# Intergenerational Education Mobility Lab

The objective of this lab is to form groups and work with actual data to examine basic patterns of intergenerational education mobility, and then to bring these patterns into conversation with the assigned readings. In addition, each group will be expected to write a short (~ 4 pages) summary that describes the results in conversation with a theoretical perspective. Please note that prior experience with this type of basic data analysis is not required. The primary objective here is to give you an opportunity to collaboratively engage with data and practice the craft of sociology.

### Initial Setup

This webpage contains all of the information and links that you will need to complete the computer lab portion of the assignment. The first step is to download the General Social Survey Cumulative Data Set, 1972 - 2016. The file will be downloaded as a zip file that contains the SPSS data set ("GSS7216_R1b.sav") and a set of release notes in PDF. Once you have downloaded the data set the next step is to download the SPSS syntax file that I have prepared for you. The syntax file contains a set of commands that will instruct SPSS to perform a variety of tasks such as variable re-coding, descriptive statistics, and simple bi-variate analyses. In addition to these commands, the syntax file also contains notes that explain each technical step of the lab. As you work with the GSS data set you may find it helpful to refer to the documentation provided through the GSS website.

### Lab Assignment

1. In the first section of the lab you will explore bi-variate correlations between respondents' education attainment and the attainment of their parents. In this section education attainment is measured as the number of completed years of formal education. Before you produce any results, write down the general trends you expect to find in the following correlations across time. Make sure your hypotheses are theoretically informed!

- respondent's education by father's education
- respondent's education by mother's education
- father's education by mother's education

Once you have written down your hypotheses, calculate the bi-variate correlations for each of the decade cohorts by running the appropriate syntax.

- Copy the output into Excel and create a single line graph that depicts the three trends in bi-variate correlations for each of the decade birth cohorts.

2. In the second section you will explore basic cross-tabulations between respondents' degree attainment and the degree attainment of their parents. Note that our measurement of attainment has changed from an interval measure to an ordinal measure. Before you produce any results, write down the general trends you expect to find in the following cross-tabulations across time:

- at least bachelor's degree attainment conditional upon a bachelor's (or more) educated father
- at least bachelor's degree attainment conditional upon a high school educated father
- at least bachelor's degree attainment conditional upon a less than high school educated father
- at least bachelor's degree attainment conditional upon a bachelor's (or more) educated mother
- at least bachelor's degree attainment conditional upon a high school educated mother
- at least bachelor's degree attainment conditional upon a less than high school educated mother

If you are interested in trends related to community college then go ahead and generate these hypotheses. However, please note that some of the older birth cohorts will not have large enough cell sizes to generate reliable results. Regardless, once you have written down your hypotheses, generate the cross-tabulations for each of the decade cohorts by running the appropriate syntax. Note that you may want to re-code these variables in order to simplify the tables and/or to boost the cell counts.

- In Excel, create two line graphs that depict the results of your hypothesized trends. One line graph should depict the trends for bachelor's+ degree attainment conditional upon the three different paternal education origins (BA, HS, <HS), and the second should depict the same trends given maternal origins.

Finally, consider whether or not the different trends identified above would change based on race and gender (including race/gender interactions). Write down a few hypotheses related to these dynamics. You do not have to test these hypotheses, but you may do so if time permits.

### Lab Write up

- Due on Friday September 29 @ 4pm (Only one copy needs to be uploaded to Canvas per group)

The written portion of the lab should be approximately four single-spaced pages and contain the following four sections:

**1. Introduction**: Briefly discuss your hypotheses. Why did you expect to find your hypothesized trends? Did your hypotheses vary depending on the type of measure (i.e. years v. degrees attained)? Why? What was the guiding theoretical basis for these hypotheses? Cite the appropriate literature from the course syllabus. You are welcome to cite sources outside of the syllabus but this is not expected.

**2. Data and methodology**: Briefly discuss the data set, the sample sizes for your analyses, and note the presence of missing data, if any. If some data are missing, how might this bias your results? What are the strengths and limitations of each measurement of intergenerational education mobility (i.e. years of education v. degrees attained)? How are these strengths and weaknesses informed by your theoretical perspective? What are the limitations of relying on bi-variate forms of analysis?

**3. Results**: Present the three graphs and briefly describe the trends. What do you notice about the trends? How do the actual trends compare with your hypotheses?

**4. Discussion**: Do the trends challenge or support the theoretical perspective(s) informing your hypotheses? How? What does this conversation between your theory and data suggest in terms of further research?