>If the breastfeeding initiation rates among middle-class
>well-educated white women go from 60% to 70%, this is only a 17% increase.
>But if the breastfeeding initiation rates among WIC clients go from 20% to
>30%, this is a whopping 50% increase in rates.  In reality, both groups have
>increased exactly the same.


>>I'm not an expert with statistics but I just don't understand the above
conclusions.

Happy to oblige.  You are interested in comparing the rate of something (in
this case, starting breastfeeding) at Time 0 (start) and again at Time 1
(after a certain amount of time has elapsed, or after an intervention, say).
You look at the difference in rates between Time 0 and Time 1 and calculate
what percentage increase has occurred.

For example, out of 100 women:
Time 0  You have 60 who are breastfeeding.
Time 1  You have 70 who are breastfeeding.

To calculate the rate of increase, you divide 10 by 60 (difference between
Time 1 and Time 0 divided by number at Time 0).  This number is 17.  You
have had a 17% increase.  An additional 10 people breastfeeding represents a
17% increase over the original 60.

Second example, out of 100 women:
Time 0  You have 20 who are breastfeeding.
Time 1  You have 30 who are breastfeeding.

Calculate the rate of increase the same way, divide 10 by 20.  This number
is 50.  You have had a 50% increase.  Half-again as many people are
breastfeeding.

Third example, out of 100 women:
Time 0  You have 10 who are breastfeeding.
Time 1  You have 20 who are breastfeeding.

This is a doubling of the number of people breastfeeding, or a whopping 100%
increase.

Fourth example, out of 100 women:
Time 0  You have 1 who is breastfeeding.
Time 1  You have 11 who are breastfeeding.

This is an increase of 1000%.

In each case, you have exactly 10 more women breastfeeding at Time 1 than
were at Time 0.  But if you start with a high number already breastfeeding
(60/100 or 60%), then this is only a small improvement of 17%.  If you start
with a low number breastfeeding (1/100 or 1%) then this is a HUGE
improvement of 1000%.

Percentage increases depend on the size of the original percent
breastfeeding.  If this is still unclear, let me know.

Kathy Dettwyler

             ***********************************************
The LACTNET mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software together with L-Soft's LSMTP(TM)
mailer for lightning fast mail delivery. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html