IOI2001 Delegation Questionnaire

Original text of questionnaire
Marked-up verion of questionnaire to show how answers are encoded
Processing program
Raw questions (input to processing program)
Raw data for responses to closed questions (input to processing program)
Statistics for responses (output of processing program)
Summary of responses to open questions

Notes

There were 74 delegations at IOI2001. Delegations that attended the IOI Feedback meeting received a questionnaire. By the end of IOI2001, we had received 32 completed forms (almost 45%). The survey was optionally anonymous. All results published here are anonymized.

Brief summary of results

1..6: The scale in the question about suitability of tasks was labeled incorrectly. Easy...Mid...Hard was intended as Not...Mid...Much. In most cases, it is not clear how the scale was actually interpreted. Therefore, the answers to this question cannot be interpreted reliably.
7..12: SCORE was considered least understandable,
followed by IOIWARI at medium, and DOUBLE and DEPOT just somewhat better;
TWOFIVE and especially MOBILES were deemed easy to understand.
13..18: Translatability for SCORE has a diffuse rating (1/5 hard);
MOBILES, IOIWARI, TWOFIVE, DOUBLE, and DEPOT are all rated more towards the easy end.
19: MOBILES is liked best by 3/10 of the respondents,
DEPOT by over 1/5, IOIWARI by about 1/6,
TWOFIVE and SCORE by 1/10;
DOUBLE by none. Compare to competitors' questionnaire.
20: TWOFIVE is liked least by 1/4 of the respondents,
SCORE by over 1/5; DOUBLE by 1/6; IOIWARI by 1/8;
MOBILES and DEPOT by just a few.
21: Well over 3/4 of the respondents liked the idea of output-only tasks.
22: More than 1/2 of the respondents had something to say about the Task Approval Procedure.
23: The FreePascal compiler was supported by the majority, with only a few doubters;
24: The FreePascal IDE has little support.
25: The Gcc/G++ compiler has no opponents.
26: The RHIDE is supported to some extent by a majority, but not as much as the FreePascal compiler, though much better than the FreePascal IDE.
27: Over 2/3 of the respondents favor a Unix-like environment for grading execution;
1/4 prefers some Windows variant.
28: In case of a single development environment for competitors, there are a few more supporters for Linux than for Windows.
29: The majority considered the grading system used at IOI2001 suitable for the competitors.
30: On the whole, the evaluation reports were rated as good.
31: After-grading support is clearly in need of improvement.
32: An overwhelming majority finds a Practice Competition necessary.
33: Also the grading info handouts have a lot of support.
34: Extra score statistics are favored by most.
35: Questionnaires have major support.
36: 1/2 of the respondents provide further comments.

IOI Secretariat / ioi-secretariat@win.tue.nl

Internet service provided by TUE = Eindhoven University of Technology