StandardsGroupReport

The work and the discussion of the standards group centered on four themes:

Dissemination

Planning and preparation of dissemination to encourage exploitation of previous results such as round-robin measurements.

• Studies should result in publications in refereed journals. This is important for community acceptance and to reach a wide audience.

• Facilities and users should learn about best practice by publishing recommendations and the web portal should point to papers and to other documents that describe the best practicesidentified by comparison between laboratories.

• Specific plans for publication on round-robin measurements on glassy carbon samples and on latex have been prepared and discussion between the authors started

Future Projects and Plans

Planning future co-operative measurement activities - round robin 'standards'

• Several fields were identified as likely to benefit from new round-robin activities. In practical terms it was thought best to initiate only two or three new projects immediately and that these should first be started by a small group of two or three participants. They can then be widened when initial materials and data are available.

• Grazing incidence scattering was seen as one priority for production of a reference sample suitable for both X-ray and neutron studies.

• Biological small-angle scattering would benefit from appropriate 'standard' samples that are understood by that community of users and are appropriate to measure in instrument configurations that used by that community.

In the longer term it will be useful to initiate project that can involve anomalous X-ray scattering and further samples to allow quantification and tests of the influence of resolution.

Uncertainties and Reliability

Major discussion concerned the importance of uncertainty in SAS measurements: both statistical counting errors and systematic deviations.

• Previous studies, particularly with the latex round-robin sample have identified that systematic errors are frequently more significant than the random errrors that arise solely from counting statistics. This needs to be more widely recognised in publications, data analysis procedures and even in guidelines for good practice.

• A longer term project to evaluate systematic errors and the uncertainty that derives from them is desirable. This will need gathering a group of appropriate people and invitations are being prepared.

Interaction with other canSAS activities

The group interacts closely to present appropriate ideas on standardisation and reliability topics to the other canSAS groups.

• Several of the ideas that have been discussed relate closely to work in other groups. New data formats should facilitate better description of uncertainty, the systematic recording of appropraite sample metadata and their uncertainties and the the reduction and analysis.

• The web portal should clearly identify good practices and provide appropriate training and educational material.

Acknowledgements

The results described arise from four full days of intensive work in small groups during the meeting in Uppsala and from preparatory work. We acknowledge the generous support of the agencies employing the particpants that have made available time and travel resources to permit the co-operative work. We are also grateful to people who have given freely of their own time to support the SAS community. Just the direct participation amounts to about two-person months of effort. These 'sponsors' are APS (Argonne National Laboratory), NCNR (National Institute of Standards and Technology), ISIS (STFC), the Diamond Light Source, Institut Laue Langevin, University of Auckland, Kyoto University, University of Tokyo and Uppsala University. We are also grateful for further participation from ANSTO, APS and ESRF in webinars and by making contributions to discussion on the wiki.