Social Limitation Structural Equation Model 12

--------------- (in press) found a relatively strong correlation between support for

English-only policies and support for social limitation. Therefore a final hypothesis is

proposed:** **

**H**

**5**

**:**

Support for English-only policies is a form of support for social limitation of

immigrants and minorities.

The proposed structural equation model (Figure 1) tests the hypotheses

described above. Based on theory, structural equation modeling provides the means to

test "causal" relations between multiple variables. As a pictorial representation, the

resulting model offers a clearer conceptualization of the theory. This process is an

improvement on alternative statistical methods (e.g., regression models) because SEM

is a confirmatory (hypothesis-testing) process rather than a descriptive one, and also it

provides explicit estimates of error variance. Additionally, while other methods are

based upon observed variables only, SEM has the power to include unobserved (i.e.,

latent variables) in the equation (Byrne, 2001).

FIGURE 1 ABOUT HERE

**Method**

Population

Using random digit dialing (RDD) (Lavrakas, 1993), a sample of households with

telephones was drawn from nine zip codes in Santa Barbara, California (provided by

Scientific Telephone Samples [STS]). The sample was weighted so that the RDD

numbers were created from telephone blocks in proportion to the number of estimated

household listings in each working block. An estimated 1.8% of households residing in

these zip codes were without telephones.