Generally, this is not a wise thing to do. Research trawls are usually smaller and have finer mesh cod ends than commercial gear. This means that the selectivity of research trawls will be very different from commercially used trawls and the resulting catch composition would, therefore, be very different. If you could find studies where a research vessels used a commercial trawl that would be more comparable.

Most of the studies in bottom trawl fisheries conducted here in Brazil have the approach Mr. Tingley referred above, using a commercial trawl with sampling design made by researchers. 

http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1516-89132005000500014&lang=pt

http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1679-87592012000100003&lang=pt

In case of finding new fishing areas in the mouth of Gulf of Kutch in Western India,( areas west of Mandvi & Port Okha) 3 Research & survey vessels had carried out surveys in different sections of the waters using 4 seam trawl nets,buldge belly trawl nets,& high opening bottom trawls in 1971 to 73. Some of the fishing grounds had & is even now giving substantial rate of CPUE. The findings of the catch data over the fishing season were disseminated to local fishermen & this had opened new areas of fishing. The findings of both the Research & survey vessels & that landed by commercial trawlers were almost the same,except for some differences. Similarly Fishery Survey of India also caries out survey of different areas using different gears & seasons. As most of the surveys are beyond the neritic zone,& if the findings are not attractive,naturally the commercial fishermen are not interested. Nevertheless you can contact the organization for their findings.

Central Marine Fisheries Research Institute, also carries out research & survey of certain areas & they may also have information related to your question.

It may not be logical to use research vessel data for estimating commercial catch as purpose and facilities available at research vessels are different than any commercial fishing vessel.

Since 1985 my institute conducted in more than 20 trawl surveys statistically planned in the space and in the time. In each of these campaigns have been carried out on average about 70 hauls with the same vessel and with the same net (commercial trawl net used of the Mazara del Vallo trawl fishery in 1985). The duration of each haul was one hour independently of the depth (from - 20 m to - 800 m). Generally, the duration of the commercial haul varies from 1.5 hours up to 4 hours in proportion to the depth (duration of haul increases with depth). When a trawl is fishing the animals try to escape by running in front of the net. When the fish get tired, slow down the run and are captured by the net. So the fish faster and stronger was able to escape capture. Since the duration of the experimental hauls is always one hour while the duration of commercial haul is at least an hour and a half the catches are not even comparable, always using the same vessel and the same fishing net.

Risky. From my experience commercial fisherman's are much better as fisherman's. The best practice is observers onboard. Reasons for bias would be essential  diferences of:

vessels; gears; fishing process, seasons, fishing target .....

As pointed out, duration of tow can affect catch composition even when the same gear is used. Survey trawls are usually very different from commercial gear. Just mesh size alone will affect catch composition both in terms of species mix and sizes captured. Thus, catch composition in survey and commercial gears used in the same area are not comparable. As noted, the only reliable source of information on commercial catch compostion is from fishery observers. I have used observer data to determine of species of interest such as ones at risk of extinction, information not present in commercial landings or log data

Contact someone at the U.S. NOAA's Northeast Fishery Science Center.  I believe they have done a great of "calibration" tows in order to quantify those differences in gear/vessel type.  There are certainly going to be differences in catchability between vessels/gear, but with a great deal of effort you can determine those differences and account for them in your results.

Saluti Leonardo. Exactly. We did underwater photography of fish behaviour during trawling and found exactly the same, that's that duration of a tow has significant effect on the size and species composition of the catch. Short tows would result in (1) underestimating the share of the larger and faster fish in the population, and (2) hence, would also underestimate the general fish abundance in comparison with commercial trawling. For Ms Maclean's study, it's important to emphasize the conditions enabling such comparison: (1) the fishing must be done at the same time and the same area as commercial fishing; (2) The trawlgear, incl. warp lengts in water, must be identical; (3) The towing speed must be the same; (4) as above: the tow duration must be the same; (5) ideally, tows should be performed simultaneously and in parallel with a commercial trawler, as in comparative fishing.   MB-Y

In addition to NMFS work in the northeast U.S. there have also been many studies conducted in the southeast U.S. by NMFS, especially for red snapper (Lutjanus campechianus) in the Gulf of Mexico.  In general, research vessel trawl catches are generally smaller than commercial vessel trawl catches using the same gear, especially if the research sampling is based on some sort of random sampling (usually not the case for commercial trawlers).  Studies have also been conducted comparing commercial fishing gear fished by biologists to the same gear fished by fishers (e.g., trotlines in Texas). 

Other contributors have emphasized the differences between commercial and survey gears but the problem is not just what gear is used but when, where and how it is used -- tow length being just one aspect of that. Commercial fishermen usually aim to maximize their catch of one or more target species. They are generally very skillful at selecting just where and when to deploy their gear, while they are able to make many detailed changes in how it is deployed. The end result is that (in general) they catch more of the target species (and sizes) than research vessels do. Even if they do not also reduce the catches of other species, more of the targets means a lower proportion of the rest. In practice, different species and sizes favour different micro-habitats and so fishing where you get more of the target species/size will tend to reduce catches of everything else. In short: research-vessel catch composition will only very rarely tell you anything useful about commercial catch composition. It would need case-specific evidence to establish a link between the two and you would need data on commercial catch composition for that study, which would negate much of the value in establishing the link.

Thanks all for your helpful answers - the point about tow duration is one I hadn't appreciated before. The Scottish trawl surveys are only 30 mins duration, which would certainly make the catch size much smaller. I will look more at NOAA's callibration studies, Charles, as it is proportions rather than numbers I'm most interested in.