CS3331 Grade Listing

Important Notes

A blank entry in the grade is a missing grade, which means you did not submit a program or did not take an exam. A zero (i.e., 0) entry indicates that you received a zero for that program or exam.

You will not receive a passing grade in the course unless you receive a passing grade on the exams alone (i.e., 60%). Likewise, you will not receive a passing grade in the course unless you receive a passing grade on the programming assignments alone (i.e., 60%).

If you receive a * next to Prog/Avg, Exam/Avg, or Avg, you should work harder for the remaining programming assignments and exams, because you are failing.

Always check if the recorded grades are identical to those marked on the graded programs and papers.

CS3331 Grade Listing
March 10, 2019

0: 0 point     space: no submission
*: may fail this course

          50 100 150        100  100           250
       #  P1  P2 Sum  AVG   EX1  Sum  AVG    TOTAL   AVG
=========================================================
       2   0   0   0    0*   58   58   58       58    23*
       4  49  75 124   83    26   26   26*     150    60
       5  50  98 148   99    42   42   42*     190    76
       6  48  84 132   88    28   28   28*     160    64
       7  48  30  78   52    64   64   64      142    57
       8  27      27   18*   60   60   60       87    35*
       9  13      13    9*   34   34   34*      47    19*
      10  46  95 141   94    35   35   35*     176    70
      11  41  74 115   77    51   51   51      166    66
      12  48  99 147   98    51   51   51      198    79
      13  39  33  72   48*   79   79   79      151    60
      14  50      50   33*   23   23   23*      73    29*
      15  50  65 115   77    51   51   51      166    66
      16  50  65 115   77    61   61   61      176    70
      17  50      50   33*   65   65   65      115    46
      18  49  74 123   82    61   61   61      184    74
      19  50  63 113   75   100  100  100      213    85
      20  50  78 128   85    58   58   58      186    74
      21  48  65 113   75    48   48   48*     161    64
      22  40  75 115   77    54   54   54      169    68
      23           0    0*   25   25   25*      25    10*
      24  50  94 144   96    82   82   82      226    90
      25  49  23  72   48*   37   37   37*     109    44*
      26  37   0  37   25*   32   32   32*      69    28*
      27  46 100 146   97    56   56   56      202    81
      28  49   5  54   36*   41   41   41*      95    38*
      29  49  95 144   96    70   70   70      214    86
      30  50   0  50   33*   21   21   21*      71    28*
      31  50  59 109   73    60   60   60      169    68
      32  47  64 111   74    96   96   96      207    83
      33  49  50  99   66    83   83   83      182    73
      34  47   0  47   31*   38   38   38*      85    34*
      35  50  60 110   73    50   50   50      160    64
      36  49  84 133   89    75   75   75      208    83
      37  48  64 112   75    86   86   86      198    79
      38  38  73 111   74    70   70   70      181    72
      39  50  98 148   99    78   78   78      226    90
      40  50      50   33*         0    0*      50    20*
      41  49  70 119   79    66   66   66      185    74
      42  49  97 146   97    88   88   88      234    94
      43  46   0  46   31*   21   21   21*      67    27*
      44  50 100 150  100    44   44   44*     194    78
      45  49   0  49   33*   73   73   73      122    49*
      46  49  64 113   75    27   27   27*     140    56
      47  38      38   25*   70   70   70      108    43*
      48  50  65 115   77    74   74   74      189    76
      49  48  55 103   69    61   61   61      164    66
      50  48  50  98   65    90   90   90      188    75
      51  20   0  20   13*   34   34   34*      54    22*
      53  20   0  20   13*   37   37   37*      57    23*
      54  48  75 123   82    61   61   61      184    74
      55      20  20   13*   48   48   48*      68    27*
      56  48      48   32*   60   60   60      108    43*
      57   0  74  74   49*   70   70   70      144    58
      58  50  65 115   77    50   50   50      165    66
=========================================================
    Max   50 100 150  100   100  100  100      234    94
    Min    0   0   0    0    21    0    0       25    10
    Avg   44  57  91   61    56   55   55      146    58
 Median   49  65 111   74    58   58   58      164    66
  StDev   12  34  45   30    20   22   22       56    22

How to Interpret the Exam I Scores

The following is a table showing the class performance of Exam I.

Here is a graphical summary.

Here is a way of interpreting the above results.

  1. The p-value of the Anderson-Darling Normality Test of 0.027 indicates that we reject the null hypothesis that the score distribution is normally distributed. Note that the mean and standard deviation are 55.98 and 20.42, respectively.
  2. The graph in the middle is a boxplot of the scores of Exam I. Note that the boxplot uses the same scale as the that of the distribution plot. The mediann 58, the vertical line segment that divides the horizontal box, indicates that the score in the middle is 58. That is, there are about 50% scores higher than 58 and also about 50% scores lower than 58. The left and right ends of the horizonal box shows the first and third quartiles Q1 and Q3, respectively. This means there are about 25% of scores lower than 37.75 (i.e., Q1 = 37.75) and about 25% higher than 70 (i.e., Q3 = 70). The mean (i.e., 55.98) is lower than the median 58. Note that the median is usually referred to as the second quartile Q2. That the mean 55.98 is lower than the median 58 suggests that the upper half has a lower weight than the higher half.
  3. From the location of the median in the boxplot, which is not at the middle of the box and is slightly to the right of the middle point, this suggests that there are more lower scores (i.e., below the median) than the higher ones.
  4. The IQR, or Interquartile range, is the range between the first and third quartiles. More precisely, IQR = Q3 - Q1. In our case, because the first and third quartiles are 37.75 and 70, respectively, IQR = Q3 - Q1 = 70 - 37.75 = 32.25. The IQR includes 50% of the scores! A small IQR means a more concentrated distribution of the scores around the median.
  5. The two line segments at both ends of the horizontal box are whiskers. The left whisker represnts the lower 25% of the scores (i.e., scores from Q1 down to the lowest non-outlier scores). The right whisker represnts the upper 25% of the scores (i.e., scores from Q3 down to the highest non-outlier scores). In our case, the left whisker goes from 21, the lowest non-outlier, to Q1 = 37.75, and the right whisker goes from Q3 = 70 to the highest non-outlier score 100.
  6. What is an outlier? In statistics, an outlier is an observation that is distant from other observations. An outlier may be due to measuring error, experiment error, corrupted dataset, or even a correct but unusualy and extreme point. Outliers in a boxplot are usually shown as small dics or asterisks. We have three outliers in our score dataset, which are shown as three asterisks. These are the three highest scores.
  7. How to identify outliers? There are several commonly used methods for identifying outliers. The graph above used a simple technique based on IQR. Here, any data point is an outlier if it is more than 1.5*IQR above the third quartile Q3 or below the first quartile Q1. More precisely, a score is less than Q1 - 1.5* IQR or greater than Q3 + 1.5*IQR is an outlier. A score higher than Q3 + 1.5*IQR is no doubt an outstandingly good one. However, a score lower than Q3 - 1.5*IQR definitely indicates a very unsatisfactory performance. In our Exam I score dataset, we have L = Q1 - 1.5*IQR = 37.75 - 1.5*32.25 = -10.625 and R = = Q3 + 1.5*IQR = 70 + 1.5*32.25 = 118.375. Hence, all scores in the range of [L, R] = [-10.625, 118.375] are not outliers, and the only outliers are the top three scores.
  8. A student whose score is below Q1 has a higher risk of failing this course. Therefore, do your best so that your performance is above Q1.
  9. The last two scales are 95% confidence intervals. The first one has the range of the true but unknown mean of this exam, which is between slightly higher than 50 and slightly lower than 62. The second one has the range of the true but unknown median of this exam, which is between 50 and about 61.
  10. Note that because Exam II is more difficult than Exam I, scores of Exam II are always lower than those of Exam I. Most students will improve in the final.

How to make a case regarding your graded programs and exams?

  • We anticipate that grading related problems will surface throughout this semester. If you disagree with what the grader did, you should do the following:
    1. Explain why you are correct. Show me a convincing argument. Something like "I did this and this by accident" and "my program ran fine on my machine and/or on one of the lab machines" are not acceptable excuses. You must show the grader missed something. Note that the grader only grade what you have in the program and its output.
    2. Mark those places you may disagree with the grader and print a detailed note explaining why you are correct.
    3. Staple your graded program and your printed notes available in my mailbox in the Department office.
    4. I will make a decision regarding if a re-grade is necessary.
  • You can make a case within one week from the day the graded programs are available for pickup. After this seven-day period, I will assume you accept the recorded grade.
  • You should follow the same procedure to request an exam regrade.