Members of Assess will remember that we have been experimenting with our workshop formats for some time. Following up a suggestion from one member, ASSESS joined with SPSS (UK) Ltd. to run a joint workshop for the first time in June. Our usual frustration over finding a suitable venue in the South ended when SPSS (UK) Ltd. kindly offered us the use of their accommodation in St Andrews House in Woking. Twenty-eight delegates arrived nice and early to start a day with a packed and varied programme aimed at Power Users of SPSS. 

Paul Callow from the Computing Service, University of Cambridge, started the day with a talk on ‘Scripting or Macros: which to use and when?’ We know many members have been using SPSS macros for some time- and indeed have attended Paul’s previous workshops on the subject. His talk offered a useful introduction to scripting, which, like macros, offer users considerable help with tedious and repetitive tasks. 

Jon Petersen, VP Product Sales and Marketing, SPSS International, followed with a discussion of the thorny topic of the use of complex sampling in those situations where elements in the sample are known to have unequal probabilities of being selected. He finished with an introduction to WesVar Complex Samples 3.0 which, unlike many conventional packages, can produce the correct estimates for standard errors and variances of statistics when complex sampling techniques have been used. After lunch, Julie Amiss, Services Manager, SPSS (UK) Ltd., gave a demonstration of ‘Trial Run. This is a Design of Experiments program which includes, amongst many other features, guidance towards an appropriate experimental design, a range of General Linear Modelling algorithms, support for 41 experimental designs and very easy to use procedures for the production of surface and contour plots. 

The last talk, on Amos, was given by Jeremy Miles, from the Applied Vision Research Unit, Derby University. Jeremy started with pen portraits of the champions and producers of Structural Equation Modelling (SEM) programs. Their products were serious and competent but perhaps not very user friendly! AMOS, Jeremy suggested, is very user friendly. It has an intuitive user-interface, which involves specifying models by drawing path diagrams. This makes it much easier for analysts, who might have hesitated to take their analyses further, to venture into the powerful world of SEM. 

As usual, we asked delegates for their comments about the workshop, so I’ll let one of them sum up. 

‘Overall, a good day! Let’s do another one soon.’