Mallard populations experienced a statewide increase of 
20% compared with 1978 and were 12% above the pre- 
vious 6-year mean; blue-winged teal increased 29% over 
1978. They are still 21% below the previous 6-year mean. 
Mallards and blue-winged teal increased in the South- 
east-Central region and the Northern High region. Both 
species experienced a significant decrease in the Northern 
Low region. 
Waterfowl Harvest Surveys 
Data supplied by Samuel M. Carney, Michael F. Sorensen, and 
Elwood M. Martin 
U.S. Fish and Wildlife Service 
This report provides estimates of waterfowl hunting ac- 
tivity and harvests during the 1978 season and compares 
them with estimates for the 1977 season. Estimates for 
both years were derived from information obtained from 
three sources: (1) the Postal Service’s report of Migratory 
Bird Hunting and Conservation Stamp (duck stamp) sales, 
(2) the U.S. Fish and Wildlife Service's Questionnaire Sur- 
vey of Waterfowl Hunters, and (3) the Service's Water- 
fowl Parts Collection Survey. 
In previous waterfowl status reports, harvest estimates 
for a particular State related to hunting by individuals 
who had purchased duck stamps in that State. No allow- 
ance could be made for the fact that some of this harvest 
may not have occurred in the State, Improved computer- 
analysis procedures now permit us to assign harvest to the 
States actually hunted, This change did not affect the esti- 
mated size of the U.S. harvest, but it did increase or de- 
crease State estimates depending upon movements of 
stamp buyers across State lines (Table C-1). Because State 
estimates changed and species composition varied among 
States, flyway and U.S. species composition also changed 
slightly. 
Another change initiated with this report affects esti- 
mated species compositions of duck harvests within States. 
Previously, weights (ducks killed per wing received) were 
calculated by using State-harvest estimates. The new 
method uses stamp sales zones (subdivisions of States along 
county lines), and wings received from in-zone stamp 
buyers are weighted by using zone-harvest estimates. 
Procedures 
Survey Sampling 
The Waterfowl Hunter Questionnaire Survey is de- 
signed to obtain answers from representative samples of 
the Nation’s waterfowl hunters that can be expanded to 
provide estimates of State, flyway, and national totals. Be- 
cause no list of waterfowl hunters exists, we obtain sam- 
ples indirectly by randomly selecting more than 3,000 of 
15 
the 16,000 post offices that sell duck stamps. Historical 
sales information is used to select post offices within each 
stamp sales zone from among three strata (<100, 
100-999, or >999 duck stamps sold annually). We usually 
subsample post offices in large cities by branches or sta- 
tions. Within States, post offices are selected so that distri- 
bution of the samples among zones and strata will be simi- 
lar to that of duck stamp sales. Each post office selected is 
sent a supply of postage-paid postcards exceeding antici- 
pated duck stamp sales, Stamp buyers use these cards to 
record their name, address, and whether their purchase 
was for hunting or stamp collecting. The last allows us to 
remove estimated portions of philatelic sales from total 
duck-stamp sales. A diary portion of the cards may be de- 
tached and used to record hunting activity. The approxi- 
mately 100,000 potential hunters who return a card are 
sent a questionnaire at the close of the hunting season, We 
send non-respondents one follow-up questionnaire. About 
70,000 contacts return a questionnaire, 
The Waterfowl Parts Collection Survey is designed to 
obtain from waterfowl hunters a representative sample of 
the waterfowl they shoot from which information can be 
expanded to State, flyway, and national estimates. An- 
nually we contact about 30,000 hunters selected from suc- 
cessful respondents to the previous year’s Questionnaire 
Survey and Parts Collection Survey; usually about 80% 
are Questionnaire Survey respondents. Within States, the 
distribution of these contacts among stamp sales zones is 
proportional to the previous year’s stamp sales. Before the 
season starts, we send these contacts a supply of postage- 
paid envelopes with instructions asking them to send one 
wing from each duck or coot and the tail feathers from 
each goose they shoot throughout the season and to record 
the place and time of kill. We include postcards for the 
hunters to order additional envelopes. About 80,000 parts 
are received and examined to determine species, sex, and 
age (immature or adult) for ducks. Only species and age 
are determined for geese and only species for coots. 
Survey Estimates 
In determining zone-level means, all questionnaires re- 
turned by in-zone stamp buyers, including originals, fol- 
low-ups, and those from different stamp sales volume 
strata, are considered equally. Zone-level means are calcu- 
lated for retrieved and unretrieved ducks, sea ducks, teal 
season ducks, geese, and coots as well as for total days 
hunted. Teal season means for individual States are ad- 
justed downward 9 to 64% based on information from 
past comparisons of questionnaires sent to all potential 
hunters; questionnaires are sent only to potential teal sea- 
son hunters. Each zone mean (including the reduced teal 
mean) is then multiplied by the number of duck stamps 
sold in the zone and this, in turn, is multiplied by a State- 
level estimate of the proportion sold for hunting. Products 
(zone means x potential hunters) are rough estimates of 
totals achieved by all potential hunters who bought duck 
