IOI2001 Competitor Questionnaire
Notes
There were 272 competitors at IOI2001.
Each delegation received 4 competitor questionnaires in their mailbox
togehter with the results for the Second Competition Day.
By the end of IOI2001, we had received 119 completed forms (almost 45%).
The survey was anonymous.
When interpreting individual responses, it should be noted that
- competitors may have misundertood questions;
- competitors may have made mistakes in completing the questionnaire;
- competitors have subjective interpretations of answers involving
a graded scale;
- competitors may have been biased in participating in the survey
(e.g. because of their competition performance, their perceived opportunity
of participating again next year, delegation pressure, etc.)
- competitors may have failed to recall all relevant information.
In particular, the last point should be kept in mind when reading
the responses to the open questions.
No doubt, competitors would have added (even) more detail,
if they had seen the responses of others to help them recall the events.
As always, it is not clear how representative the set of respondents
is for the whole ensemble.
However, because of the large number of respondents,
the overall picture painted by the whole set seems fairly accurate.
Brief summary of results
- 1, 2: More than 1/3 of the respondents used
Linux on both competition days;
less than 2/3 used Windows;
very few used both or switched between competition days.
- 3...5: Many useful suggestions were offered
for improving the Linux and Windows environments,
as well as the grading system and web services.
- 6, 7: About 1/2 of the respondents used Pascal on both
competition days;
about 1/5 used C,
and about 1/4 used C++;
only one competitor indicated to have used more than one programming language;
none switched between days.
- 8...14: The FreePascal IDE was used by
more than 1/3 of the competitors;
Turbo Pascal IDE by more than 1/4 (some used both FP and TP IDE);
RHIDE by almost 1/3;
Turbo C++ IDE was hardly used;
Vi by about 1/10;
Emacs by about 1/5;
as other editor Notepad was mentioned most often.
- 15: Just over 3/4 of the respondents indicated that they had used some
form of debugging aid, mostly the integrated debugger of an IDE.
- 16: Many used some form of calculator;
also mentioned are scripting, including one case of perl.
- 17...21: Web services were generaly easy to understand,
except that Test was considered less clear.
- 22...26: The perceived response time of web services
averages at medium,
with Submit/Test more at the slow end.
- 27...31: The majority seemed to like the web services,
but there are also clear signs of dislike.
- 32...37: On the whole, all tasks were considered relatively
easy to understand, with MOBILES as easiest.
- 38...43: More than 2/3 of the respondents rate TWOFIVE as
very difficult for finding an algorithm;
MOBILES, SCORE, and DOUBLE end in the middle;
IOIWARI is considered slightly easier, and DEPOT slightly harder.
- 44...49: Writing a program is considered hard for TWOFIVE
(though not as pronounced as for finding an algorithm);
MOBILES, IOIWARI, SCORE, and DEPOT end up around the middle;
DOUBLE is clearly considered easy in this respect.
- 50: DOUBLE is mentioned as task liked best
by almost 1/4 of the respondents,
followed by SCORE (almost 1/5), IOIWARI (1/6), and MOBILES (1/8);
TWOFIVE and DEPOT each get (almost) 1/10.
Consequently, every task had its fans.
- 51: TWOFIVE is mentioned as task liked least
by 45% of the respondents;
MOBILES, IOIWARI, SCORE, DOUBLE, and DEPOT get around 1/10.
- 52: Less than 1/4 of the respondents is subscribed to the
IOI mailing list.
- 53: More than 1/5 of the respondents provide further comments.
We have not looked at possible correlations between answers
(e.g. program language versus platform), but this can be easily
obtained from the raw data.
Other correlations (e.g. final score versus language) cannot be uncovered
for lack of information (due to the anonymous nature of the survey).
IOI Secretariat /
ioi-secretariat@win.tue.nl
Internet service provided by
TUE = Eindhoven University of Technology