The multivariate linear model is

His t o r y Co r n e rThe Histor y of the Cluster Heat Map Leland WILKINSON and Michael FRIENDLY The cluster heat map is an ingenious display that simultaneously reveals row and column hierarchical cluster structure in a data matrix. It consists of a rectangular tiling, with each tile shaded on a color scale to represent the value of the corresponding element of the data matrix. The rows (columns) of the tiling are ordered such that similar rows (columns) are near each other. On the vertical and horizontal margins of the tiling are hierarchical cluster trees. This cluster heat map is a synthesis of several different graphic displays developed by statisticians over more than a century. We locate the earliest sources of this display in late 19th century publications, and trace a diverse 20th century statistical literature that provided a foundation for this most widely used of all bioinformatics displays.

Imagery and concreteness norms and percentage noun usage were obtained on the 1,080 verbal items from the Toronto Word Pool. Imagery was defined as the rated ease with which a word aroused a mental image, and concreteness was defined in relation to level of abstraction. The degree to which a word was functionally a noun was estimated in a sentence generation task. The mean and standard deviation of the imagery and concreteness ratings for each item are reported together with letter and printed frequency counts for the words and indications of sex differences in the ratings. Additional data in the norms include a grammatical function code derived from dictionary definitions, a percent noun judgment, indexes of statistical approximation to English, and an orthographic neighbor ratio. Validity estimates for the imagery and concreteness ratings are derived from comparisons with scale values drawn from the Paivio, Yuille, and Madigan (1968) noun pool and the Toglia and Battig (1978) norms.The Toronto Word Pool (TWP) is a collection of 1,080 common English words originally selected from the Thorndike-Lorge (1944) word counts (see Murdock, 1968Murdock, , 1974. This pool has been used for some time in a number of laboratories in verbal learning, memory, and psycholinguistics studies, although normative data on the items have never been collected.In experiments that require random selection of a number of lists in order to exclude list-specific effects or to assure sufficient sampling of materials, it is particularly useful to have normative data available. Thus, information on item frequency allows lists to be matched or balanced on this variable. Similarly, if a moderate

To investigate the properties that make a word easy to recall, we added to existing norms for 925 nouns measures of availability, goodness, emotionality, pronunciability, and probability of recall in multiple-trial free recall. Availability, imagery, and emotionality were found to be the best predictors of which words were recalled. This result, which is stable across recall data collected in three separate laboratories, argues for the importance of availability as apredictor of recall and questions the role of the correlated variables of word frequency and meaningfulness. Consistent with earlier work on a smaller sample of words, six factors describe the numerous properties of words studied by psychologists. The six factors are composed of variables based on orthography, imagery and meaning, word frequency, recall, emotionality, and goodness.What factors determine whether a particular word will be recalled by a particular subject under a specific set of experimental conditions? Over the history of the study of memory and learning, four separate aspects of the recall task have received attention (Jenkins, 1979): (1) the nature of the material to be learned and the conditions of presentation; (2) the type of response or criterion task used; (3) the activities engaged in by the learner during study and at test, including depth of processing, rehearsal, and mnemonic strategies; and (4) the characteristics of the learner, including general knowledge, skills, and motivation.In the study presented here, we addressed the first aspect of recall tasks, the nature of materials and the conditions of presentation, by trying to predict the probability of recall of individual words when the other three aspects of the recall task were randomized or held constant. Most of our emphasis was on the material itself because here, and in other studies (Rubin, 1980(Rubin, , 1985, presentation effects were minimal.Consider a typical study investigating one or more item variables, such as frequency, concreteness, or age of acquisition. Words are selected to represent different portions of the range of the variables of interest (e. g., high frequency vs. low frequency). The recall performance of subjects learning these words is then compared. For our purposes, this form of experiment has four disadvantages. First, in selecting fixed lists of stimulus items, the researcher must deal with the confounding relationships between the variable(s) under study and other correlated variables. Holding the correlated stimulus variables constant eliminates the confound, but may produce lists of items which are not representative of the variable being studied, or which are confounded in new ways. For example, Paivio, Yuille, and Madigan (1968) noted that varying imagery while holding concreteness constant results in a set of words that varies in emotionality. Second, so many properties of the experimental procedure (e.g., presentation time, modality, etc.) and the subject population are held constant that the generality of the results may be questio...

No abstract