Advanced Training Institute Sorts the Data!

The APA Science Directorate’s latest foray into advanced training for psychological scientists – a hands-on experience in mining large-scale databases – had a great start at the August ATI in Chapel Hill, North Carolina.

With the generous support of the National Institute of Child Health and Human Development (NICHD), APA organized this ATI to help investigators learn the intricacies of performing secondary data analysis, using the large longitudinal dataset from the ongoing Study of Early Child Care (SECC) as an example. Research Triangle Institute, a primary contractor in building this dataset for NICHD, partnered with APA in organizing the ATI.

The SECC data are from 1,364 families, followed since their infants' birth in 1991. The study covers demographic, family, maternal, paternal and caregiver characteristics; child social and emotional outcomes; language development; cognitive skills; school readiness; growth and health measures, and much more. It includes data collected by observation, testing, and face-to-face and phone interviews. This ATI was designed to introduce the SECC study and its available databases (currently from birth through the third year of school) so that researchers may independently use and train others to use the NICHD databases for original scholarship and publication. Among the individuals in attendance were Ty Partridge and Jennifer DeGroot Hanawalt from Wayne State University in Detroit. When asked to comment on the ATI, here is what they said:

Ty Partridge
“Given my enthusiasm for the NICHD Study of Early Child Care and the opportunities it affords for the field, I was delighted to find out that there was an APA Advanced Training Institute regarding the dataset. I have worked with a number of archived datasets in secondary data analysis and all of them present a myriad of intricate difficulties. Often it can be a substantial challenge to discover even the most basic concepts, such as figuring out what variable a particular variable name refers to. Trying to understand why values seem to have been transformed from the descriptions of the original data or why the sample size of a given analysis doesn’t seem to make sense can lead to weeks of frustration. So, the value of having the opportunity to sit down with the people who know the NICHD-SECC data most intimately cannot be overstated.

“The quality level of this ATI exceeded my expectations. The most impressive aspect of the training institute was the supporting material. Overall, the ATI was valuable to researchers who had a wide range of experience with secondary data, large databases, and statistics. I was fortunate to have the graduate student coordinator of my lab, Jennifer DeGroot Hanawalt, attend this training institute as well.

“It is clear that countless hours of work and exceptional foresight went into the documentation of this dataset. Now rather than spending six-months to a year trying to understand the data well enough to address even basics questions, we have been able to begin analyses on several projects that will be in the manuscript phase by the end of the semester. This research can be integrated with a cadre of interdisciplinary prevention, intervention and research efforts organized into a comprehensive program referred to as the Wayne State University Children’s Bridge program.”

Jennifer DeGroot Hanawalt
“As a graduate student, I found the ATI to be a comprehensive and user-friendly introduction to secondary data analysis, the details of this particular longitudinal data set, and the advanced statistics necessary to address complicated longitudinal questions. At the workshop we learned detailed information about the data collection process, the database, and the data itself. We were trained extensively on how to use the wealth of dictionaries and detailed manuals that accompany the dataset in order to find and understand data. The individualized statistical training and support provided by the group of statisticians was a rare and valuable opportunity to learn a variety of sophisticated statistical techniques and have individualized support as I attempted to employ these new techniques to address my hypotheses. The workshop was an excellent opportunity to meet researchers with similar interests and build a network of colleagues with whom to share ideas and research.”

Applications for the 2004 ATI on using the SECC data will be available in Spring 2004. Visit www.apa.org/science/ati-info.html for more information.