When calculating statistical significance, there was some overlap between groups in the middle category and groups in the lower and higher categories. For example, while the difference between Middle Eastern households in the lower use category and East Asian households in the middle category was not significant, the difference between Eastern European households and East Asian households was significant. From the middle to the upper levels, the difference between East Asian households and Southeast Asian was significant, but the differences between East Asian households to South American and African to South American households were not significant.
Figure 2 shows rates of public library use in the past year. Concerning this use, during the past year European households were once again the least likely to have used the library at 37.8%. South Asian households were again the most likely library users at 57.9%, a range of 20.1 percentage points. While rates of public library use for all households went up when comparing use in the past month to the past year, the placement by geographic region in low-, middle-, and high-use categories stayed the same for all groups except for Canadian or other North American, which jumped from the low-use to the high-use group. This suggests that the factors that affected decisions to use the public library remained consistent from a monthly to a yearly basis.
While the low-, middle-, and high-use group cutoffs in figures 1 and 2 were chosen on the basis of calculations of statistical significance, looking at the tables would suggest that a practical division might move the group boundaries slightly. This is particularly clear in figure 2, in which it seems that Middle East should be moved from the lower to the middle category and Canadian and other North American should be moved down from the upper to the middle category.
SocioDemographic Analysis
In the literature, the needs of newly arrived immigrants differed from those of established immigrants. New immigrants need information on jobs, housing, literacy/ESL, and citizenship. As immigrants become more established, their library needs blend with those of U.S.–born patrons. Year of entry by world region indicates that there is quite a difference in the immigration patterns of people from different world regions. Table 1 shows, for example, that more than 75% of European immigrants arrived prior to 1989, so patrons from that group will likely have fewer specialized library needs than other groups. About two-thirds of Canadian and other North American respondents also immigrated by 1989. Conversely, half or more of respondents from the following regions came to the United States after 1989: Africa (69.7%), South Asia (62.7%), East Europe (59.4%), Central America or Mexico (55.0%), and South America (51.1%).
Factors affecting trends in immigration and citizenship are complex, and a lengthy discussion of the political history of world regions, immigration legislation, and immigration policy will not be engaged in here. Very briefly, it can be seen from table 1 that the patterns of immigration by decade vary significantly between geographic groups. As mentioned above, this can be an important consideration for librarians who are working on collection development because established immigrants have different library needs from newly arrived immigrants.
Of the demographic variables examined in previous studies, education level had the strongest impact on whether people used libraries. When the U.S. Census Bureau calculates education level, they do so for adults ages 25 and over, presumably to give time to reasonably assume educational completion. For the adults ages 25 and over in these CPS data, the following was the education distribution at the lowest level:
- 60.0% of adults from Central America or Mexico completed less than high school education.
- 31.4% of adults from the Caribbean completed less than high school education.
- 20.2% or fewer adults from all other geographic groups completed less than high school education.
- For groups with the highest education levels, the following was their distribution:
- 80.2% of adults from South Asia had completed at least some college.
- 70.9% of adults from Africa had completed at least some college.
- 50% of adults from all other groups except Central America or Mexico, the Caribbean, and South America had completed at least some college.
Effective public library advertising campaigns for groups with higher education levels might differ from those aimed at groups with lower education levels, as will appropriate collections, programs, and services. Of course, librarians must remember that group data cannot be applied to individuals. For example, while 60% of immigrant adults from Central America or Mexico did not complete high school, it is not appropriate to assume that any particular patron falls into that category. These numbers are only useful for library planning to best serve the community, not to erroneously peg any individual as having a group characteristic.
Table 2 examines library use in the past month by education level within geographic categories. When all households in the dataset were examined together, there was a direct and linear relationship between education level of household reference person and likelihood that someone in the household used the public library in the past month. The data for all respondent households together were as follows (n = 43,440 households):
- For reference persons with less than a high school education, 16.1% of households used the library in the past month.
- For reference persons with a high school diploma or GED, 24.2%.
- For reference persons with some college or an associate’s degree, 33.9%.
- For reference persons with a bachelor’s degree, 41.2%.
- For reference persons with a graduate degree, 46.0%.
As illustrated in Table 2, while this linear relationship was predominately reflected no matter which immigrant group was examined, there were a few interesting differences. Particularly interesting was that at the lowest education level, households of respondents from South America, Southeast Asia, and Central America or Mexico were far more likely to have used the library in the past month than other groups in this education category. In other words, while education has been shown in many studies to be the best predictor of library use, this does not apply in the same manner to some immigrant groups. This difference is most likely the result of an intervening variable such as presence of children in the household, family size, or another unidentified variable.
Some other points of interest from the table include the fact that European and East European households were least likely to have used the public library in the past month than other groups in nearly all categories. Between the categories of “some college” and “bachelor’s and higher,” those with bachelor’s degrees or higher from Southeast Asia appear to have used the library less than those who had completed only some college. Geographic regions that did not have enough households to calculate library use within these multiple levels (twenty-five or fewer households in some table cells) were Middle East, Canada or other North America, Africa, and South Asia.