Overview of the Data Collection  (an outline of the methods and information sources)
The 500 Composers, Placed in Rank Order  (the 500 composers ranked statistically on the basis of the methods applied here, plus an alternate, subjectively assessed, "top 222" list)
The 111 Most Influential Composers  (three secondary compilations based on collected "influences" data for the 111 most influential composers)
"Musical Influences" Statistics  (statistics of composers' influence for all 500 composers, organized alphabetically by composer name)
Charts Depicting the Allocation of the 500 Composers Over History  (graphs showing how the 500 composers treated here are distributed over time)
An Alternate Approach to the "Has Influenced" Statistics  (a manipulation of the influences data that makes an adjustment compensating for historical remoteness)
Composers' Similarities  (planned for the future: an investigation into which composers are most similar to which other ones, based on the descriptive data presented here)

Overview of the Data Collection

Over the ten years that Version 1 of this service operated several individuals asked me to supply more description of its methodological underpinnings. The information provided at this site was statistically arrived at; i.e., decisions as to which composers and which of their works should be included were based on objective criteria, not subjective preferences. The particular 500 composers now included scored highest on a combination of eleven (unweighted) variables; these were: (1) length of composer entry in the Schwann Opus catalog (2) length of composer entry in the Grove's Dictionary of Music (3) length of composer entry in the (British) RED Classical Catalogue (4) for each composer's existing published sheet music, the number of libraries in the OCLC WorldCat database (covering the sum holdings of over 50,000 libraries in the U. S. and worldwide) holding his/r 20th-ranking work in his/r overall list (5) from the same source, the total number of sheet music publications for each composer over the past five years only (6) from the same source, the total number of Library of Congress subject headings referring to each composer’s name (name authority record) (7) same as #6, but taking the total over the past five years only (8) from the same source, the total number of recordings referring to each composer (9) same as #8, but taking the total over the past five years only (10) as for #4, the number of libraries holding his/r 25th-ranking recording (11) combining data from the Opus and RED catalogs, the mean length of record for all the works by each composer.

These data were reduced to rank values across some 700 initial composer names, and the geometric mean taken across the ranks, establishing an overall score. Some small adjustments were made to compensate for outlier values and national affinities, then the resulting top 538 scorers were re-subjected to the process to obtain a penultimate top 500.

For practical reasons I ultimately decided to allow ten composers (ranking in the range #501 to #515) who had been part of Version 1 to remain in the list, at the expense of what would have been ten newcomers to the list (with ranks in the range of #487 to #500). With the expansion from 444 to 500 composers in Version 2, this meant that thirteen composers from Version 1 were dropped from Version 2, with sixty-nine new names being added. The top-ranking new composer (i.e., who was not included in Version 1 at all) was Josef Rheinberger, at #226. Regarding the top 222, twenty composers were newly included (and twenty others dropped, of course), the highest new entry, Astor Piazzolla, being at #98.

Decisions as to which individual works of a given composer should be included were based, as in Version 1, primarily on relative numbers of currently available recordings (as determined from standard music recordings catalogs), relative numbers of entries and holdings in the OCLC WorldCat database (recordings and sheet music), and the composite opinions of about a dozen standard reference sources. The (very) approximate number of works to be included for a given composer was determined on the basis of the relative scores obtained as described in the first paragraph above.

Version 2 also contains a good deal more information on influences. In the original effort, more than a thousand biographical, analytical, and reference sources were examined (and supplemented by information retrieved from reviews of recordings and database searches). For Version 2 I was able to supplement the original data with new database searches, further investigations of reference works, and a good deal of internet work focusing on online-available dissertations, album liner notes, and concert notes and reviews. In all, opinions from at least five thousand sources have contributed to the “influences” compilations. Over the period 2015-2017 I was able to carry out a further investigation of sources on influences with the result that the influences noted are based on a total of more than 20,000 opinions.

One of my objectives in creating this structure was to come up with a set of variables that I felt could be used to identify a rank order of “current significance” (broadly stated). I am happy with the results, yet I would be remiss were I not to admit that I still feel there are some biases in the data. On the basis both of my own judgment and various ranking systems I have seen over the years, it appears to me that the present system slightly: (1) underestimates the place of composers with small outputs, (2) overestimates the place of opera composers, and (3) underestimates the place of recent composers. But what variable(s) might deal with this problem all at once!?

For this reason I have also supplied a "top 222" list based both on the results of Version 2, and my knowledge of other ranking efforts--call it a "gut appraisal" if you will, but at the very least it generates one more starting point for discussion!

The reader is invited to investigate the other compilations linked to from the top of the page, as they contain some additional ways of looking at the data that might stimulate further thought.

Return to Main