So I was able to find a source for timing around the 1926 date:
The Chicago Booth Center for Research in Security Prices
states that their initial correction for cash dividends came from:
- 1926-1936: Moody’s Quarterly Dividend Record
Since the only known complete file of this last publication was in
Moody’s New York offices, the data from the earlier period were recorded
in the Moody’s offices by trainees working for the research division of
Merrill Lynch, Pierce, Fenner & Smith Inc.
See: http://www.crsp.com/products/documentation/data-sources
Curiously this last part appears to be a quote from 1964 article titled "Rates of Return On Investments In Common Stocks" by L. Fisher and J.H. Lorie
1964 article from the University of Chicago Journal Of Business: http://www.crsp.org/50/images/rates%20of%20return%20paper.pdf
More of a side note:
While Robert Schiller states on his Online Data page at Yale that for his book Irrational Exuberance:
Monthly dividend and earnings data are computed from the S&P four-quarter
totals for the quarter since 1926, with linear interpolation to monthly figures.
Dividend and earnings data before 1926 are from Cowles and associates
(Common Stock Indexes, 2nd ed. [Bloomington, Ind.: Principia Press,
1939]), interpolated from annual data.
So it appears for academic purposes approximations from annual data are used for pre-1926 analysis and quarterly values for post-1926 analysis.