Pub quiz Hospodský kvíz categorized their results into various topics.

Those topics are:

 1	aktualitova		news
2	filmova			movies
3	historicka		history
4	hrava			puzzles
5	kulturni		culture
6	obrazkova		pictures
7	poslechova		listening
8	prirodovedecka		natural science
9	spolecenska		society
10	sportovni		sport
11	vedecko-technicka	sci-tech
12	zemepisna		geography


The results of individual teams can be represented as a matrix where each row is one team and each column is a topic according to the list above.

Not all teams took part on all contests (and not all contests presented all topics), so the missing results are filled with zeros for the purposes of the statistics.

Figure 1 Mean and standard deviations per topic. Note that some topics were presented less frequently, so their appear as more difficult (I don’t have data for normalization).

Figure 2 Topic correlations. Interesting are weak correlations of puzzle-history, society-history and puzzle-sport. This may be affected by “cohabitation” of some topics more often than others.

Figure 3 Principal component analysis (columns are topics, rows are the principal components, weights of them are in in Figure 4).

355.0922202481993
16.36153546791306
11.97173965876528
9.799514147814509
9.152375537498369
7.212498668027225
6.729483237792453
5.730609693783117
4.466064954700245
4.328148525381178
3.693248457998539
2.565359833986063


Figure 4 PCA weights, we can see there is no much differentiation. Teams are mostly spread along one axis – overall skill of a team.