How America adopted radio: demographic differences in set ownership reported in the 1930-1950 U.S. censuses.

Radio set ownership data reported in the U.S. Censuses of 1930, 1940, and 1950 suggest that several factors affected the rate of radio adoption during these decades. Although a majority of U.S. households were radio-equipped by 1931, substantial numbers of poorer Americans, especially those living in the South, could not afford sets until much later. A majority of African American families, the poorest of all, did not own radios until well into the 1940s. It was not until 1950 that 95% of all American households owned receivers.

**********

There is little doubt that when radio first arrived in the 1920s, it caused a great deal of excitement. Sterling and Kittross (2002) describe it as a time when "the country was afire with radio fever" (p. 69). Barnouw (1966) writes of the early "euphoria" of radio set manufacturers who "couldn't produce fast enough" (p. 91), and Hilliard and Keith (2001) maintain that "people were buying radio receivers as fast as they could afford to" (p. 31). Such descriptions are justified since the number of licensed broadcasting stations jumped from just 5 in 1921 to over 500 by 1924 (Sterling & Kittross, 2002, p. 827) and by 1927 Americans were spending $198 million annually on radios (U.S. Bureau of Foreign and Domestic Commerce, 1930, p. 841). (1)

The decade of the 1920s was a prosperous period for many Americans, especially those who lived in the urban areas of the North and West. Good economic times meant that many city dwellers had the luxury of buying the new radio equipment as soon as it appeared on the market. In addition, early radio stations were usually built near urban centers where signals could be received by the maximum number of affluent consumers. By 1928, there were 28 stations broadcasting in the New York City metropolitan area, 36 in Chicago, and 24 in Los Angeles (U.S. Federal Radio Commission, 1928, pp.117-123).

Yet America's phenomenal radio boom was primarily experienced in the cities of the North and West. In the nation's poorer regions, radio's arrival was much slower. In 1928, Atlanta had only three radio stations and New Orleans seven. For the 44% of Americans who still lived outside the cities, service was spotty or non-existent. Several large rural states had fewer than six radio stations, and these tended to have smaller transmitters than their urban counterparts. For example, in 1928, the entire state of South Carolina had only two stations transmitting with an aggregate power of 90 watts (U.S. Federal Radio Commission, 1928, pp. 117-123).

While the coming of radio may have been a sensation in some parts of the country, other Americans were suffering from economic hard times--even in the years before the Great Depression struck in 1929. In these areas, many families were in economic crisis, few entrepreneurs were willing to build stations, and widespread radio adoption was still years away.

Despite the considerable interest in radio's early industrial history, little detailed research has been reported on how receivers were adopted in the United States. Starch conducted surveys on radio adoption in the 1920s (Spalding, 1963-1964, pp. 36-37) and Cantril & Allport (1935, pp. 85-86) made early use of Census data to describe the growing radio audience. Lichty and Topping (1975, pp. 451-455, 521) and Sterling & Kittross (2002, p. 862) described overall adoption rates for several years based on data from industry and Census sources. (2) Jellison (1993, p. 61) used 1930 Census data to examine different rates of radio ownership among farm families in several midwestern states, and Craig (2001) discussed the relative slowness of early adoption by rural families. But none of these sources investigate in depth the differences in radio adoption patterns nor do they attempt to detail the social factors that helped determine when families decided to buy their first radio.

Without a clear understanding of how American families adopted radio, historians have an incomplete picture of the audience during these years. The fact that adoption rates varied widely among different regions and between urban and rural areas means that the demographic make-up of the radio audience changed as time went on. A fuller explanation of the sociological dimensions of these changes can shed light on historical decisions that took place in the industry in such areas as advertising, programming, and regulation. Further, an understanding of the factors that affected radio adoption rates may provide insight into audience acceptance of contemporary and future media technologies.

This paper attempts to fill the gap in our understanding of radio's adoption by using the available U.S. Census data on radio ownership in the years 1930, 1940, and 1950. Rogers' (1995) past work in the diffusion of innovations provides a framework for the examination of geographical and ethnic differences in patterns of radio ownership over the three critical decades it took for radio to be adopted in 95% of American homes.

Method

United States Census Bureau reports for the decennial Censuses of 1930, 1940, and 1950 were examined for data on the ownership of radio by U.S. households. (3) In each of these years, census takers determined whether or not households "contained a usable radio set or one only temporarily out of repair" (U.S. Bureau of the Census, 1943, Vol. II, Pt. 2, p. 5). In 1930 and 1940, attempts were made to gather these data for each U.S. household enumerated, but in 1950, a sample of 20% of occupied households was used. (4) Radios in places of business, institutions, and hotels were not counted. Findings were compiled by geographic location and, in some years, by race and parentage of the head of household.

Geographic Categories

The Census Bureau presented summaries of its results using various state groupings to yield several different geographic categories. To allow for comparison among the three decades, the data in this paper are reported based on the four regions used in the 1950 Census: Northeast (with about 27% of U.S. households), North Central (with about 32%), South (with about 29%), and West (with about 12%). (5)

The Censuses also reported radio ownership based on whether the households were "urban," "rural non-farm," or "rural-farm." For 1930 and 1940, the Census Bureau defined "urban" as a city or other incorporated place having 2,500 inhabitants or more. (6) Rural households were defined as those located outside these urban areas and further divided into "rural non-farm" households and "rural-farm" households. Since preliminary analysis indicated that radio ownership patterns in rural farm and non-farm households were similar, this distinction was collapsed into the single category of "rural."

Race, Nativity, and Parentage

In all three Censuses, the Bureau gathered information based on the "race and color" of the head of household. (7) Unfortunately, the method of tabulation and presentation of these data varied from Census to Census, and in 1950, radio ownership data were not reported by race at all. This paper presents the available findings on race and parentage for 1930 and 1940 separately in the form made available by the Bureau.

Some results categorized households based on whether the head was "White," or "Nonwhite," with the latter category including "Negroes, Indians, Japanese, Chinese, and other Nonwhite races."(8) A label of "Mexican" was used for "persons of Mexican birth or ancestry who were not definitely Indian or of other Nonwhite race." Such "Mexicans" were classified as "White" in the 1940 and 1950 data, but as "other races" in the 1930 report (U.S. Bureau of the Census, 1953b, Vol. II, Pt. 1, p. 35).

In the 1930 Census, White heads of households were categorized based on their nativity and parentage. The category "native White-native parentage" included U.S.-born Whites both of whose parents were also U.S.-born. Other categories included "native White-foreign or mixed parentage," and "foreign-born White." Nativity and parentage was not determined for those classified as "Negro" or "other races" (U.S. Bureau of the Census, 1933, Vol. VI, p. 6.).

Findings

U.S. Census data from 1930, 1940, and 1950 present a picture of radio ownership in relation to three major variables: Geographic region (Northeast, North Central, South, or West), whether located in an urban or rural area, and, for 1930 and 1940, by race of the head of household.

Finding 1: Radio adoption occurred at different rates in different parts of the country with patterns consistent in all three Censuses.

Finding 1a: Rates of radio adoption were fastest in the Northeast, followed closely by the North Central and Western regions, and with the South lagging appreciably behind.

Finding 1b: Rates of radio adoption were faster in urban areas than in rural areas.

Finding 1c: The pattern of difference among these variables was consistent across all three Censuses. That is, in all years, adoption was fastest in the urban Northeast and slowest in the rural South.