Depositors should review this list prior to depositing. If you have questions or concerns, please email icpsr-nacda@umich.edu
Ease of Understanding:
File and folder structure named with unique, descriptive titles; if providing multiple datasets, include dataset to documentation legend or name correspondingly Any publication or promotion deadlines related to the plan for the release of the data and materials Intended access type; public or restricted access If restricted access is needed, plan to discuss dissemination with archive and collaborate on terms of use agreement Do you already distribute data with us? If so, is this deposit an update to the data you supplied in the past? Please provide ICPSR study number and dataset relation.
Complete Documentation:
Original questionnaire Codebook; with unweighted frequencies is best for data comparison Identify and describe computed and derived variables Interviewer instructions User Guide Final Report Citations of related publications the data were used in Question text, DDI, or another text-source for the question Variable groupings (Especially for large datasets, it is useful to categorize variables into conceptual groupings)
Complete Study-Level Metadata:
Clear and consistent title across files and metadata. NOTE: to conform to metadata standards, the title will appear on the ICPSR (NACDA) site as Title, Location, Years, such as “Aging in Women [United States], 2005-2006”. Description of the project; goals, main topics, etc. PI full names and organizational affiliations (we reference this site to match names: Virtual International Authority File ) Data collection dates Intended unit of analysis (a description of who or what is being studied) Sample description Universe description Does the project/study have a website? If so, please provide it. Funding source and grant number
Complete Dataset(s):
Ensure variable names DO NOT contain characters that are not acceptable across stats packages, such as periods Ensure and names, addresses, phone numbers, and any other direct identifiers are removed Include complete, descriptive, and unique variable labels Include complete value labels (for SAS, a formats catalog is required) Missing designated or at least labeled clearly; if assigned in SAS or Stata, provide numeric codes that match the letter missing values for ease of conversion across stats packages Weighting information; weight variable names and syntax, if needed for replication Longitudinal weights; weight variable names and syntax, if needed for replication Identify merging variable(s) Provide necessary syntax for merging (based on origin stats package)
Please refer to our deposit pages for information about deposit options, tips, and expectations: