The figures don't add up

life expectancy, getting worse or improving???

I was reading an article about Okonjo-Iweala (Nigeria's finance minister) on the BBC news site but got sidetracked by the stats listed about Nigeria. Life expectancy seems to have dropped from 51 in 2000 to 47 in 2005.

Now I love facts and figures (i'm that sad), and i was quite interested in doing a comparison of the life expectancy of the average Nigerian over the years.
A quick seach on the net reveals the following figures

Year	1955	1960	1970	1980	1990	2000	2005 (prognosis)	2015 (prognosis)	2050 (prognosis)
Life expectancy at birth (Number of years)	36.5	38.2	42.0	46.1	50.2	52.5	51.5	52.3	63.9

The calculations were done in 2000
source: http://globalis.gvu.unu.edu/indicator_detail.cfm?IndicatorID=18&Country=NG

I'm a bit curious about how the figures are calculated, what criteria is used in measuring life expectancy in countries like Nigeria where records are not kept. (that's why we have to shut down the whole country for a week to do a census). What population sample are these figures based on?

According to wikipedia
'Life expectancy is the average number of years remaining for a living being (or the average for a class of living beings) of a given age to live. Life expectancy is also called average life span or mean life span, in distinction to maximum life span. Life expectancy is also defined as the age at which 1/2 of age cohorts have died.'

But here's the interesting part
'Life expectancy is heavily dependent on the criteria used to select the group. In countries with high infant mortality rates, the life expectancy at birth is highly sensitive to the rate of death in the first few years of life. In these cases, another measure such as life expectancy at age 10 can be used to exclude the effects of infant mortality to reveal the effects of other causes of death. Typically, life expectancy at birth is specified. To calculate it, it is assumed that current mortality levels remain constant throughout the lives of the hypothetical newborns.'

What i find surprising is that life expectancy was much worse in the 50s,60s and 70s than it is now.
This is strange 'cause if you were born in the 60s or 70s you were likely to know, not only your grandparents but your great grandparents (and in some cases your great great grandparents, Ok so people had kids earlier).
If you also factor in the fact that many people did not know their date of birth or the high infant mortality rates during that period, it really makes you wonder where they got the figures from.

methinks just like the population figures bandied about, this figure is just a guestimate.