The management of data security is of critical importance for all data collectors and each one has comprehensive processes to ensure that access to data collection and dissemination systems is tightly controlled. Every instuitution will have ‘record contacts’ who have to manage colleagues’ access to these systems. Some of these account management systems are complex with various levels of permissions for each data collection.
For each data collection the burden associated with this activity is broadly consistent across institutions.
The work involved in building the data files and subitting them through the relevant web portal. These data collection systems now use standard web browser technology and are generally easy to use. Upload and system response times have improved significantly over the years.
The burden associated with this activity will be determined largely by the work necessary to extract/build the data file for submission. This can vary significantly according to the systems and processes that are involved.
Queries and review
Data collectors usually run multiple stages of validation and data checking and at each stage errors and queries are fed back to the institution for investigation and resolution. In many cases this work forms the bulk of the submission process and it is not unusual for institutions to make multiple resubmissions before all the issues are resolved.
There are two factors that influence the scale of this work. Failures in data quality by the instiutution – missing data or erroneous/impossible values – are relatively easy to trap in data collection systems and are usually easily understood. The burden associated with these issues will be driven largely by the quality of the data within the institution.
Instances where the submitted data does not meet the expectations of the data collecting organisation can be significantly more complex to investigate and often result from data that appears to be unusual but is not incorrect. These issues usually result from poor fit of the data model to the institutions reality or from an incorrect expectation by the data collector of what the institutions data should look like. The amount of burden associated with these cases is significantly random.
Most data collections involve some kind of final sign-off process. Some have very specific requirements about who has authority to sign-off the data and how and when that must be done. Some of these processes involve printing, signing and scanning sheets of paper. For each data collection the burden associated with this activity is broadly consistent across institutions.