LSAL Project Design Highlights
The design of the longitudinal study is characterized by a number of critical features:
A statistically representative sample adults ages 18-44 who, at the time of screening, had not finished or were not currently enrolled in high school, or gotten a GED and were proficient speakers of English (not necessarily native speakers) allows generalization to the target population of interest.
Student and Non-student comparisons
In order to better understand the contribution of adult literacy programs to the learning and literacy development of adults, LSAL compares the experiences of two groups within the target population: adults who participate and do not participate in the formal programs. People entering an ABE program in 1998 at one of several area community colleges were recruited into the study selection pool as a way to over-sample students for adequate statistical comparison.
This is a learner-centered panel study rather than a program evaluation design. Programs are not directly involved in collecting data about their students for the LSAL or trying to follow them over time. Previous studies using strategies dependent on programs have had methodological problems because programs are not set up to follow individuals over time. This is an important feature of the study because it shifts the focus on inquiry from a program frame of reference to the experience of learners who move in and out of programs.
LSAL is a panel study that has six data collection points over nine years. The panel of 940 people who were interviewed in wave one are retained in the study and were interviewed once a year between 1998 and 2001 (waves 1 through 3) and once every two years for waves 4 through 6. This design gives 6 points of measurement for modeling change over time and a spread of nine years over which development and the outcomes of development can occur.
An extensive background survey instrument organizes learning in multiple contexts: formal education programs include ABE, work contexts, home and family, and personal interests. Additional one time modules have been added on including learning disabilities, turbulence in daily life, health, and computer literacy.
Multiple outcome measures
Multiple outcomes will be tracked over time, including basic skills, changing patterns of literacy and learning practices, and changes in a range of social and economic activities. Both "hard" and "soft" indicators are being collected wherever possible. For example, employment and earnings data are being collected both through interviews and with participants' permission from state administrative records; growth in basic skills is being assessed through paper and pencil tests as well as through interview questions and self report
Many questionnaire items are identical or highly similar to ones used in other major research projects in order to promote comparability and utility of the data sets and research findings.
Multivariate longitudinal modeling
The research questions will be addressed with multivariate and multilevel models. Because the LSAL is longitudinal the primary research questions cannot be answered until all the data are collected and analyzed.
A high quality sample has been recruited and retained despite the expectation that it is difficult if not impossible to do so with this population which is notoriously mobile and hard to locate. Through Wave 5, about 90% of the panel is still in the study (even though some respondents are not interviewed in a given wave, they return to participate in later waves). This outstanding figure is due to the attention, planning and resources invested in the tracking and detective operations. High retention improves the statistical power and validity of the final models and findings.
|Wave Number||Retention Rate|
The project has a qualitative component that is working with a smaller number of study participants in a more informal and intensive manner. These qualitative case studies are being carried out with pilot subjects from the survey, individuals who are not part of the formal survey sample but who nevertheless receive the same interviews and assessments as individuals in the formal sample. Because both the survey data and the richer qualitative data are being collected independently from these individuals, the two types of information will complement each other and help the project better understand the strengths and limitations of the survey data.