The Alignment Framework for Data Visualization: Relationships Among Research Goals, Data Types, and Multivariate Visualization Techniques

Dr. Yu, Chong Ho and Dr. John Behrens


Table of Contents

  1. Alignment Framework
    1. Dimensionality
    2. Noise-smooth continuum
    3. Research Goals and Visualization
    4. Data Type and Visualization
    5. Summary of the Alignment Framework
    6. Wickens' Theory and Studies
      1. Proximity Compatibility Principle
      2. Separable Display, Integral display and Configural display
      3. Serial Processing and Parallel Processing
      4. Focus Task and Integrative task
    7. Criticism to Wickens' Studies
      1. Pattern-seeking versus value-reporting
      2. Small number of data point
    8. Extension of Wickens' Studies
    9. Purpose and Hypotheses
    10. Method
      1. Design
      2. Subjects
      3. Materials
      4. Procedure
      5. Instrument
    11. Results
      1. Results of Outlier Identification Tasks
      2. Results of Relationship Examination Tasks
    12. Discussion
      1. Outlier Detection
      2. Relationship Examination
    13. References


Since the term "scientific visualization" was coined by a panel of the Association for Computing Machinery (ACM) in 1987 (McCormick, DeFanit, & Brown, 1987), both hardware and software developers have invested resources to enhance computer graphing capabilities. These developments can be seen as a logical extension of the work of John Tukey (1977, 1980, 1986a, 1986b, 1986c, 1988), who argued for the value of high dimensional graphics in social sciences.

While many scientific researchers believe that dynamic visualization is a promising data analysis tool for a variety of applications (e.g. Alexander & Winarsky, 1989), psychological studies concerning advanced graphical representation have been unable to support the effectiveness of these displays (Marchak & Whitney, 1990; Marchak & Marchak, 1991; Wickens, Merwin, & Lin, 1994). To address this discrepancy, this paper suggests a alignment framework to explain these different opinions, and proposes and reports the results of an empirical test of this explanation. The central thesis of this work is that visualization is only effective when there is an appropriate alignment of data type, task type (research goal) and type of graphical representation.

Among such researchers in the field of visualization, C. Wickens and his colleagues have established one of the most theoretically grounded and programmatic research endeavors (cf. Wickens, 1986; Carswell & Wickens 1987; Barnett & Wickens, 1988, Wickens, Todd, 1990; Goettl & Wickens & Kramer, 1991; Wickens, 1992; Wickens, Merwin & Lin, 1994). Building on his theory of the proximity compatibility principle, Carswell and Wickens (1987) predicted that integral displays are suited for integrative tasks while separable displays facilitate focus tasks. An integral display combines several dimensions in a single object whereas separable displays show data of different dimensions in different panels.

However, in several empirical studies this prediction was not fulfilled. Wickens interpreted these results as a failure of high-dimensional graphical devices in general. In this paper I argue that Wickens' negative results are not due to the general ineffectiveness of visualization procedures, but rather to the misalignment of data type, research goal (task type) and graphical type employed in his research. The supposition of this paper is that the quality of graphical display is dependent on the proper alignment of these three aspects of data visualization, as introduced in the following section.

Alignment Framework

I have recently proposed that effective data visualization is an outcome of the proper combination of graphical technique, research goal, and data type. In each of these three aspects, there are several sub-categories that must be considered as portrayed in Figure 1. Each of these aspects and sub-categories are discussed briefly below.

Figure 1. Taxonomy of three aspects of visualization

Taxonomy of Visualization Techniques

To make sense out of different graphical representations one can consider them in terms of dimensionality and noise-smooth (Yu & Behrens, 1995). As shown in Figure 2, visualization techniques can be considered to fall somewhere between two poles of the noise-smooth continuum, as well as the level of dimension. Noise level and dimensionality together dictate the complexity of the graph.

                                     One-dimensional
                                            |
                                            |
                                            |
                                     Two-dimensional
                                            |
                                            |
                                            |
               Noise--------------------------------------------------Smooth
                                            |
                                            |
                                            |
                                    Multi-dimensional

Figure 2. Noise-smooth continuum and level of dimensionality

Dimensionality

The number of dimensions of data is a common criterion for classifying graphical formats (e.g. Watson & Driver, 1983; Yorchak, Alison, & Dodd, 1984; Lee & MacLachlan, 1986; Barfield & Robless, 1989; Spence, 1990). According to these researchers graphical formats correspond to the number of variables, and therefore, can be categorized as one-dimensional, two-dimensional, and multi-dimensional.

The level of data dimension conditions the choice of graphing technique. For instance, a single histogram is often an effective summary of one-dimensional data (Scott, 1992). With bivariate data a scatterplot is a logical candidate for visualization because histograms cannot depict bivariate functions. Multivariate data sets usually require even more sophisticated visualization techniques.

Noise-smooth continuum

Apart from dimensionality, noise level is another useful criterion to classify graphical displays. Noise-smooth can be thought of as how much summary versus raw data is portrayed. When accuracy of prediction is required, data analysts have to display every single datum point (e.g. Palya, 1991). In many situations, however, depicting all data is not desirable, because too much information may hinder the researcher from unveiling the underlying data pattern. Therefore, the researcher often faces the tension between retaining and sacrificing detail.

In short, this taxonomy of visualization technique guides the researcher to contemplate questions such as "how many dimensions do I want to display?" "Should I reduce or integrate several dimensions?" Decisions about dimension reduction are often driven by the research goal and data type, which are discussed below.

Research Goals and Visualization

In addition to the preceding taxonomy of noise-smooth (raw data versus summary), I propose a classification of task type/research goal in terms of how much data need to be depicted. Generally speaking, there are six major categories of research: spotting outliers, discriminating clusters, checking distributional and other assumptions, examining relationships, comparing group differences and observing a time-based process. Only spotting outliers and examining relationships will be address here because they are examined in the proposed experiment. As the six tasks fall somewhere along the noise-smooth continuum, spotting outliers and examining relationships are located near the two opposite poles of the continuum i.e. spotting outliers requires focusing on all raw data whereas examining relationships focuses on a smooth structure underlying the data.

In contrast to the above taxonomy, Wickens et al. (1990, 1994) classified task type in terms of the degree of information integration. However, his categorization is of little practical use because pure focus and medium integration tasks such as reporting data values are seldom used in data visualization, and his high integration tasks emphasize on individual observations. This will be more fully discussed below in the section regarding Wickens' research.

Data Type and Visualization

Besides research goals, data type also affects the choice of visualization techniques. There are at least four relevant aspects of data type, namely, origin, format, complexity, and distribution. Data format is concerned with measurement scale while data distribution is related to normality and non-normality. In this study only data complexity, which is relate to Wickens' studies, will be discussed.

Data complexity is governed by the level of dimension, the number of observations, and the structure of the data. The need for dimension integration and noise filtering in graphics arise from the complexity of data sets. If all data sets are one-dimensional, contain as little as six to eight observations, and provide a nicely bell-shaped curve or a linear function, there will be little struggle between noisy and smooth displays. However, when the data set is multi-dimensional, contains hundreds or even thousands of observations, has non-normal distributions and non-linear relationships among variables, advanced visualization techniques should be used to deal with all three sources of data complexity.

Summary of the Alignment Framework

The proper combination of graphical technique, research goal and data type is crucial to a successful visualization. For example, if the task is to spot multiple outliers or discriminate clusters from a huge data set, a 3D plot that integrates three dimensions and depicts noisy data can be helpful. On the other hand, a smoothed graph such as a density curve or a histogram with large bandwidth runs the risk of hiding outliers. In light of the above taxonomies, we can predict under which circumstance particular techniques can succeed or fail. In the following section I will discuss the work of Wickens and his colleagues, and point out the mismatch of visualization technique, goal and data type in their research.

Wickens' Theory and Studies

Proximity Compatibility Principle

Wickens and his colleagues conducted a series of studies concerning the appropriateness of graphical display, task type, and the integration of dimensions. They termed their theory the "proximity compatibility principle." In this view, an optimal display should be both physically and perceptually proximate and compatible. Proximity is defined in terms of sharing of features between displayed attributes such as closeness in space, identity in color or similarity of semantic meaning (Carswell & Wickens, 1987).

Separable Display, Integral display and Configural display

Carswell and Wickens (1987) classified graphical displays into two categories based on proximity, namely, separable and integral displays. In separable displays, different dimensions of data are presented in separate graphical panels, and thus there is a low degree of proximity among different dimensions. In contrast, integral displays combine several dimensions into a single graphical object, and consequently entails a higher degree of proximity. In their terminology, an integral display is also referred to as an object display. For example, when there are two variables, X and Y, one may show their values by two bar charts as separable displays, or compress the information into a single object--a scatterplot with X and Y axes. When there are three variables--X, Y and Z, one may use three separate pairwise scatterplots (X,Y; X,Z; Y,Z), or employ a 3D plot integrating all three dimensions.

If different physical dimensions in a graph correspond to a single cognitive code, a graph is considered integral. A successful collection of dimensions would bring out an emergent feature that facilitates data analysis. However, occasionally these dimensions may be too artificial to be blended mentally. In such cases a graph that maintains separate perceptual codes in different dimensions is said to be configural (Carswell & Wickens, 1990).

Serial Processing and Parallel Processing

According to Wickens, in an integral display, the dimensions are perceptually compatible i.e. cognitive effort devoted to one dimension would lead to the increase of cognitive resources invested into the other. However, if the graph is separable, cognitive energy spent in one dimension is at the expense of the others. Using a computer analogy, the cognitive processing of separable graphs is thought of as serial processing while the integrated perceptual processing is viewed as parallel processing (Carswell & Wickens, 1987). Although parallel processing seems to be a good thing for human performance, this ability can be detrimental if parallel processing is mandatory rather than optional (Wickens, 1992).

Focus Task and Integrative task

According to the proximity compatibility principle, the merits of separable and integral displays, as well as serial processing and parallel processing, are dependent on the task nature--focus tasks versus integrative tasks. If the visualization task requires a focus on independent dimensions such as locating individual data values, a separable graph that is likely to facilitate serial processing is desirable. If the job demands a judgment based upon integrated information from various dimensions, then an integral graphical object, which can enhance parallel processing, should be superior to a separable one.

Wickens and his colleagues conducted a series of experiments to verify these predictions. Throughout these experiments, the proximity compatibility principle was only partially supported. In a recent study aimed at simulating scientific visualization, Wickens, Merwin and Lin (1994) found that certain integral displays were not more supportive than separable ones in integrative analysis. This study consisted of two experiments. The first one was a comparison between 2D (separable) and 3D (integral) displays with six data points on each graph whereas the second one examined the effectiveness of stereo, mesh and rotation with eight data points on each graph. While the first experiment was similar to Wickens et al.'s previous studies, the second one carried new aspects of greater consequence. In the second experiment, three types of display were employed--stereopsis, 3D plot with the feature of rotation, 3D plot with a mesh surface. A still 3D plot was used as the control condition. The task types were also classified in terms of the degree of information integration:

1. Pure focus task, e.g. What is the earnings value of the blue company?

2. Integration across dimensions in one observation, e.g. Is the green company's debt value greater than its earnings value?

3. Integration across observations in one dimension, e.g. How much greater is blue's price than red's price?

4. Integration across dimensions and observations, e.g. Which company has the highest total value of all three variables?

It was found that the long term retention of abstract knowledge of the data failed to benefit from the 3D display exposure. Also, the rotation of a 3D graph and the presence of a mesh surface did not support performance in integrative processing. Wickens et al. stated that "it is an article of faith in many scientific visualization products that scientists should be able to explore their data interactively" (p.47) while in their empirical study animated motion did not provide any benefit for understanding data.

Criticism to Wickens' Studies

I argue that the major problem of Wickens' studies is a mis-match of graph, task, and data.

Pattern-seeking versus value-reporting

Wickens classified research goal/task type in terms of degree of information integration. In practice, graphical presentations are rarely useful in pure focus tasks and medium integration tasks such as reporting single values. Several researchers pointed out that graphs should be used to convey an overall pattern while tables are better for looking up data points (Ware and Beatty, 1986, 1988; Kosslyn, 1994).

It is doubtful whether focus tasks and medium integration tasks should be implemented with visualization at all. For instance, in the study reported above (Wickens, Merwin & Lin ,1994) the focus question is: "What is the earning value of the blue company?" This question can be easily answered by simply looking up the values on a spreadsheet. Even in the supposed high integration task, the focus is still on individual values rather than the relationships among variables: "Which company has the highest total value of all three variables?" Indeed it is more efficient to sum the values of all three variables, sort the data by summed values, and then list all the cases.

Although a 3D spin plot can be used for closely examining particular observations such as detecting outliers, the major concern of spotting outliers is the relative distance between the extreme cases and the majority of the observations. In other words, the analyst cares more about the overall data structure than individual cases. However, in Wickens et al.'s experiment, the focus on individual observations led subjects away from the global picture. Accordingly, the graphical format of the 3D spin plot and the task type of value-reporting are misaligned and poor performance should be expected.

Similar concerns arise when Wickens employed the 3D mesh. A 3D mesh is usually a smoothed surface. However, in Wickens et al.'s graph only individual data points are connected. It is qualified to be a surface plot or a perspective plot, rather than a mesh plot. A surface plot is not appropriate for aiding data value reporting, because in a large data set the surface would appear to be rough and tracing the exact co-ordinate is difficult. Even in a small data sets as Wickens et al. portrayed, exact co-ordinates are difficult to perceive. Again, Wickens et al.'s experiment tasks represent a misalignment of research goal and graphing technique.

Tasks given in Wickens et al.'s experiments, ranging from pure focus to so-called high integration, are not compatible with the goal of pattern seeking in data visualization. Given all the preceding misalignments, it is not surprising that Wickens et al. found no advantage of 3D spin and 3D mesh, because they are the wrong tools for the assignments that concentrate on individual values.

Small number of data point

It is important to note that in most of Wickens' studies the data sets in the experiments were as small as two to eight. As mentioned in the taxonomy of graphical format, the choice of visualization techniques are often driven by the goal of handling massive and messy data. Triangle displays in the first and second studies have little practical value because in most situations either the number of variables or observations would exceed three.

For example, in the experiment discussed above, Wickens reported no advantage for use of a 3D plot. This is not unexpected under the circumstances described by Wickens, because 3D plots were developed primarily to solve the problem of overplotting and perspective limitation in multiple dimensions. In this case of small sample size, no overplotting and viewpoint obstruction occured and the advantages of the 3D plot could not be gained. In this way Wickens' tasks represent a misalignment of the plot made for high volume data with a very small data set.

The size of the data set is also an issue determining the appropriate use of a mesh surface. When there are many observations in a 3D plot, the perception of trend, which is based upon our perception of depth, is not easily formed. In this case connecting neighboring points to construct a mesh surface is helpful because it can provides depth cues. Moreover, if there are many peaks and holes in the plot, a smoothed mesh surface derived from a function is also desirable. In the third study reported above Wickens et al. (1994) found no advantage to this plot, but again, they had not properly aligned the plot with the type of data set size appropriate for the plot. An advantage of the type Wickens might have expected is only likely to occur when the size of the data set is large--an attribute that did not hold when only eight data points were used.

Extension of Wickens' Studies

This study tested these suppositions by exposing subjects to aligned and mis-aligned graph-task-data situations. This study is an extension of Wickens et al.'s study (1994) with three modifications. First, the interest of this study is on 3D spin and 3D mesh plots rather than stereopsis displays. This is because stereopsis displays are rare and the availability of 3D plots is increasing greatly. Second, the data sets to be used in the experiment will be composed of eight, fifty and one hundred cases, respectively, in contrast to only six to eight cases in Wickens et al.'s study. This is important because a large data set is more realistic in data visualization work and effective plot use depends on appropriate alignment with data set size. Third, as in Wickens' study the task types (research goals) will also be classified as obtaining an overall impression of data or dealing with individual cases, the latter task type will be spotting outliers instead of recalling values. As mentioned above, focused and medium integration tasks such as reporting values should be accomplished with the use of tables. Therefore in this study spotting outliers, which is a highly integrative task, will be used instead. Outlier detection is considered highly integrative for two reasons. First, in a multivariate case, outliers are defined in terms of their coordinates in more than one dimensions, and therefore spotting outliers demands a high level of information integration. In addition, for detecting outliers an analyst is interested in the relationship between the extreme cases and the majority of the data. Thus, the task is global in nature.

Purpose and Hypotheses

In light of the alignment framework presented above, it is expected that superior performance would be found for 3D spin and mesh plots for outlier detection and relationship examination across medium and large data sets.

Method

Design

This study used a 3 data-size X 3 plot-type X 2 research-goal completely within-subjects factorial design. The factor of data size has levels of small (eight cases), medium (50 cases), and large (100 cases). The factor of graphical format includes 2D, 3D spin and 3D mesh plots with spotting outliers and examining relationships as the levels of the research goal/task.

Subjects

The sample is twenty-three graduate students with prior coursework in quantitative research methods, including low-dimensional data visualization, at Arizona State University.

Materials

Nine different fictitious data sets were randomly constructed for the experiment. All were tri-variate data sets with three each consisting of one hundred, fifty, and eight cases. In this experiment participants were asked to spot outliers and examine relationships. Thus, the data sets were designed to match the task nature. As indicated in the preceding functions, two-way interaction effects were present in the data sets in order to make the conditional relationships among the three variables fairly complicated.

Each data set was displayed using three types of graphs on Macintosh computers. These graph type were: (a) A set of 2D scatterplots portraying three variables in a pairwise manner, (b) a 3D plot portraying a cloud of data points with a spin option, (c) a 3D plot with a mesh surface conforming to the underlying function is shown in Figure 3.

Figure 3. 3D mesh plot

Procedure

The experiment was administered in several computer labs at the Arizona State University. One to five subjects participated in the experiment at a time.

In this experiment I will refer to a combination of graph(s) and question as a scenario. Subjects were exposed to all eighteen scenarios as described in the design section. In order to avoid carry over effects, the data used in each scenario were randomly drawn from data sets. The order of scenarios was likewise randomized for each individual.

For each scenario, subjects viewed a graph or several graphs, and a dialog box with a multiple-choice question. They were told to answer the question according to the information shown on the graph(s), and were permitted to manipulate the graphics as appropriate to the graph type. After an answer was selected, another set of graph(s) and problem were presented. The process ended when all eighteen conditions were exhausted. The subjects were allowed to explore the data and answer the online question for about thirty minutes. Afterwards, they repeated the same procedure with a different sequence of the scenarios and different data sets drawn randomly.

Instrument

An objective test, which was embedded within the computer program, was given to the subjects as they were exposed to each condition. The test consisted of eighteen questions--one for each scenarios. The formats of Question 1 to 9, which was concerned with outlier detection were the same but the outliers varied. For the task of spotting outliers, the graph showed one observation each colored red and green. Participants chose one of the following answers: (a) the red observation is an outlier, (b) the green observation is an outlier, or (c) there is no outlier.

For the task of examining relationship, participants were given values of two variables and asked for a third. Again, the subjects had three choices: high, medium, and low.

Results

Tests of repeated measures analysis of variance (ANOVA) are reported using both univariate and multivariate statistics. In order to verify these results assuming a binomial model, logistic regressions with exact tests and 95 percent confidence intervals of proportion were computed.

Results of Outlier Identification Tasks

For the repeated measures ANOVA, the response variables consisted of the mean correctness of each question across Test 1 and 2. As seen in Table 1, in both univariate and multivariate statistics the graphical format factor yielded a significant effect, as was the interaction between graph type and data size. The sample size effect was not significant.

Table 1

Repeated Measures ANOVA for Outlier Detection

                                Univariate                                
Source              df1  df2    MSe             F             p           
Graph Type            2  44     .1280           36.78         .0001       
Data Size             2  44     .0984           1.14          .3287       
Graph * Data          4  88     .1030           4.55          .0022       
                                Multivariate                              
Source              df1  df2    MSe             F             p           
Graph Type            2  21                            33.12  .0001       
Data Size             2  21                             1.05  .3634       
Graph * Data          4  19                             3.15  .0380       

A logistic regression with exact test using Test 1 scores found a significant graph effect for the medium sample size and the large sample size. However, no significant graph effect was found for the small data size. Exact tests using Test 2 scores yielded significant results across all three sample sizes. Summary of exact tests are reported in Table 2.

Table 2

Results of Exact Tests for Outlier Detection

Sample size                     p  
            Test 1                 
Small                       .1104  
Medium                      .0011  
Large                       .0004  
            Test 2                 
Small                       .0186  
Medium                      .0037  
Large                       .0001  

The confidence intervals of proportion regarding outlier detection are shown in Figure 4. Patterns of Test 1 and 2 were slightly different. For the small sample size of Test 1, the confidence bands of three types of graphs overlapped, and are not distinguishable in terms of performance. For the small data size of Test 2, however, 3D mesh plots were superior to 2D scattergrams. For the medium data size of Test 1, 3D spin plots were superior to 2D plots. Nonetheless, for the medium data size of Test 2, performance difference between 3D spin plots and 2D graphs were trivial.

a. Test 1

		small		   medium		large

b. Test 2

Figure 4. Confidence intervals of graphs for outlier detection

Results of Relationship Examination Tasks

For the task of relationship examination, significant effects were found for size, graph and the size*graph interaction. Univariate and multivariate statistics are reported in Table 3.

Table 3

Repeated Measures ANOVA for Relationship Examination

                                Univariate                                
Source              df1    df2             MSe             F           p  
Graph Type            2     44           .1290         32.69       .0001  
Data Size             2     44           .0883         16.37       .0001  
Graph * Data          4     88           .1092          4.08       .0051  
                                Multivariate                              
Source              df1    df2                             F           p  
Graph Type            2     21                         37.99       .0001  
Data Size             2     21                         11.26       .0005  
Graph * Data          4  19                     8.93          .0003       

Exact tests reported here, which were stratified by subjects, were analogous to those of repeated measures ANOVA. In both Test 1 and 2 significant results were found in small and medium sample sizes. However, in the large sample size of both tests the three graphical formats did not differ from each other. Summary of exact tests are reported in Table 4.

Table 4

Results of Exact Tests for Relationship Examination

Sample size                     p  
            Test 1                 
Small                      .00001  
Medium                     .00250  
Large                      .38970  
            Test 2                 
Small                      .00010  
Medium                     .00010  
Large                      .17030  

As shown in Figure 5, confidence intervals of proportion pertaining to relationship examination resemble those of repeated measures ANOVA. Here differences in effectiveness of graph type were congruent with those illustrated in Figure 5a. In the medium data size of Test 1, variation of performance using 3D spin plots and 2D graphs were indistinguishable. This pattern did not hold in Test 2, in which 3D spin plots had an advantage over 2D plots.

		small		   medium		large

a. Test 1

		small		   medium		large

b. Test 2

Figure 5. Confidence intervals of graphs for relationship examination

Discussion

The results largely conform to the predictions derived from the alignment framework and substantiate the effectiveness of 3D graphics for data visualization with appropriate data and tasks. These findings counter the conclusions of Wickens et al. (1994) that 3D graphs do not have an advantage over 2D graphs for scientific visualization. In this study 3D graphics outperformed their 2D counterparts for both outlier detection and relationship examination across all sample sizes. While I hypothesized that the superiority of 3D displays would be realized in medium and large data sets only, it turns out that the superiority of 3D graphics was also present in small data sets.

Outlier Detection

In both medium and large data sets, the results conform to the prediction that 3D graphics are better than 2D ones for spotting outliers. In both cases 3D mesh was as effective as 3D spin. Based on my review of the literature, it was expected that mesh surfaces are used primarily for examining relationships rather than detecting outliers. Nonetheless, it seems that users found the mesh surface helpful as a depth cue, which is an important feature for determining distances among observations.

Scores of Test 1 and 2 were computed separately for confidence intervals. In the small sample size of Test 1, performance among the three types of graph were indistinguishable. In the medium sample size performance of 3D spin was even better than that of 3D mesh plots. However, in Test 2 3D mesh plots led to superior performance than 2D plots in both small and medium sample sizes. Also, in the small data size of Test 2 the difference between 3D spin and 3D mesh plot approached significance. Further, in the medium data size of Test 2 the relationship between 3D spin and 3D mesh plots was opposite to that in Test 1 i.e. 3D mesh plot outperformed 3D spin plot. This is consistent with the idea that students improved their skill for interpreting 3D mesh plots after some practice.

Relationship Examination

Expectations were fulfilled in both medium and large data sets. According to correlated t-tests, 3D mesh plot was the best for the task of relationship examination, 3D spin, the second best, and 2D plots, the worst. In the large data set, the mean score of 2D plot was slightly higher than that of 3D spin plot. The possible explanation of the equal performance between 2D and 3D spin plots in the large data set is that when there were adequate observations on the graph, the pattern became obvious even in 2D scattergrams.

It was found that virtually in all sizes of data sets, performance improved as the sample size increased. The exception is that in 2D plots scores of small and medium sample sizes were exactly equal. One explanation is that the large data set provided enough observations to form a more obvious function while the small and medium data sets failed to suggest a pattern.

It is interesting to notice that in both outlier detection and relationship examination, sample size did not make a significant difference in 3D mesh. However, the large data set was still the best, the medium, the second, and the small, the worst. It is concluded that regardless of sample size, a mesh surface was helpful in examining relationships among variables and also detecting multiple outliers.

Most of the findings in this study confirm the notion that the usefulness of visualization technique is tied to the task nature and the data type. Although the effectiveness of 3D mesh in the small data set was not foreseen, this result further supports the alignment framework rather than weakening it. At first it was argued that the ineffectiveness of 3D graphics in Wickens et al.'s studies (1994) is due to the use of a few observations on the graphs. However, this study counteracted Wickens' conclusion because even with a small data set 3D graphics are still superior to 2D ones.


References

Alexander, J., & Winarsky, N. (1989, April). Interactive data visualization. Information Display, 14-22.

Barfield, W., & Robless, R. (1989). The effects of two- or three-dimensional grphics on problem-solving performance of experienced and novice decision makers. Behavvior and information Technology, 8, 369-385.

Barnett, B., & Wickens, C. (1988). Display proximity in multicue information integration: The benefits of boxes. Human Factors, 30, 15-24.

Carswell, M., & Wickens, C. D. (1987). Information integration and the object display: An interaction of task demands and display superiority. Ergonomics, 30, 511-527.

Carswell, M., & Wickens, C. D. (1990). The perceptual interaction of graphical attributes: Configurality, stimulus homogeneity, and object integration. Perception and Psychophysics, 47, 157-168.

Casey, E. J., & Wickens, C. D. (1986). Visual display rep[resentation of multidimensional systems (Tech. Report ARL-90-5/AHEL-90-1). Savoy, IL: University of Illinois Willard Airport, Aviation Research Laboratory, Institute of Aviation.

Goettl, B. P., Wickens, C. D., & Kramer, A. F. (1991). Integrated displays and the perception of graphical data. Ergonomics, 34, 1047-1063.

Kosslyn, S. M. (1994). Elements of graph design. New York: W. H. Freeman Company.

Lee, J. M., & MacLachlan, J. (1986). The effects of 3D imagery on managerial data interpretation. Management Information Systems Quarterly, 10, 257-269.

Marchak, F. M., & Marchak, L. C. (1990). Dynamic graphics in the exploratory analysis of multivariate data. Behavior Research Methods, Instruments, & Computers, 22, 176-178.

Marchak, F. M., & Marchak, L. C. (1991). Interactive verus passive dynamics and the exploratory analysis of multivartiate data. Behavior Research Methods, Instruments, & Computers, 23, 296-360.

McCormick, B. H., DeFanti, T. A., & brown, M. D. (Eds.). (1987). Visualization in scientific computing [Special issue]. Computer Graphics, 21(6).

Palya, W. (1991). Laser printers as powerful tools for the scientific visualization of behavior. behavior Research Methods, Instrument, & Computers, 23, 277-282.

Scott, D. W. (1992). Multivariate density estimation: Theory, practice, and visualization. New York: John, Wiley & Sons.

Spence, I. (1990). Visual psychophysics of simple garphical elements. Journal of Experimental Psychology: Human Perception and Performance, 16, 683-692.

Tukey, J. W. (1977). Exploratory data analysis. Reading, MA: Addison-Wesley Publishing Company.

Tukey, J. W. (1980). We need both exploratory and confirmatory. The American Statistician, 34, 23-25.

Tukey, J. W. (1986a). Data analysis and behavioral science or learning to bear the quantitative man's burden by shunning badmandments. In L. V. Jones (Ed.), The collected works of John W. Tukey, Volume III: Philosophy and principles of data analysis: 1949-1964. Pacific Grove, CA: Wadsworth.

Tukey, J. W. (1986b). The collected works of John W. Tukey, Volume III: Philosophy and principles of data analysis: 1949-1964. L. V. Jones (Ed.). Pacific Grove, CA: Wadsworth.

Tukey, J. W. (1986c). The collected works of John W. Tukey, Volume IV: Philosophy and principles of data analysis (1965-1986). L. V. Jones, (Ed.). Pacific Grove, CA: Wadsworth.

Tukey, J. W. (1988). The collected works of John W. Tukey, Volume V: Graphics. W. S. Cleveland, (Ed.). Pacific Grove, CA: Wadsworth.

Ware, C., & Beatty, J. C. (1986). Using color to display structures in multidimensional discrete data. Color and Resereach Application, 11, S11-14.

Watson, C. J., & Driver, R. W. (1983). The influence of computer graphics on the recall of information. Management Information Systems Quarterly, 7, 45-53.

Wickens, C. D. (1986). The object display: Principles and a review of experimental findings (Tech. Report CPL-86-6/MDA903-83-K-0255). Champaign, IL: University of Illinois, Cognitive Psychphysiology Laboratory.

Wickens, C. D. (1992). Engineering psychology and human performance (2rd ed.). New York: Harper Collins Publisher.

Wichens, C. D., & Todd, S. (1990). Three-dimensional display technology for aerospace and visualization. In Proceedings of the Human Factor Society 34th Annual Meeting (pp.1479-1483). Santa Monica, CA: Human Factors and Ergonomics Society.

Wickens, C. D., Merwin, D. H., & Lin, E. L. (1994). Implications of graphics enhancements for the visualization of scientific data: Dimensional integrity, stereopsis, motion, and mesh. Human Factors, 36, 44-61.

Yorchak, J. P., Allison, J. E., & Dodd, V. S. (1984). A new tilt on computer generated space situation displays. In Proceedings of the Human factors Society 28th Annual Meeting. Santa Monica, CA: Human Factors Society, 894-898.

Yu, C. H., & Behrens, J. T. (1995). Applications of scientific multivariate visualization to behavioral sciences. Behavior Research Methods, Instruments, and Computers, 27, 264-271.