Quanics is correct. I would never recommend paying your own way at a graduate program--go where you are funded. A graduate stipend will net to about $20k/year (there are some amazing fellowships that will fund $50k+ but they are very, very competitive). In a "real job" you'll often be making much more than $20k, as I was when I was a higher ed IT professional (and could have qualified for free tuition as an institutional employee). PhD programs are rated by several factors including graduation rate and time to completion. It is obvious to me now that the people who are choosing the students to fill the very limited number of graduate student seats in any given year will choose those who are committing to work full time on graduating.
Many people do complain about academia being hostile to family life (see http://100rsns.blogspot.com/2010/09/15-marriage-and-family-u... for a particularly negative view). However, many of my colleagues had children during graduate school. They took time off classes but then used their "down time" (e.g. child napping) to focus on writing and came out ahead. I got married immediately before entering grad school and had my first child as I was writing my dissertation. The timing worked really well for me. I was lucky that my wife could support me through graduate school (on top of my stipend).
Many people do complain about academia being hostile to family life (see http://100rsns.blogspot.com/2010/09/15-marriage-and-family-u... for a particularly negative view). However, many of my colleagues had children during graduate school. They took time off classes but then used their "down time" (e.g. child napping) to focus on writing and came out ahead. I got married immediately before entering grad school and had my first child as I was writing my dissertation. The timing worked really well for me. I was lucky that my wife could support me through graduate school (on top of my stipend).