Thompson, Paul2011-03-162011-03-161983Some missing data patterns for multidimensional scalingdoi:10.1177/014662168300700106https://hdl.handle.net/11299/101624Most tasks used to gather information for multidimensional scaling analysis are quite difficult for people to perform. An experiment was run to determine if systematic limits existed in such data collection situations and to determine the form that these limitations assumed. The solution obtained from a complete ordering of stimuli to targets, using the conditional rank order paradigm, was compared to solutions obtained from certain partial orders, constructed from the complete orders by setting certain rankings equal. The partial orders were found to reproduce the complete order solution quite accurately when about one half of the information was eliminated. The information eliminated about similar items produced more differences in the obtained solutions than did the information about dissimilar stimuli. Suggestions about efficient techniques for gathering information for multidimensional scaling purposes are discussed.enArticle