e. How are the data analyzed? You should state the method and why this method is appropriate. Also note if descriptives, graphics, or supplementary analysis are used.

The two surveys are examined to make conclusions on the patterns of adaptation of the Mexican immigrants’ children. Having gathered and merged the data on the degrees of speaking mother tongue among several generational cohorts of immigrants, the authors of the article analyze these data to derive linguistic “survival curves”, as they call them. It means that they investigate the chances for a language to survive in the first, second and third generations and then apply life table methods in which they determine the average number f generations through which a mother tongue is expected to survive in the United States. The last variable is defined as “linguistic life expectancies”. At the same time, this is not a technical analysis, but rather heuristic point derived from a demographic technique. The data derived from the two surveys (IIMMLA and CILS) is merged in the Table entitled “Numerical distribution of the IIMMLA-CILS merged sample by population group and generation.” In this table, there are nine groups of immigrants by origin and six generational cohorts. Linguistic survival curves are shown in the Figure, where the x-axis shoes the generation cohorts and the y-axis specifies the proportion of those still speaking their mother tongue. Each curve stands for a separate population group. Another figure shows the proportion of immigrant group members speaking their mother tongue by generation. Further, linguistic life expectancies for each generation group are visually presented in a bar graph.

f. What are the findings? Do they support the hypotheses?

Linguistic survival in Mexican neighbourhood of Southern California has been studied by two indicators: by what language, native or English is preferred in household, and how good the ability to speak English fluently is. It has been found out that despite a strong ability to maintain mother tongue, even the Mexican immigrants in the third generation fail to choose it in household or speak it better than Spanish. Linguistic death is stated as a fact in the United States. Thus, Huntington’s argument is contradicted directly and the hypothesis of the study is supported.

g. Do you think the author(s) answered the research question(s)? Why or why not? (Here you can talkabout limitations of the work and/or suggestions for future research.)

The authors seem to have answered the question fully, and even more than that, they have projected important issues for further research. They have shifted attention from the supposed threat to English dominance in the United States to the future of bilingualism and the languages of the immigrants. As the authors alarm, these languages require protection from proliferation, but provide no solution for the problem. In fact, that is not a limitation, but rather a space for prolonged studies. What is more, the problem of languages in Southern California has been entered to the global context of language death problem, which is also a constructive remark worthy of special attention.

2.   (9 points) You are a research assistant helping a professor use GSS data to answer the question “What  factors influence the likelihood of getting divorced?”Â Â  What does the table below tell you about the data?  In a logistic regression analysis using these data, what will be the highest  possible N?  Who would be excluded from answering the question “Have you ever been divorced?”

Descriptive statistics is intended to show the number of computations (N), the average score, lowest and highest scores, and standard deviation (SD). The table includes five variables, which are the fact of ever being divorced, age, socioeconomic status, attendance of religious service and being Catholic. An answer to the question whether a person has ever been divorced can have only two variants, “yes” or “no”, and they are quantitatively expressed by numbers 1 and 0, respectively. The same logic refers to the fact of being Catholic. Minimum age of the respondents is 18 and maximum is 89. The mean score is the most important because it provides us with an image of the average respondent. Here this is a 44 year old person who is not likely to be divorced and not likely to be Catholic, but attending some religious service. As there is no standard deviation for being divorced and Catholic, these variables provide clearer impression of likelihood, and the highest number thus will be 1500. Only people of younger age can be excluded from answering the question about being divorced.

Leave a Reply