notes age heaping and Myer's index

notes age heaping and Myer's index - Eco 572: Research...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Eco 572: Research methods in Demography Digit Preference The datasets section has the age distribution of the Philippines in 1990 from ages 0 to 99, which appears as table 7.7 in Siegel and Swanson (2004), p. 103. (Thanks to Tom Pullum for providing an electronic version as well as a do file to analyze it.) . infile age pop using /// > http://data.princeton.edu/eco572/datasets/phpop1990.dat (100 observations read) . gen pm = pop/1000000 . line pm age, ytitle("Population (millions)") /// > title(Philippines 1990) . graph export phpop1990.png, replace (file phpop1990.png written in PNG format) The spikes are likely to represent some form of age misreporting. To investigate a preference for terminal digits such as 0 or 5 we could tabulate the last digit of age . gen lastdigit = mod(age,10) http://data.princeton.edu/eco572/digitpref.html (1 of 6) [2/12/2008 10:21:15 AM] Eco 572: Research methods in Demography . tab lastdigit [fw=pop] lastdigit | Freq. Percent Cum.------------+----------------------------------- 0 | 7,611,712 12.57 12.57 1 | 6,374,857 10.53 23.10 2 | 6,424,913 10.61 33.71 3 | 6,042,858 9.98 43.69 4 | 6,003,475 9.91 53.60 5 | 6,140,517 10.14 63.74 6 | 5,546,993 9.16 72.90 7 | 5,664,798 9.35 82.25 8 | 5,344,036 8.82 91.08 9 | 5,401,935 8.92 100.00------------+----------------------------------- Total | 60,556,094 100.00 We see strong preference for 0, and some for 1, 2 and 5. Note, however, that because of mortality we should expect more people at age 0 than 5 or 9, and more at 10 than 15 or 19, and so forth, with more at 80 than 85 or 89. Myers' adjustment is to tabulate the last digit several times starting at different values. In our example we could tabulate it only for ages 0 to 89, and then 1 to 90, until we do 9 to 98, and sum the tabulations, obtaining a "blended" population which helps control for a trend....
View Full Document

This note was uploaded on 02/12/2008 for the course ECON 572 taught by Professor Rodriguez during the Spring '06 term at Princeton.

Page1 / 6

notes age heaping and Myer's index - Eco 572: Research...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online