Analysis Plan

Singer (1989) suggested that divorcing couples equally share the differences in their incomes and for a period of half the length of the marriage. She also suggested that each party contribute equally to the cost of caring for children. However, Singer did not specify how this should be accomplished, if it should be done through the courts, or through the parents. To determine the award based on Singer’s formula, the difference between the pre-divorce gross incomes of the divorcing parties, as stated in the final decree, was divided in half and 50% of the income was allocated to each party.

No child support award was subtracted because Singer did not specify how this should be determined. Each party’s income was divided by the poverty level income guideline for a one-person household, so that child support deductions would not be subtracted from either household Independent Variables There were three independent variables: (a) gender of custodian, (b) length of marriage and (c) pre-divorce total gross income. Gender of custodian had two levels, male or female.

Length of marriage had four levels: 0-4. 9 years, 5-9. 9 years, 10-19. 9, and 20 plus years. Pre-divorce total gross income ranges were divided into 20% quantiles to produce five levels: $0-$43,020. 00; $43,020. 01-$55,236. 36; $55,236. 37-$68,466. 00; $68,466. 01- $91,178. 00; and $91,178. 01- $2,000,000. The current research was mainly exploratory because the alternative strategies for allocating income had not been evaluated empirically. There were three main research questions in the current study.

First, how equitable were the income-to-needs ratios of the actual court order and the alternative strategies and how do they compare to one another? Second, do the strategies provide adequate incomes for male and female-headed households based on the needs standard of poverty level income guidelines? Third, what effects do strategy, household/gender, length of marriage, and pre-divorce total gross income have on post-divorce income-to-needs ratios? A few relationships were expected based on previous literature.

Female-headed households were expected to have lower income-to-needs ratios than male-headed households (Bianchi, Subaiya, & Kahn, 1999; Peterson, 1996); and male and female-headed households in the lowest pre-divorce income category were expected to have more equivalent income-to-needs ratios than households in the highest pre-divorce income categories, where female-headed households were expected to have the greatest decline in income-to-needs ratios post-divorce (Arditti, 1997; Holden & Smock, 2001).

The first research question asked: How do the income-to-needs ratios based on alternative strategies, compare with the actual court order, and to one another? Plots of the mean male and female income-to-needs ratios, based on the different strategies, were used to make the comparisons. The second research question asked: Do the strategies provide adequate incomes for male and female-headed households, based on the needs standard of poverty level income guidelines?

A new variable was created for each strategy to code income adequacy as a one if a household’s post-divorce income-to-needs ratio was less than one (below poverty). Income adequacy was coded as a two if a household’s post-divorce income-to-needs ratio was equal to or greater than one, but less than two, a code of three indicated the ratio was equal to or greater than three, but less than four; a code of four was equal to or greater than four, but less than five.

Ratios were coded a five if equal to or greater than five, but less than six; and coded six if it was equal to or greater than six, but less than seven. A code of seven indicated the ratio was equal to or greater than seven, but less than ten; and an eight indicated it was equal to or greater than ten, but less than one hundred. Frequencies were generated for each strategy to evaluate the percentage of male and female households’ post-divorce income-to-needs ratios using each strategy that were below, above, or at poverty level incomes.