permission. All tabulated data are subjeeted to an 
extensive diselosure review prior to publieation. Any 
tabulated item that identifies data reported by a 
respondent or allows a respondent's data to be 
aecurately estimated or derived, was suppressed and 
coded with a 'D'. However, the number of farms 
reporting an item is not considered confidential 
information and is provided even though other 
information is withheld. 
DATA COMPARABILITY 
Data definitions are comparable between the 2013 
and 2005 aquaculture censuses, with the exception 
that the 2005 Census of Aquaculture did not include 
the algae categories of microalgae and sea 
vegetables. For 2013 algae was included in the data 
for total sales and miscellaneous sales. Specific data 
changes from 2005 are listed in Appendix B. Dollar 
figures are expressed in current dollars and have not 
been adjusted for inflation or deflation. 
The census of aquaculture data are not directly 
comparable to the census of agriculture, due to 
different priorities and data definitions. A census of 
agriculture priority is the value of production of all 
agriculture (including aquaculture) at the county 
level. A census of aquaculture priority is a more 
specific look at U.S. and State-level aquaculture 
sales and aquaculture distributed for conservation. 
In the 2012 Census of Agriculture all agriculture 
production moved off the farm had a value of sales 
reported or assigned. Aquaculture which was moved 
for distribution, conservation, recreation, etc. was 
assigned a value. In the 2013 Census of Aquaculture, 
farms with aquaculture which was produced and sold 
are included in tables 1 through 21, and 24. 
Aquaculture which was not sold, but distributed for 
conservation is included in tables 22 and 23 only. 
The number of farms for each category is also 
affected between the two censuses. For county level 
data the census of agriculture attempts to get a 
response for each location. The census of 
aquaculture allowed one respondent to report for 
multiple locations, which reduces farm counts. 
Another difference with the census of agriculture is 
the minimum level of production. The census of 
agriculture has a minimum of $1,000 of production 
2012 Census of Agriculture 
USDA, National Agricultural Statistics Service 
or potential production of all agriculture items. For 
example, a farm with $200 of crayfish and $900 of 
rice is included. The census of aquaculture minimum 
is $1,000 worth of aquaculture production either sold 
or distributed for conservation, which could reduce 
the number of farms. 
The last difference is that the census of agriculture 
food fish category excludes catfish and trout. The 
census of aquaculture includes catfish and trout in 
the food fish totals. 
MEASURES OF CENSUS QUALITY 
There are two main types of estimation error that 
affect all estimates obtained from almost any survey. 
These errors make it unlikely that estimates obtained 
from the 2013 Census of Aquaculture will exactly 
match the true value in the population for a given 
farm characteristic. 
The first type of error, referred to as non-observation 
error, occurs in any estimate generated from a survey 
in which nonresponse occurs or data are not 
potentially obtainable from every unit in the target 
population. Statistical weighting as described in the 
Estimation section is used to reduce the effects of 
this type of error. 
The second type of error is called non-sampling 
error. There are many sources of nonsampling error. 
Respondent reporting errors, data collection errors, 
data keying errors, data editing errors are all 
examples of errors of this type. Quality controlled 
data processing is used to keep the effect of non- 
sampling errors to a minimum. 
Census Response Rate 
The response rate is one indicator of the quality of a 
data collection. It is generally assumed that if a 
response rate is close to a full participation level of 
100 percent, the potential for nonresponse bias is 
small. Because the aquaculture mail list contained 
both farm and non-farm records, the response rate is 
an indicator of replying to the data collection effort, 
but does not reflect whether those responding 
records qualified for data summarization. The 
response rate for the 2013 Census of Aquaculture is 
90.2 percent. 
APPENDIX A A- 3 
