### Table 1: Distribution of student grades in a particular paper

"... In PAGE 3: ... Hence a statistical monitoring procedure of the student grade distribution is needed to detect the presence of such special cause variations. The distribution of grades for six different consecutive offerings of one particular paper is shown in Table1 . Evidently the percentage of A grades is above the guideline limit of 17% for the first three time periods.... In PAGE 4: ... The midpoint of the guideline grade ranges will be used to obtain the expected counts. For example, the expected number of A grades for 100 passing students is 15 (for the first period shown in Table1 ). This expected count is then used to compute the chi-squared statistic for the Grade A category as (18-15)2/15 = 0.... In PAGE 5: ... Hence a control chart with a false alarm probability of 0.05 can be drawn for Table1 data as shown below: 0 1 2 3 4 5 6 7 123456 Period Chi-squared measure of deviation Figure 1: Chart for monitoring student grades Obviously the above chart does not signal the presence of a special cause variation in grading for the paper after all. Unfortunately, a signal rule based on a single plotted point is not particularly sensitive to small upward or downward deviations from the grading guidelines.... In PAGE 5: ... That is, the faculty is not taking any corrective steps for successive paper offering has to be viewed seriously. Figure 3 provides such a control chart for Table1... ..."

### Table 1 Comparison of the several fitting methods investigated in this paper on the basis of the variances for tuning curve values in a particular stimulus condition Distribution Variance of tuning

"... In PAGE 6: ... In practice, this distribution can only be approximated. However, as our simulations show, the introduction of a single outlier into a Gaussian distribution, for which the least-squares fit would otherwise be optimal, can change the picture totally (see Table1 ). Using our method, we need not calculate an op- timal fitting method for every distribution of the residuals, since we provide a method that performs well across a wide range of residual distributions.... In PAGE 6: ... Stability of the fitting method: numerical simulations The stability of the fit of a tuning curve across several repetitions of the entire experiment depends on the type of noise distribution, and on which approximation criterion is used (see Section 3) to approximate the ideal tuning curve. Table1 shows simulation results for the variance of es- timators for the mean. These simulations are for slightly contaminated normal distributions.... ..."

### Table 8.2. Distributions (at least 30%) of class values in particular clusters

2007

### Table 3: Performance of the fast evaluation method (d = 2). The N centers are positioned within 1=10 unit along the diagonal of the unit square and the M = N evaluation points are are uniformly distributed in the unit square. This test case highlights the fact that the method does not appear to be sensitive to the particular distribution of centres and no observable di erences are noted to the uniformly distributed case. The table shows times in seconds for the approximate and direct evaluation of the Hardy Multiquadric interpolants.

2004

### Table 2 The Probability of a particular sample value exceeds a predefined threshold x for a normal distribution.

"... In PAGE 7: ... The bad news is that it is not a very tight upper bound. Table2 shows some examples ... ..."

### Table 1. Credential distribution used in the evaluation scenario. The variable F rep- resents a particular funding agency.

2006

"... In PAGE 16: ...2N of the hosts are resources. Users and resources are randomly generated and assigned credentials and credential release policies in accordance with Table1 ; in situations where multiple release policies are indicated for a single credential type, one is chosen uniformly at random for each credential generated. Resources are also assigned resource access policies according to Table 2.... In PAGE 22: ...2. According to Table1 , users can have at most 19 credentials described by the ontology shown in Figure 2; resources can have at most 10 credentials described by this ontology. Note also that only resources will have BBB or PrivacyPolicy credentials.... ..."

Cited by 2

### Table 1. Regression estimates of power law: log y = log a + b log x demonstrate that the power law holds but the particular distribution changes over time in no systematic way. The exponent b is not significantly different in the trading cycle 29 from the value in trading cycle 19, but they are significantly diggerent a 90% confidence from either of the other two which are themselves significantly different.

### Table 3 illustrates the resulting mapping of SEEDS to the MLM model, when 2D and 3D visualisation aspects are of particular interest; it can be seen that for the interfaces and distributed objects, both CORBA and DIS entities have to be considered.

"... In PAGE 9: ... Table3 . Mapping of SEEDS components to the MLM model 0-7695-0981-9/01 $10.... ..."

### Table 2: Transaction Characterization: The # rows indicate the number of data points in a particular distribution (e.g., number of complete transaction in a given trace); in the case of the average column, the # entry represents the total number of data points in all of the traces. The (percentile) lines indicate the percentile within the distribution that the average represents (e.g., the 5.864-second average complete transaction duration for the Transport trace represents the 95th percentile of that distribution.) Complete Transactions refer to those which conclude with an end record. All transactions includes those which had not ended at the end of the trace.

1996

Cited by 8

### Table 2: Transaction Characterization: The # rows indicate the number of data points in a particular distribution (e.g., number of complete transaction in a given trace); in the case of the average column, the # entry represents the total number of data points in all of the traces. The (percentile) lines indicate the percentile within the distribution that the average represents (e.g., the 5.864-second average complete transaction duration for the Transport trace represents the 95th percentile of that distribution.) Complete Transactions refer to those which conclude with an end record. All transactions includes those which had not ended at the end of the trace.

1996