文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Bayesian inference with probability matrix decomposition models
侵权投诉

Bayesian inference with probability matrix decomposition models

check, psychometrics Probability Mcatrix Decompositioni models mtay bve uised to model observed binary associations between two sets of elements. More specifically, to explain observed associations betweeni two elements, it is assumed that B laitent Bernou

Meuld/es et (it.

Probability Matrix Decomposition (PMD) models have beeni introduced by Maris, De Boeck and Van Mechelen (1996) as a method of data analysis for two-way two-nmode frequency data (using Tucker's termninology of types of data, the rows and the columns of the data see Carroll& Arabie. 1980). In general, refer to different types of elemnents, denoted as objiest and (attributes.In matrix nmost applications of PMD models, the observed freqtuenicies reflect the number ol raters who have the opinioIn that objects and attributes are related. However,

o)ne may also consider other types of repeatedl measuremenits, such as the ju

dgments of one rater at different occasions. In these cases we use the term replications for repeated measurements. By making a specific choice of ob?jects, attributes, and ieplicationis a wide variety of pheniomenia in educational and

behavioral -research rmay be inivestigatedi. For- instance, Maris, De Boeck. and Van Mechelen (1996) used PMD models to study the process of psychiatric diagnosis. In this study, different clinicians had to judge whether patients have a certaini symptom. Candel and Mans (1997) illustrated the use of PMD mnodels in marketing research and nade a comparison with other methiods of data analysis for two-way two-mode frequency data, nameley, latent class analysis and correspondence analysis. To illustrate the use of PMD imodels in marketing research they asked respondents to judge whether products have a certain attribute. De Boniis, De Boeck, Perez-Diaz. and Nahas (1999) used PMD models to study emotion perceptioni in facial expressions. These authors asked raters to jusdge whetlher they perceive a certain emnotioni in a photographed iacial expression. in addition, PMD models

may also be useful in

personality research. For instance, in order to analyze the hostile belhavior of people in different frustrating situations persons were asked to judge whether they would display a certain hostile reaction in a certain frustrating situation

(Meulders, De Boeck,& Van Mechelen, 200(0). Furthermore, PMD m,odels could also be used to analyze some br,oacd categories of judgments that are often studied in educationial and behavioral research. For instance. they could be used in the context of social network analysis (Wasserman& Faust, 1994) to explain thne ties between mebenbers of a network oni sonic social relation (e,g. frienidlship, communticating, etc.). TlLhe two-way two-mode data matrix to be used in such analysis may be obtained by asking each actor in the network to report all those to wnorn hie or she is tied, on the social relationship in question, and by afterwards counting the nunmber of ties between each pair of actors. Another research area where PMI) nmodels could be meaningfully applied is that of consumer behavior. More specifically, one could record the nrumiber of times consumers buy a certain product and analyze the resulting consumers by prodiucts imiatrix. As a final example, PMD models could be used in the context of educational measurement to study the extent to which ptipils imiaster certain attairment targets. in this case the data may reflect the

numbear of judges who indicate that a particular pupil masters a certain attainment target.

154

Bayesian inference with probability matrix decomposition models

上一页第2页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email