Archived Information

Appendix E: Example of a Customized National Long-Term Degree Completion Report

There are many ways to cut the national data on bachelor's degree completion. This appendix presents one example of a customized descriptive report. It was originally prepared for the National Center on Public Policy and Higher Education in January, 1999.

For any descriptive report such as this, one must always ask who the "constituent" wants in the denominator of the ratios. In this case, while the data base, the High School & Beyond/ Sophomore cohort, is the same, the resulting universe is different from that used in the text of the "Tool Box" monograph. The degree completion rates, then, are also different--very different. What follows is the report to the Center.

Students in the universe of analysis were those who (1) had graduated from high school or earned a high school equivalency diploma by the end of 1987; (2) were of known race; (2) attended one or more 2-year and/or 4-year colleges at any time between 1982 and 1993; and (3) earned more than 10 credits. The weighted N is 1,855,866. In the following table, the universe is constricted for selected variables, e.g. SES, highest level of parents' education, etc., where we do not have information for all the students in the sample. The weighted N is stated for these variables.

All differences in columns are statistically significant at p<.05 unless a pair of estimates is indicated with asterisks. What p<.05 means is that the odds of the difference occurring by chance are less than one in twenty. This is a standard threshold for significance. This threshold is modified, in the case of each variable, by accounting for the number of categories and comparisons in the variable.

For each variable, the percentage distribution among values is found in the furthest column to the right. For each variable, this column adds to 100.0%. Otherwise, the rows add to 100.0, except where affected by rounding.

[Student reports of the level of parents' education are notoriously inaccurate. In this case, reported level of mother's/father's education was adjusted with reference to reported occupation--something late adolescents understand more clearly. Thus, for example, a mother who was reported to be a school teacher but with no education beyond high school had her highest level of education recorded as college graduate. This type of adjustment could be made only for those occupations requiring at least a college education, if not a degree--or a graduate/first professional degree. Since some children have only one parent or report occupation and level of education for only one parent, this variable refers to the highest level of education to be found among any parents.]

By Census Division of
Student's High School

"True First Institution" excludes dual enrollment in high school and false starts.

[The years covered by these variables are the four immediately following scheduled high school graduation. Not all students from this cohort who entered higher education were in school during those years.]

[+A difficult category because the loop of questions was confined to students who said they worked to contribute to their education. "Work-Study," as defined here, excludes summer jobs and covers only that work taking place while the student was in school.]

ADDENDUM: States in Census Divisions*

New England: ME, NH, VT, MA, RI, CT

Mid-Atlantic: NY, NJ, PA

E. North Central: OH, MI, IN, IL, WI

W. North Central: MN, ND, SD, IA, MO, KS, NB

South Atlantic: DE, MD, DC, VA, WVA, NC, SC, GA, FL

E. South Central: KY, TN, AL, MS

W. South Central: LA, AR, TX, OK

Mountain: MT, ID, CO, NM, UT, AZ, NV, WY

Pacific: CA, OR, WA, HA, AK

*Divisions used by the Bureau of the Census in the Current Population Series (CPS).

Appendix D: So They Got a Degree! Why Did It Take So Long? Or Did It?