Poster #12 - Are the grandkids all right? Using secondary analysis for hard-to-find participants

• In Event: 3-001 - Poster Session 2 during Breakfast

Abstract

This poster presents issues raised when secondary analysis was used to study adult outcomes of being raised by grandmothers. Cycle 20 of the General Social Survey collected by Statistics Canada (2006) was used.

Secondary analysis allowed study of a sample that is rare. In Canada, only 4.8% of children lived with a grandparent (Statistics Canada, 2011). Yet, this is an important population to study because the percentage of children in this situation is rising and literature suggests that being raised by grandparents puts children at risk for difficulties.

The 76 adults raised by grandmothers had a larger social network than the randomly selected comparison group, t(128) = 7.67, p < .001, but social network did not moderate their life satisfaction. Given that they both received and gave out less help than the comparison group [t(141) = –3.35, p < .001 for help received; t(147) = –6.2, p < .001 for help given], they could be encouraged to utilize their network to greater effect.

Adults raised by grandmothers had lower life satisfaction than the comparison group, t(123) = –2.96, p = .004, but those with higher income had higher life satisfaction, β = .361, p < .05, suggesting income can help compensate for negative experiences.

However several issues arose. Deciding upon a database posed challenges because there are a large number to choose from, not all are available to all researchers, it can be difficult to learn the codes of the database, and not all contain information that would answer the research question. Furthermore, databases can go out of date quickly.

In hindsight, propensity score matching (Dehejia & Sadek, 2002) may have been a better idea than random selection of a comparison group to ensure that the groups were similar. Our samples differed on number of visible minorities and aboriginals, level of education, and income.

Finally, we were limited in the outcome variables with which to compare groups. We found no group differences in health and mobility limitations, which suggests that although the odds are stacked against them due to their nontraditional upbringing, there are areas in which family structure does not make a difference in adult outcomes. It would have been nice to have access to measures such as quality of success in occupation, marriage, or as an active member of the community. Despite limitations of secondary analysis, we were able to gain insight into this hard-to-find but important population.

Authors

• Cheryl Kier, Athabasca University
  Presenting Author

• Tak Fung, University of Calgary, Alberta, Canada