, M=281, F=412) and retired (n=1631, M=630, F= 1001). The correlation was calculated for the dependent variable and predictors to indicate the strength of each predictors. A multi-regression model fitted for the predictors while controlling others was used to examine the effects of predictors on alcohol consumption. The possible problem of multicollinearity with the model was tested by variation inflation factors (VIFs). The residual plots could not be examined in this analysis for the independent variable is a count variable and would have been explained better using Poisson's

Get Access