 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Currently Released
Data
Data from AHAA will be made available in multiple stages through Add Health (as a part of Wave III of Add Health). Information about future releases is available on the Future Release page. Currently released data is described below. For more extensive information about how variables were constructed and their distributions, see the Users' Guides.
- Linking Indicators
- Academic Courses Indicators
- Academic Networks
- Contextual Data
- Transitions
- Curriculum Data
- Primary Data
- Graduation Status
- Sample Weights
Linking Indicators
Due to the multi-cohort design of Add Health, students’ high school careers overlap differently with the Add Health survey years. Thus, AHAA created a set of variables that consist of survey to transcript matching indicators. Importantly, these variables enable analysts to link students’ course-taking information to the school year 1994-95 – when the in-school survey and Wave I were conducted – and therefore connect existing survey data from Add Health to academic data from AHAA. Specifically, these variables allow analysts to discern when each student was in high school, the duration of each student’s high school career, what grade level each student was in during the first Add Health survey year, and each student’s transcript-indicated grade level when high school course-taking began.
Academic Courses Indicators
A second set of released variables are constructed academic courses indicators, which measure aspects of students’ course-taking enrollment and performance in each year and cumulatively across all years of high school. The range of academic courses measures available for future release include course sequences, course type, course grades, course failures (failure index variables), semesters attempted, and credits earned variables for the key academic subjects of math, science, foreign language, English, history/social science, and physical education. Academic courses measures are also available for the composite category “overall” which refers to all coursework (including non-core and non-academic courses) taken by students per year and by the end of high school. Additional academic indicators will be released in the future including
more detailed foreign language indicators and physical education GPA and
credits earned indicators.
The following five sections describe the currently-released academic-course indicators.
Course Sequences. Course sequence indicators were developed to capture the academic level of students’ coursework in highly differentiated and sequentially ordered core high school subjects, serving as key measures of student academic achievement in these subjects. Because high school math and science courses are typically organized into hierarchical sequences – meaning successive courses are recognized as being more advanced and generally requiring more prerequisites – variables indicating students’ location in these subjects’ course hierarchies per year, as well as the ultimate level attained by the end of high school were constructed.
GPA. The Grade point average (GPA) variables capture students’ academic performance in each year of high school, as well as cumulatively across all years of high school. Separate math and science, as well as overall GPA indicators which measure students’ performance in coursework taken across all subjects (including electives) were constructed. The GPA variables are calculated as the average grade across semester-length courses in a given year (for the yearly indicators), or across all years of students’ course-taking (for the cumulative indicators). In contrast to self-reported data, these are official indicators of performance as recorded on students’ high school transcripts. As such, they provide analysts with parallel measures of academic performance for students for all years of high school from all the various cohorts included in Add Health/AHAA.
Course Failures. While GPA captures the range of student performance, the course failures or failure index variables convey information about the extreme end of low academic performance. These variables are proportions that correspond to the number of semester-length courses failed (in each year or across all years of high school) divided by the number of semesters-length courses attempted (in each year or cumulatively). Courses not assigned a grade of A-F are not included in the calculation of the failure index. As with the GPA indicators, separate failure index measures for math, science, and overall were constructed.
Course Type Indicators. Course type variables facilitate identification of the degree of difficulty of coursework taken by Add Health students, shedding insight into the grouping of these students according to ability (elite, average, at-risk/below average). Specifically, analysts can explore these variables to determine if students took regular, honors, advance placement (AP), international baccalaureate (IB), or remedial courses in English and history/social science in each year of high school course-taking. Because course sequence indicators were not generated for English and history/social science (these subjects are not typically hierarchically ordered), the English and history/social science course type variables are significantly detailed – representing particularly important measures of students’ academic position in these two core academic subjects.
Given that the course sequence variables enable identification of the placement of students in the math and science course trajectories per year and by the end of high school, math and science course type variables were not produced. Although course sequence variables were constructed for foreign language, they do not fully capture students’ foreign language course taking experiences. To address this limitation, three supplemental foreign language course type variables were developed. Analysts can use foreign language course type variables to identify which students ever enrolled in AP/IB level foreign language during their high school careers, which students took multiple foreign languages, and to discern which specific foreign languages were collapsed into the category “Other”.
Semesters Attempted and Credits Earned Indicators. Semesters attempted and credits earned variables convey information about the intensity of students’ exposure to course content. Specifically, these variables inform the analyst of the number of semester-length courses students attempted in each of the core academic subjects (math, science, foreign language, English, and history/social science) per year of high school, and the amount of credit, if any, students received for them. Overall semesters attempted and credit earned variables were also produced to enable analytical assessment of the total amount of coursework (including electives) students took and received credit for in each year and cumulatively. This information is critical to understanding Add Health students’ academic experiences because learning opportunities and the benefits of education are largely shaped by the quantity of course work taken and assimilated. Importantly, these variables are not restricted to graded courses as in the case of the GPA and failure index variables, but capture information about all types of high school coursework taken by students including pass/fail and non-graded courses. Thus, they enable a comprehensive evaluation of students’ high school course taking patterns and academic achievement in the key curricular subjects and overall.
Academic Networks
Extensive educational research has shown that the internal academic organization of schools shapes within-school academic and social processes and students' outcomes. AHAA adapted and developed two systematic approaches for measuring academic networks within high schools for two academic years, 1994-1995 and 1995-1996. (1) Course-overlap indicators, akin to the friendship data in Add Health that captures dyadic ties between two individuals, were constructed, enabling analysts to examine the effects of the relationships that occur when students take similar courses. (2) Local positions were estimated using network methodology to produce non-overlapping course-taking clusters for all the original Add Health schools (Field, Frank, Schiller, Riegle-Crumb, and Muller 2006). Students are placed in one local position per academic year, based upon the students' transcript-recorded course-taking history. These local positions represent particularly salient, intermediate social contexts within the larger school environment. Through the use of hierarchical-linear modeling techniques, these local positions can be used to estimate the effects of the social milieus within schools on student academic, social, and health-related outcomes.
Contextual Data
Institutional characteristics and the surrounding context of schools clearly impact students’ academic experiences. Therefore, data from the Common Core of Data (CCD) survey, the Private School Survey (PSS), the U.S. census, and the 2000 Office of Civil Rights (OCR) data which describe school and/or district features were attached to participating schools. Currently, no variables from the OCR data have been released, though the variables constructed from the PSS are released almost in full (one variable will be released at a future date). The indicators constructed from the CCD are partially released. To see complete documentation of how these data sources (the CCD, PSS, and OCR data) were linked to AHAA and Add Health, how variables were constructed, and a complete list of the specific variables that are currently available, see the Contextual Data User's Guide on the User's Guides page.
Transitions
As adolescents move through school, they are confronted with transitions that impact their educational trajectory. Most students transition from middle school to high school by changing schools. In addition, some students transfer to a new school during high school. These transitions can be used to trace students through middle school and high school as well as estimate the effects of schools on academic and health-related outcomes for incoming students. Available measures include middle school students’ transitioning to high school, student transfers between Wave I and Wave II, and district indicators of last high school attended for transfer students.
Curriculum
The impact of coursework on student educational and health-related outcomes is intrinsically related to curricular content. The textbook coding is based on curriculum frameworks developed for the Third International Math and Science Study (TIMSS) and is used to create summary measures for the topics covered (content) and the types of tasks students are asked to do with specific topical information (performance expectations) (Schmidt, McKnight, Houang, Wang, Wiley, Cogan, and Wolfe 2001; Schmidt, McKnight, and Raizen 1997). Using information from textbooks and other instructional materials for high school-level mathematics and science courses, William Schmidt and colleagues developed a method to measure two fundamental aspects of math and science curriculum: (1) content, and (2) performance expectations. See http://ustimss.msu.edu/ for more information. Applying this method to AHAA transcript data, AHAA constructed several relatively fine-grained indicators of the learning opportunities made available to AHAA participants through their math and science courses. These indicators are based on detailed coding of textbooks schools reported using in each course in these two subjects. These measures can be linked to students’ academic performance, attainment, and to their health-related attitudes and behaviors as reported in the Add Health surveys (Schiller, Picucci, Schmidt, and Houang 2004).
Primary Data
The primary or raw data indicators are based on information collected from participating schools, and listed directly on student transcripts. One grouping of primary measures conveys comprehensive information about the specific materials gathered from schools during the data collection process. Analysts can examine these measures to determine which and how many data collection instruments were submitted by each school for AHAA. Student-level disposition variables were also produced to enable assessment of the availability of data per Add Health/AHAA student. A third set of primary indicators concern school characteristics as ascertained from school information surveys completed by school administrators. The final grouping of primary or raw data indicators relate to pertinent items recorded on student transcripts such as details about student high school exit status, last school attended, and standardized test information (which is available for only a very limited number of students). Currently, one variables from the Primary Data is released - the last school attended. See the Primary Data User's Guide on the User's Guides page for more detailed information about the Primary Data measures that are released and that will be released at a future date.
Graduation Status
A third type of first-release indicators concerns students’ exit status from high school and month and year of graduation. These indicators came directly from student transcripts or related materials, and were not constructed or altered by AHAA researchers. Categories for exit status include: graduated with a standard diploma; graduated with honors diploma; graduated with special education diploma; received certificate of attendance; still enrolled; dropped out; non-graduate; certificate of completion; and graduate equivalency diploma. Information for these variables came from student transcripts when available. When the transcript was not available, information entered on the transcript request form or any additional information that the school provided was used.
Sample Weights
Education component sample weights are also included in the first release. The Add Health attempted to collect Transcript Release Forms from all Wave III participants. Although special efforts were made to collect this information, transcripts were unavailable for some students. The AHAA education weights correct for bias due to non-response. Adjusted weights were created for two sets of respondents: longitudinal Wave II-III respondents and cross-sectional Wave III respondents.
Persons interested in obtaining first-release AHAA data files from Add Health should contact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill, NC 27516-2524 or visit their website at http://www.cpc.unc.edu/projects/addhealth/data/contract.