Incarceration In My Family

Data
The data used in this project came from the Returning Home dataset, collected between 2004 and 2005, by the Urban Institute .The data are comprised of both males and females who were incarcerated in the state of Texas with a total sample size of 676 individuals at Wave 1(during incarceration). About 30 days before release, respondents completed a self-administered survey regarding a host of measures assessing pre incarceration dimensions encompassing family life, employment, and experiences outside of prison. Respondents were also asked about life during incarceration including experiences such as various activities and programs they may have participated in, family contact, and disciplinary issues. Approximately 2 to 5 months after release,
Respondents were asked a series of questions about their interactions, connections, and contact with their family. These 12 items asked respondents to choose from a 4-item scale (strongly disagree, disagree, agree, and strongly agree) indicating how much they agreed with the following statements:
• My family was a source of support for me.
• I had someone in my family I could count on to listen to me when I needed to talk.
• I had someone in my family to have a good time with.
• I had someone in my family to talk to about myself or my problems.
• I had someone in my family whose advice I really wanted.
• I had someone in my family to share my most private worries and fears with.
• I had someone in my family to turn to for suggestions about how to deal with a personal problem.
• I had someone in my family who understood my problems.
• I had someone in my family to love me and make me feel wanted.
• I had someone in my family to get together with to relax.
• I had someone in my family to do something enjoyable
At Wave 2, approximately 2 to 5 months after release, respondents were asked the same set of questions assessing their family relationships after release. Researchers explored whether the items represented a single dimension, or multiple dimensions, through the use of a rotated factor analysis. A rotated factor analysis was used to extract a set of factors from a data set, in which individual covariates were loaded together in a manner in which maximum variance was explained. Researchers used .5 as the threshold for construct validity and observed two distinct factors. Researchers also noted that in addition to using .5 as the cut off for the loading threshold, construct validity could be further observed through the eigenvalues as both factors report eigenvalues above 3.0. The relatively high eigenvalues for each factor (much higher than 1.0) suggested the total explained variance of the combined measures was sufficient to be used as a single factor. The first factor captured questions demonstrating that the respondent wanted to, or spent time with, family members, which researchers labeled familial interactions. The second factor captured dimensions indicating the respondent believed he or she could rely on family for advice and support, which researchers labeled familial emotional