The estimation procedure combined information from the telephone and intercept 

 surveys. The completeness of the resulting data matrix was occassionally affected by the 

 presence of "missing cells" in which no information was obtained. A cell is defined as State x 

 mode X species x wave in the intercept survey, and State x mode x wave in the telephone 

 survey. The presence of missing cells can result in an underestimate of the total number of 

 fish, or an estimate of number of fish but no corresponding estimate of the weight of these 

 fish. Missing weights were estimated by length-weight equations or using a protocol to input 

 an average weight for the species from the closest adjacent cell, such as the adjacent mode in 

 the same area and State. 



Special care must be taken when comparing catch estimates for the 8 years (1979-86) of 

 MRFSS data because of differences in sampling coverage. The 1981-84 sampling differs from 

 1979-80 in that the January - February waves in 1981-84 were not sampled on the Atlantic 

 Coast north of Florida. The 1985-86 sampling differs from 1981-84 since the January - 

 February wave of 1985-86 was sampled in Georgia. The 1981 sampling differs from all other 

 years since the January - February wave of 1981 was not sampled in the Gulf and on the east 

 coast of Florida. The 1982-84 sampling differs from 1979-81 and 1985 in that Texas boat mode 

 data were not included in 1982-84. For 1986, party boats in the South Atlantic and Gulf 

 subregions were not sampled by the MRFSS but were covered by a program conducted by the 

 NMFS Beaufort Laboratory in North Carolina.J_/ In addition, Texas was not sampled by the 

 MRFSS in 1986 but was covered by the Texas Parks and Wildlife Department 2.'. 



Large increases or decreases of the estimated total catch of individual species groups 

 are often due to the inclusion in the sample of unusually large catches of a species by one or a 

 few fishermen. Calculation of an estimated catch for the species for an entire State then 

 results in an artificially high estimate. Analysis of data and removal of outlines has minimized 

 this occurrence. In addition, examination of the standard errors or coefficients of variation 

 'associated with the estimates will indicate if significant differences exist between the two 

 estimates. Trends in the catch estimates from 1979 through 1986 should also be considered 

 when examining the total catch estimates. 



1/ For further information contact: NMFS Beaufort Laboratory, Pivers Island, P.O. Box 570, 

 Beaufort, NC 28516. 



2/ For further information contact: Texas Parks and Wildlife Department, Coastal Fisheries 

 Branch, 4200 Smith School Rd., Austin, TX 78744. 



12 



