In a new study there is a requirement to specify the general purpose of a research endeavor. This will often include the research question(s), the reason(s) for the study, goals, outcomes, subjects of the study, and other high level aspects of the work. This stage is often characterized by ongoing discussion among the principal researchers and requires a good deal of investigation into the current state of understanding within the domain being investigated. What data exist? What has been published? What impediments need to be overcome? What data are being asked for?
In a study that repeats, the retrospective evaluation will provide many inputs that do not exist in the early stages of a new study, and which can result in specific changes to the study. These changes may include an extension of the scope, changes to the universe, modified or new research questions, etc.
1.1 DEFINE RESEARCH QUESTIONS, UNIVERSE & HIGH-LEVEL CONCEPTS
As a study begins and as each new wave or phase begins, high-level conceptualization underpins what follows. This work may be based on researcher knowledge and experience, new theoretical work, prior studies (or earlier waves), a preliminary literature review, and more. At the beginning of new phases of a project, the results of retrospective evaluation (steps 9.x) may initiate reconceptualization.
1.2 EVALUATE EXISTING DATA AND PUBLICATIONS
A more rigorous literature review may follow an initial one. Existing data might be evaluated in terms of comparability of universe, concepts, and categories, as well as in terms of access restrictions, cost, and format. This activity highlights the importance of good metadata for the existing data.
1.3 ESTABLISH OUTPUTS & NEEDED INFRASTRUCTURE
This step may include preliminary estimates of needed staff, equipment, space, travel, and so on. Preliminary development of a project team to include people with a variety of skills may occur here, as potential collaborators are recruited. It may also be desirable to consider the general form of needed analyses and outputs. Will data be published as static tables, dynamic Web sites, or as a Web service? Will there be access control issues? How long will the data need to be preserved? Where might long-term preservation be housed and funded?
1.4 DEFINE SPECIFIC CONCEPTS TO BE MEASURED
Measures for dependent and independent variables, controls, and other classification measures need to be chosen. Measures that repeat across waves can be documented in a DDI ResourcePackage and used by reference. This practice can extend to using measures that have been used in other studies – documenting their
comparability. The specific universe to be measured may need to be defined.
1.5 PLAN, CREATE TIMETABLE & IDENTIFY NEEDED INFRASTRUCTURE
More detailed planning than in 1.2 above may be necessary either due to the complexity of the project or through the need to develop a formal proposal. A wide range of complexity of the planning process is possible, ranging from a relatively simple document to the development of detailed project management models. This step may also involve identification of potential funding sources and developing at least an informal timetable for the proposal process. Several funding sources now require a detailed data management plan as part of a proposal. Developing the data management plan may require some preliminary work in several of the “later” steps in the model. A data management plan may require a description of how the data will be captured, which in turn will require at least an initial consideration of sampling, instruments, data elements, and processing from steps 2, 3, and 5. The plan may require some consideration of data and metadata standards and file formats, archival setting, and distribution policies and procedures (steps 5, 6, and 7). Part of developing the data management plan will involve decisions about for how long the data will be preserved and at least preliminary arrangements with the organization which will ultimately preserve and curate the data.
1.6 PREPARE PROPOSAL & GET FUNDING
In many cases involving human subjects research, approval will be required from an institutional review board (IRB), necessitating a formal project proposal. Funding agencies will have their own requirements for the content of a proposal. Increasingly this may require a formal data management plan in addition to a detailed budget proposal. This step may require travel to meet with potential funders.