William Sidney Rossiter was born at Westfield, Massachussets, the only child of Samuel E. Chadwick and Margaret Harvey (Rossiter). When the boy was five years old his parents were divorced, the mother being given custody of the child, and four years later his name was changed to that of his mother's father. His mother soon removed to Washington where she lived for many years with two unmarried sisters.
The influence of a family friend, Julius H. Seelye, drew Sidney to Amherst College in 1880.
After graduating (1884) Rossiter entered the business office of the New York Tribune and continued for sixteen years in newspaper work or publishing. In 1900 he was given charge of printing and publishing for the Census Bureau, his duties being those of a liaison officer between the Bureau and the Government Printing Office. The appointment proved a turning point in his life.
His new duties brought him into intimate contact with a group of statisticians who were interpreting the population statistics of the Census of 1900; this experience gave him what was virtually post-graduate training in economics and statistics; he wrote the reports on printing and publishing for the censuses of manufactures of 1900 and 1905.
His lack of formal statistical training proved no handicap, for it emboldened him to go beyond the actual meaning of his figures. William Rush Merriam, the director of the Census, leaned heavily on him, and Merriam's successor, Simon Newton Dexter North, made him his chief clerk and acting director. Rossiter's management of the Census printing and his magazine articles on the subject led to his appointment in 1908 by President Theodore Roosevelt to investigate the increasing cost of government printing.
While superintending the publication of the names of heads of families recorded at the census of 1790, he prepared A Century of Population Growth from the First Census of the United States to the Twelfth, 1790-1900 (Department of Commerce and Labor, Bureau of the Census, 1909), which contained an estimate of the proportion of English, Scotch, Irish, German, Dutch, and French blood in the population of the United States, based upon classification of families according to the stock indicated by the last name. Although he failed to allow adequately for the effects of translating or transforming foreign into English names and so overstated the proportion of English blood, he presented in this study an approximate solution of a problem previously thought insoluble, and his estimate in improved form is now used in determining immigrant quotas from European countries to the United States.
When Director North resigned, in 1909, as the result of a departmental controversy, Rossiter resigned also, and for some years thereafter divided his time between the dry-goods business of R. H. Stearns & Company in Boston and the Rumford Press of Concord, N. H. The latter enterprise throve under his management until it absorbed all his energies. During the twenty years of his connection with it as treasurer, vice-president, and president, its employees increased ten-fold, the value of its capital twenty-fold, and the quality of its publication was equal to the best.
In 1918 he persuaded the secretary of commerce to invite the American Statistical Association and the American Economic Association to cooperate with the director of the Census by appointing a joint advisory committee. Rossiter became its chairman and retained that post until his death. He had long made his summer home in Vermont, first at Greensboro and later at Morgan, and in 1911 he printed "Vermont: an Historical and Statistical Study of the Progress of the State" in the Publications of the American Statistical Association. This had been planned as a state document and prepared with the cooperation of the federal Census Office. It was followed by several less technical articles and addresses, all contributing to the organization of state councils and the New England Council, which aimed to deal constructively with social and economic problems in that group of states.
He was chairman of the New Hampshire Council and an active member of the New England Council. In 1922 he was president of the American Statistical Association and in this same year published Increase of Population in the United States, 1910-1920 (Census Monograph No. 1). He was a loyal member of his college class, attended forty-eight of the annual reunions held for fifty years after graduation, and prepared and published the annual class record. He died in Concord, N. H.
Rossiter was known for the preparation of the "A Century of Population Growth" from the First Census of the United States to the Twelfth. His report on investigation of the increasing cost of government printing contributed to the resignation of the Public Printer, and influenced the subsequent administration of the Government Printing Office.
On Oct. 21, 1891, Rossiter married Nellie C. Budd of New York. They had one child, a daughter.
