This outcome study, conducted during the 1960s, followed up on 100 adoptions arranged by four New York agencies decades earlier, between 1931 and 1940. The study is interesting not only because it surveyed very long-term outcomes, but because thinking about adoption had changed so much in the intervening period. For example, the researchers were preoccupied with factors such as infertility and the unique struggles that adopters faced in feeling entitled to their children. This understandably incorporated the emphasis on difference that was featured in recent works of adoption theory, like Shared Fate, and in therapeutic perspectives on adoption that reached their peak after midcentury. Yet when the adoptions studied here were initially arranged, during the 1930s, infertility and its psychological consequences had been minor considerations, if they had been considerations at all.

Jaffee and Fanshel nevertheless set out to explore and measure adults’ “ability to undertake parental role obligations without neurotic conflict.” They hypothesized that attitudes toward child-rearing were a sensitive barometer of parental psychology and would be strongly correlated with outcomes. Stricter parents (who resorted to spanking, for instance, and rarely left their children alone) were less able to handle separation. This illustrated that they had not overcome the handicaps of adoptive parenthood. On the other hand, parents who took more risks and allowed their children more freedom (by hiring more babysitters, for instance) had vanquished “the psychological insult” of infertility. They had achieved authentic parenthood.

Contrary to their hypothesis, the data that Fanshel and Jaffee gathered suggested that the degree of strictness had little to do with adoption outcomes. Money rather than psychology determined the extent of substitute care. The study utilized extensive tape-recorded interviews that were then coded and manipulated electronically. The researchers also examined original case records and devised an instrument that interviewers used to evaluate parents’ overall feelings about adoption and infertility. The study focused on school performance, quality of past and present child-parent relationships, health, vocational history, heterosexual adjustment, and parental satisfaction. Outcomes were summarized, an Overall Adjustment Score was calculated, and the 100 adoptees were divided into three groups: “low-problem,” “middle-range,” and “high-problem.”

The researchers assumed that difference in adoption made all the difference, so it astonished them that most of the parents they interviewed (73 percent) insisted that the problems they had encountered had nothing to do with adoption. Even parents of “high-problem” children were unlikely to blame adoption itself for whatever had gone wrong.

There were other surprises too. Age at placement, which had declined steadily during the previous four decades because of studies like How Foster Children Turn Out and adopters’ demand for babies, did not influence outcomes as the researchers expected. In fact, “high-problem” children in the sample had been placed younger than “low-problem” children.

Jaffee and Fanshel were also perplexed to discover that telling did not emerge as a significant variable (Marshall Schechter’s “ Observations on Adopted Children” predicted that it would.) While the vast majority of parents had told the children about their adoptive status, they varied widely in when they told, how they told, and how often (if ever) they returned to the topic. Few parents had given children full and honest information about illegitimacy, for example, and many deliberately withheld embarrassing details. Seven tables that presented statistical correlations related to telling revealed that none of this mattered. Only one thing appeared to influence outcomes. Adoptees who expressed more curiosity about their natal backgrounds were more highly clustered in the “high-problem” group. This finding confirmed a view that was widespread at the time. Indifference to genealogy was a sign of success. Adoptees who gave little or no thought to the facts of their birth had turned out well.

How They Fared in Adoption demonstrated a paradox in the evolution of outcome studies over time. As researchers utilized more sophisticated design and statistical methodology to control for variance, they also became more reluctant to make causal claims and less confident that knowledge would translate into progress. Studies that set out to banish uncertainty from family-formation were more and more likely to conclude that little could be known for sure.