DiscussWorldIssues - Socio-Economic Religion and Political Uncensored Debate - View Single Post

**UKkoXJvF** · 08-12-2009, 03:28 PM

Just found the actual study. The median, which they report for men as 6.8 (?), is a crude number which is uncorrected even for the amount that the age structure of their survey does not correspond to the age structure of the 20-59 cohort in the US population.

The medians for age subgroups in their samples are as follows:

20-29: 5.4
30-39: 7.6
40-49: 6.9
50-59: 6.9

They don't appear to specify, but I'm interested in knowing what type of median they are using. The sample median will always be either an integer or half-integer for data restricted to the integers.

08-12-2009, 03:28 PM	#4
UKkoXJvF Join Date Oct 2005 Posts 453 Senior Member	Just found the actual study. The median, which they report for men as 6.8 (?), is a crude number which is uncorrected even for the amount that the age structure of their survey does not correspond to the age structure of the 20-59 cohort in the US population. The medians for age subgroups in their samples are as follows: 20-29: 5.4 30-39: 7.6 40-49: 6.9 50-59: 6.9 They don't appear to specify, but I'm interested in knowing what type of median they are using. The sample median will always be either an integer or half-integer for data restricted to the integers.
	Quote