The study found that the results in a previous research paper are influenced by how the model is set up. By using different methods to account for initial conditions and censoring in the data, the researchers were able to get a more accurate estimate of how a woman's decision to work is affected by her past employment history. The new approach provided a clearer understanding of the factors influencing married women's labor supply.