'Estimates from the March 1985 CPS reflect the civilian noninstitutional population of the United States and members of the Armed Forces in the United States living off post or with their families on post, but excludes all other members of the Armed Forces.

2 Data from the 1980 census are based on 100-percent tabulations of the resident population in the United States on April 1, 1980.

3In the 1980 census, the "Other Spanish" category included persons from Spain, the Spanish-speaking countries of Central and South America, and Hispanic persons who identified themselves generally as Latino, Spanish-American, Spanish, etc. In the CPS, the category "Central or South American" is listed as a separate origin. *For the CPS, this category includes persons who did not know or did not report on origin.

March 1982 and 1985. This improvement, however, did not lessen the gap in educational attainment between the Hispanic and non-Hispanic populations because the non-Hispanics also experienced an improvement in their educational attainment level. Specifically, the proportion of Hispanics 25 years old and over who had completed 4 years of high school or more in 1985 was 48 percent, compared to 45 percent in 1982. The comparable proportions for non-Hispanics are 76 percent and 72 percent, respectively (table D). The apparent differences between 1982 and 1985 in the proportion who completed 4 years of high school or more among Hispanic subgroups are not statistically significant (table E).

The proportion of Hispanics with 4 years or more of college showed no improvement between 1982 and 1985, while the proportion among non-Hispanics showed a significant improvement. Specifically, the proportion of non-Hispanics completing 4 or more years of college in 1985 was 20 percent, compared with 18 percent in 1982. The comparable proportions for the Hispanic population were 9 percent and 8 percent, respectively. The apparent changes between 1982 and 1985 in the proportion who completed 4 years or more of college among the Hispanic subgroups were not statistically significant.

In March 1985, differences in educational attainment levels existed among the Hispanic subgroups. Persons of Central and South American and other Spanish origin had the highest proportion of persons 25 years old and over who had completed 4 years of high school or more (about two-thirds each) this relationship also was evident in 1982. For 1985, Mexican-origin persons had the lowest proportion of persons with 4 years of high school or more (42 percent).

Among the Hispanic subgroups, Mexicans and Puerto Ricans had the lowest proportions with 4 or more years of college completed. Cubans, Central and South Americans, and persons of other Spanish origin had significantly higher proportions than the other two groups (table E).

School enrollment. Almost all of Hispanic and nonHispanic children of compulsory school age (usually 5 to 15 years old) were enrolled in school in October 1984. However, the school enrollment rate differed between Hispanic and non-Hispanic children of non-compulsory school age. Hispanic children 3 and 4 years old, were less likely to be enrolled in nursery school (24 percent) than non-Hispanic children of the same age (37 percent). A difference was also evident for children above the ages of compulsory school attendance; about 86 per

Voting and registration. Voting information published in other Current Population Survey Reports is based on all persons, 18 years old and over, regardless of citizenship status. However, 32 percent of the Hispanic population of voting age were not citizens of the United States and were not eligible to vote in November 1984. Therefore, it is misleading to use all persons, 18 years old and over, as the base for describing the voting and registration characteristics of the Hispanic population. In the following discussion, only citizens of the United States, 18 years old and over, are considered.

In November 1984, the registration and voting rates of Hispanic citizens were lower than those of nonHispanics. Specifically, 59 percent of eligible Hispanics, 18 years old and over, reported that they were registered to vote in the 1984 election, compared with 72 percent of eligible non-Hispanic persons. Only 48 percent of eligible Hispanics reported voting in the election, compared with 63 percent of those not of Hispanic origin.

Certain social and economic characteristics appear to be directly related to reported voter participation among both Hispanic and non-Hispanic populations:

• Older people were more likely to vote than younger people. (This comparison applies to age groups shown in figure 2.)

Figure 2.

• Persons with higher educational attainment levels were more likely to vote.

• Employed persons were more likely to vote than unemployed persons.

• Persons in families having incomes of $25,000 or more were more likely to vote than persons in families having incomes below $25,000.

The lower voter participation rate among Hispanics appears to be related in part to a younger, less educated, more unemployed, and less affluent population.

Fertility. The June 1985 CPS showed that Hispanic women, on the average, begin childbearing at earlier ages, and have more children than their non-Hispanic counterparts. Among women 18 to 44 years old, Hispanics averaged 1.7 children per woman as compared with 1.3 children per non-Hispanic woman in the same age group. Also, about 40 percent of Hispanic women ages 18 to 24 years old had borne children, compared with only 28 percent of their non-Hispanic counterparts. For the 25-to-34 age group, 75 percent of Hispanic women had borne children, compared with only 65 percent of non-Hispanic women. However, in the 35

to-44 age group, a very high proportion of both the Hispanic and non-Hispanic women had borne children (90 percent and 85 percent, respectively) (table G).

Family size. The higher fertility levels of Hispanic women as compared with non-Hispanic women contributes to the larger average family size of Hispanic families. In March 1985, Hispanic families averaged about 3.88 persons, compared with 3.18 persons for non-Hispanic families (tables 1 and 16).

Between 1982 and 1985, the mean size of Spanishorigin families did not statistically differ (3.74 and 3.88 persons, respectively), while there is some evidence that the mean size of non-Hispanic families decreased (from 3.22 to 3.18 persons). During this period changes in the mean family size within Hispanic subgroups were also not significant. The size of families among the Hispanic subgroups in March 1985 ranged from 4.15 persons per family of Mexican origin to 3.13 persons for Cuban families (tables D, E, 1, and 16).

Family composition. Families of Spanish origin in the United States totaled 3.9 million in March 1985. About 23 percent of these families were maintained by women with no husband present; the comparable proportion for non-Hispanic families was 16 percent. Some statistical evidence exists that suggests the proportion of Hispanic families maintained by married couples was lower in 1985 than in 1982 (72 percent and 74 percent, respectively). The apparent change between 1982 and 1985 in the proportion of Hispanic families maintained by women with no husband present, however, was not statistically significant (22 and 23 percent in 1982 and 1985, respectively) (tables D, 1, and 16).

There are differences among the Hispanic subgroups in the proportion of families maintained by women with no husband present. In 1985, Puerto Rican families were twice as likely as families of other Hispanic origin

6A Hispanic family is defined as a family in which the family householder is of Spanish origin.

to be maintained by women with no husband present. The observed changes in the proportion of families maintained by women with no husband present for the specific subgroups between 1982 and 1985 were not significant (tables E, 1, and 16).

ECONOMIC CHARACTERISTICS

Employment. Employment improved for Hispanics between March 1982 and March 1985, reflecting the general improvement in the economy which began in 1983. In March 1985, the unemployment rate was 11.3 percent for Hispanics 16 years old and over, down from 13.4 percent in March 1982. There is at least some evidence that there was a drop in the unemployment rate for each Hispanic subgroup. Despite this improvement, the unemployment rate for Hispanics was still higher than that for non-Hispanics (11.3 and 7.4 percent, respectively) (tables D and E).

Changes in the labor force participation rates between March 1982 and 1985 differed by sex. Hispanic men experienced a lower labor force participation rate in 1985 than in 1982; in contrast, Hispanic women experienced a higher participation rate in 1985 than in 1982.

The labor force participation rate for Hispanic men 16 years old and over was 78 percent in 1985, compared with 81 percent in 1982. Significantly lower labor force participation rates in 1985 were evident for men in the prime working years. Specifically, the participation rate of Hispanic men 25 to 44 years old was 94 percent in 1982 and 91 percent in 1985; for Hispanic males 65 years and older, the respective rates were 21 percent and 16 percent.

Even though Hispanic men experienced a negative change in labor force participation, their participation rate in 1985 still exceeded that of non-Hispanic men (78 percent and 74 percent, respectively). In particular, the labor force participation rate of Mexican-origin men was very high in 1985, exceeding the labor force participation rate for total Hispanic men (tables D, 2, and 9).