MTB > read 'scitest.dat' c1-c5
     120 ROWS READ

  ROW    C1    C2   C3   C4   C5

    1    35     1    1    2    1
    2    34     1    1    1    1
    3    38     1    2    2    1
    4    38     1    2    1    1
   .  .  .

 MTB > name c1 'score' c2 'student' c3 'task' c4 'rater' c5 'school'
 MTB > anova score=student|task;
 SUBC> random student task;
 SUBC> ems.

 Factor     Type Levels Values
 student  random     30     1     2     3     4     5     6     7     8     9
                           10    11    12    13    14    15    16    17    18
                           19    20    21    22    23    24    25    26    27
                           28    29    30
 task     random      2     1     2

 Analysis of Variance for score

 Source          DF         SS         MS       F      P
 student         29   16081.37     554.53    4.18  0.000
 task             1     136.53     136.53    1.03  0.319
 student*task    29    3843.47     132.53   30.24  0.000
 Error           60     263.00       4.38
 Total          119   20324.37


 Source          Variance Error Expected Mean Square
                component  term (using unrestricted model)
  1 student       105.499   3   (4) + 2(3) + 4(1)
  2 task            0.067   3   (4) + 2(3) + 60(2)
  3 student*task   64.075   4   (4) + 2(3)
  4 Error           4.383       (4)


Here's another way to use these data, but probably less interesting.

 MTB > # analysis using task & school as random factors
 MTB > table c3 c5;
 SUBC> means c1;
 SUBC> stdev c1.




  ROWS: task     COLUMNS: school

            1        2        3      ALL

   1   35.500   33.500   36.550   35.183
       11.105   14.096   14.218   13.055

   2   33.850   30.800   34.500   33.050
        9.349   14.388   15.188   13.105

  ALL  34.675   32.150   35.525   34.117
       10.166   14.125   14.558   13.069

   CELL CONTENTS --
             score:MEAN
                   STD DEV

 MTB > anova score=task|school;
 SUBC> random task school;
 SUBC> ems.

 Factor     Type Levels Values
 task     random      2     1     2
 school   random      3     1     2     3

 Analysis of Variance for score

 Source         DF         SS         MS       F      P
 task            1      136.5      136.5   48.62  0.020
 school          2      246.5      123.3   43.89  0.022
 task*school     2        5.6        2.8    0.02  0.984
 Error         114    19935.7      174.9
 Total         119    20324.4

 Source         Variance Error Expected Mean Square
               component  term (using unrestricted model)
  1 task           2.229   3   (4) + 20(3) + 60(1)
  2 school         3.011   3   (4) + 20(3) + 40(2)
  3 task*school   -8.603   4   (4) + 20(3)
  4 Error        174.875       (4)