LOW TEMPERATURE SMOKED FISH FILLETS:
A POTENTIAL NEW PRODUCT FORM
FOR FLORIDA FISH
W. Steven Otwell, John A. Koburger,
and Robert L. Degner
Technical Paper No. 19
Department of Food Science and Human Nutrition
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, FL 32611
Technical papers are duplicated in limited quantities for
specialized audiences requiring rapid access to information and
may receive only limited editing. This paper was compiled by
the Florida Sea Grant College with support from NOAA Office of
Sea Grant, U.S. Department of Commerce, grant number 04-8-M01-76.
It was published by the Marine Advisory Program which functions
as a component of the Florida Cooperative Extension Service,
John T. Woeste, Dean, in conducting Cooperative Extension work in
Agriculture, Home Economics, and Marine Sciences, State of Florida,
U.S. Department of Agriculture, U.S. Department of Commerce, and
Boards of County Commissioners, cooperating. Printed and distributed
in furtherance of the Acts of Congress of May 8 and June 14, 1914.
The Florida Sea Grant College is an Equal Employment Opportunity-
Affirmative Action Employer authorized to provide research, educa-
tional information and other services only to individuals and
institutions that function without regard to race, color, sex, or
The objective of this work was to develop a new product form which
could expand the use of traditional and underutilized Florida fish. The
product is a skinless and boneless fish fillet which has been flavored
by a low temperature smoking (LTS) process. The LTS process requires
less heat energy for production and provides a greater product yield than
the traditional hot smoking process. The LTS fillets can be frozen
and are cooked prior to serving. Market surveys with over 400 respondents
have indicated a very positive consumer reaction. The LTS fillets are
a unique seafood form which provides consumers with a new seafood choice.
Key words: Smoked fish, underutilized fish, new product.
TABLE OF CONTENTS
LIST OF TABLES..................................................... iv
LIST OF FIGURES.................................................. v
PRODUCT DEVELOPMENT............................................... 4
Brining..................................... ............. .... 4
Smoking Time................................................ 6
Packaging and Storage................................. ...... 7
Cooking Method............................................... 8
RECOMMENDED PRODUCTION PROCESS................................... 10
Packaging and Storage...................................... 11
Cooking.................................... ................. 11
PRODUCTION WITH VARIOUS SPECIES.......... ....................... 12
STORAGE STUDIES ............................. ..................... 14
CONSUMER SURVEY.................................................. 18
LIST OF TABLES
1 Effect of brine concentration on panel preference for fried
LTS mullet fillets previously smoked for 1 hours............. 6
2 Effect of smoking time on panel preference for fried LTS
mullet fillets previously soaked in 4 percent brine.......... 7
3 Effect of cooking procedure on panel preference for LTS
mullet fillets.............................................. 9
j4 Sensory evaluations of LTS fillets prepared from various fish
species ..................................................... 13
15 Effect of frozen storage on panel preference for fried LTS
mullet fillets............................................. 15
6 Sensory evaluations of influence of frozen storage on LTS
fillets prepared from various fish species.................. 17
7 Consumer ratings of characteristics of the test product and
previously eaten smoked fish................................. 20
8 Consumer ratings of physical attributes of the smoked fish
9 Primary food shoppers' purchase intentions for the test
10 Primary food shoppers' preferred package sizes for frozen
fish fillets................................................. 22
11 Primary food shoppers' indicated substitution of the test
product for currently available frozen fish fillets.......... 23
12 Respondents' intentions to order the test product if avail-
able in restaurants by current frequency of fish fillet
orders ..................................................... 24
13 Respondents' ability to correctly identify species from which
test products were made...................................... 25
LIST OF FIGURES
1 Standard process for the production of LTS fish
Financial support for this product development work was provided
cooperatively by the Agricultural Experiment Station, Institute of
Food and Agricultural Sciences, University of Florida and the State
University System of the Florida Sea Grant Program. Financial support
for the consumer survey was provided by immediate response funds from
the State University System of the Florida Sea Grant Program.
The completion of this work would not have been possible without
the valuable assistance of others which deserve specific recognition.
Sam May, laboratory technologist, provided many hours of assist-
ance in preparing the products and conducting a series of taste panels.
Frank Lawlor, Florida Cooperative Extension Service and Sea Grant
Marine Advisory Agent, helped in obtaining fish for producing the
Ricacdo Alvarez, graduate student, provided general laboratory
assistance and helped conduct various taste panels.
Barbara Prichard, secretary, prepared all drafts and the final
Various members, too numerous to list, of the Department of Food
Science and Human Nutrition served as taste panelists.
The market survey firms, Carmen Jones Market Research, Inc. and
Tampa Bay Opinion Mart conducted the well organized consumer surveys.
Judy King, statistician, prepared the informative statistical
analysis of the consumer survey data.
The objective of this work was to develop a new product form to
expand the use of traditional and underutilized Florida fish.
The product is a skinless and boneless fish fillet which has been
flavored by a low temperature smoking (LTS) process. Prior to serving,
the finished product must be cooked. The combination of first low
temperature smoking to impart flavor, then cooking results in a final
product that differs from the traditional hot smoked fish.
Product development work has outlined a recommended product
process as determined with product ratings by preference panels. The
basic recommended procedure is a 4 percent brine soak for 30 minutes,
air drying, smoking for 1 hours at temperatures from 800F to 1200F,
then frozen storage. All various cooking methods, including baking,
broiling, and frying, were rated acceptable.
Acceptable LTS fillets were produced from mullet., grouper, snapper,
Spanish mackerel, and cod. Frozen storage of the whole fish or 3 months
frozen storage of the finished fillets was not detrimental to the qual-
ity and panel preference for the LTS fillets.
Consumer surveys by professional market survey firms in Tampa and
Jacksonville indicated a very positive consumer reaction to LTS fillets.
Ninety-one percent of the primary food shoppers interviewed said they
would buy LTS fillets if they were available. A high percentage of the
402 respondents indicated they would order LTS fillets if available
The LTS fillets are a unique seafood form which could provide
consumers a new seafood choice. There is no one best method for pro-
duction of LTS fish fillets. Industry production of LTS fillets will
require necessary process modification to assure volume production with
Low Temperature Smoked Fisti Fillets:
A Potential New Product Form For Florida Fish
W. Steven Otwell, John A. Koburger and Robert L. Degnerl
A new product form has been developed to expand the use of tradi-
tional and underutilized fish species harvested in Florida. This
product is a skinless and boneless fish fillet which has beer flavored
by a low temperature smoking process. The combination of first low
temperature smoking (LTS) to impart flavor, then cooking either by frying,
baking, or broiling results in a final product that differs from traditional
hot smoked fish.
The basic production process was derived from modifications of the
cold smoking techniques commonly used in Europe to produce kipper fillets
from herring. The recommended LTS process requires only 1 hours
smoking time at temperatures between 80F to 1200F (270 to 490C); whereas,
hot smoking processes usually require a smoking time between 4 to 12
hours at temperatures in excess of 1400F (600C). The smoked flavor and
color are similar to that associated with traditional hot smoked fish,
but the LTS process requires less heat energy during production and
minimizes product dehydration. The boneless and skinless features of the
product should be more appealing to the typical American consumer, and
the finished product can be cooked by all che various methods typically
used to prepare regular, uncured fish. Thus, this new product form
1W. Steven Otwell is assistant professor and John A. Koburger is a profes-
sor in the Department of Food Science and Human Nutrition, and Robert L.
Degner is an assistant professor in the Department of Food and Resource
Economics. Both Departments are part of the Institute of Food and Agri-
cultural Sciences, University of Florida, Gainesville.
combines the desirable features of both regular fish and traditional
hot smoked fish.
Although the recommended LTS temperatures are lower than the re-
quired smokin- temperature specified in the U. S. Food and Drug Admin-
istration's Good Manufacturing Practice Regulations (GMP's) for hot smoked
fish2, this product is essentially cold smoked and subsequent!handling
procedures, and frozen storage must comply with the BMP's for handling perish
able foods. Currently, the GMP's for smoked fish are being revised and
should include guidelines for cold smoked products. These guidelines
should provide considerations for lower smoke temperatures if proper
sanitation and storage are employed. Thus, LTS fillets provide product
quality and safety to assure consumer satisfaction.
The recommended LTS process can provide a final product yield from
the raw fillets in excess of 98 percent as compared to lower product
yield, 50 to 75 percent from raw fish products entering the smokehouse
for preparation by traditional hot smoking methods. The finished LTS
product can be packaged whole or in controlled portions for frozen
storage. Tests have demonstrated that LTS fillets can remain frozen for
over three months with no detrimental effects on flavor or texture.
Thus, this new product is well suited to meet the typical requirements
for producers, retail outlets, restaurants, and fast food firms, i.e. high
yield, portion control, and extended shelf-life.
The LTS fillet is a non-species specific product which permits use
of a variety of fish species to produce the same product form. The
original product development was conducted using mullet, Mugil cephalus,
2Smoked and Smoked-Flavored Fish, Code of Federal Regulations, Title 21,
Part 128A, subpart A, effective date February 12, 1970.
but subsequent tests have verified successful production of the LTS
fillets from grouper, snapper, cod, and Spanish mackerel. These results
indicate the LTS fish fillets can be made from lean or fatty fish, and
from popular or underutilized fish. The fish species used would be
based on the economic discretion of the producer. The producer must
consider fish availability, fillet yield, fillet size, and labor
This product is an excellent way to utilize 'spent' (roe removed)
mullet carcasses. During recent years, foreign seafood firms have
expanded their demand for mullet roe. More than 250 thousand pounds of
roe was purchased in 1978 form the southeastern region of the United
States (personal communications, 1979). This accounts for about one
million pounds of whole mullet. Florida is the principle mullet roe
exporting state. Market value for the roe alone provides an adequate
net revenue and high enough return on investment to mullet fishermen
to encourage catching the mullet, even if no market exists for the
carcass. Since roe production is highly seasonal, and 'spent' or
cut mullet carcasses cannot be successfully stored, large volumes of
carcasses are available during roe season and must be sold at very low
prices. The fillets, an excellent protein source, could be recovered as an
LTS product for additional profit.
The LTS fillets are a unique seafood form which could provide
consumers with a new seafood choice. A broader product choice should
increase the overall demand for seafoods. Further, it may be possible
to introduce certain underutilized fish, i.e. jacks, croakers, bluefish,
etc., in the form of LTS fillets.
Initial product development work concentrated on the utilization of
mullet, Mugil cephalus. Large mullet were harvested in the fall of 1979
with commercial gill nets set along the West coast o. Florida. Ripe roe
was removed from the female mullet and marketed through commercial channels.
Boneless and skinless fillets were hand cut from the remaining fresh
carcasses. The fillet yields were approximately 20-25% of the live weight.
Tests were designed to determine the effects of brine concentrations,
smoking times, methods of cooking, and frozen storage on acceptance of
LTS mullet fillets. Figure 1 outlines the standard production process.
After the fillets were cleaned and/or thawed, the standard production
process consisted of soaking in a brine solution, air drying, low tem-
perature smoking, packaging, and storage (Figure 1).
The effects of various brine concentrations (0, 2, 4, and 6 percent
salt) on product preference was determined using the standard production
process. A two percent brine concentration equalled two cups of salt
per nine gallons of fresh water, four percent equalled four cups, etc.
The smoking time was 1 1/2 hours at 1200F (490C). The LTS
samples were deep fat fried at 350F (1770C) for two minutes. Unidentified
samples of the finished products were presented to taste panelists, who
scored their level of product preference per brine concentration. The
highest percentage of panelists preferred LTS fillets presoaked in a four
percent brine (Table 1) which indicated the four percent brine provided
,the preferred flavor. The results indicated the panelists expected a salt
flavor with smoked fish but excessive salt was objectionable. On the
(Boneless and Skialess)
CLEAN AND/OR THAW
COLD SMOKING (1200F)
PACKAGE AND STORAGE (frozen)4
PACKAGE AND STORAGE (frozen)
I COOK AND SERVE
Figure 1. Standard process for the production LTS fish fillets.
basis of these results, a brine concentration of four percent was selected
for use in continuing studies.
Table 1.--Effect of brine concentration on panel preference for fried
LT mullet fillets previously smoked for 1 hours.
Percent Liked Best Liked Least
Brine 1 2 3 4 Totalsa
0 0 2 13 85 100
2 12 30 52 7 101
4 53 38 5 3 99
6 35 30 30 5 100
aTotals percentages may not sum to 100 percent due to rounding.
percentages are based on 30 preference panelist scoring obtained
in three test replications for a total of 90 observations per
The effects of various smoking times (no smoking, 1, and 3 hours)
at 1200F (490C) on product preference was determined using standard
procedures. All fillets were presoaked in a four percent brine for 30
minutes prior to smoking. All smoking was done in a Koch Grandprise
smokehouse with an external smoke generator. Smoke was produced from
hickory dust smoldering on electric resistance coils. The smoking
chamber was vented for air flow, but there were no humidity controls.
The air vents were fully opened during the first half hour of smoking and
closed during the final hour.
Unidentified fried samples of the finished product were presented to
a 20 member sensory panel. Fillets used in this test were not previously
frozen. The panelists rated product aroma, color, texture, flavor, and
overall preference based on a 9 point scale (9 = excellent; 1 = extremely
poor). The results (Table 2) indicated the panelists preferred some smoke
time, but there was no significant difference in preference for product
smoked 1 or 3 hours. The longer smoke time, however, produced a slightly
darker, tougher product. Based on the highest overall rating and energy
conservation, the 1 hour smoke time was selected for further studies.
Table 2.--Effect of smoking time on panel preference for fried LTS
mullet fillets previously soaked in four percent brine.
Time of Smoking
Product Characteristics 3 hrs. 1i hrs. None
Aroma 7.1 7.1 6.7
Color 6.4 7.3 6.4
Texture 6.8 7.3 7.0
Flavor 7.2 7.0 6.2
Overall preference 7.2* 7.5* 6.3t
aRatings are based upon a preference scale where 9 = excellent and
1 = extremely poor. Mean ratings are based on 20 preference panelist
ratings obtained in two replications per smoking time.
bAccording to a Duncan's Multiple Range test, mean ratings for
overall preference followed by the same symbol are not different
at the 0.05 level of significance.
Packaging and Storage
Routine packaging methods were used to prepare the finished LTS
fillets for frozen storage. Three to five pounds of fillets were
initially wrapped in freezer paper, then wrapped with aluminum foil.
The packages were placed on freezer shelves in single layers to assure
rapid cooling. The packages were noticeably hardened in less than eight
hours after e.. sure to the freezer temperature of -200F (-290C). The
packages remained frozen (-200F or -290C) until further testing. No
tests were conducted on packaging methods but subsequent tests were con-
ducted to determine the effect of prolonged frozen storage.
To determine if there was a preferred method of final preparation,
the LTS fillets were fried in peanut oil for two minutes at 3500F
(1770C), broiled for six minutes in an electric oven, and baked in an
aluminum foil wrap for 20 minutes at 4500F (2320C). The fillets were
prepared by the standard production process using a four percent brine
and 1 hour smoke time at 1200F (490C). Fillets used for these tests
were not previously frozen. Panel preference for cooked samples were
rated on the same nine point scale (9 = excellent; 1 = extremely poor).
Panelists rated product aroma, color, texture, flavor, and overall
preference. The results (Table 3) indicated all three cooking methods
are acceptable. There was no significant difference in overall preference
for any one final product preparation, but the fried product was rated
the highest preference for color and flavor. Frying tended to darken the
product from a pale yellow to an attractive golden brown color. Continued
frying caused excessive browning. Broiling and baking caused a white,
gelatineous exudate to form on the surface. These proteinaceous substances
gave an objectionable product appearance.
Table 3.--Effect of cooking procedure on pinel preference for LTS
Aroma 7.1 6.5 7.0
Color 7.2 6.3 6.4
Texture 6.6 7.0 6.7
Flavor 7.4 6.7 7.0
Overall preference 7.1* 6.7* 7.0*
aRatings are based upon a preference scale where 9 = excellent and
1 = extremely poor. Mean ratings are based on 20 preference panelists
ratings obtained in two replications per cooking method.
bAccording to Duncan's Multiple Range test, mean ratings for overall
preference followed by the same symbol are not different at the
0.05 level of significance.
RECOMMENDED PRODUCTION PROCESS
Based on the product development work, the recommended basic pro-
duction process is outlined below. This process was developed utilizing
mullet fillets, but has been successfully used on fillets from grouper,
snapper, cod, and Spanish mackerel. Random check weighing for the various
species indicated the proper procedure should provide a product yield in
excess of 98 percent of the initial raw fillet weight.
It should be noted that this procedure has been developed using only
one particular smokehouse. Variations in thermal characteristics, air-
flow rates, and humidity control in different smokehouses could influence
the required smoking time and exact smoking temperature. Likewise, the
thickness of the fish fillet could influence the required brine concen-
tration and smoking time. Variations in the ratio of dark muscle to white
muscle tissue could affect product appearance and flavor. Production
procedures must be determined for respective types of smokehouses and
fish, but the recommended production process should serve as an approximate
Clean fish fillets (skinless and boneless) are soaked in a prechilled
(400F or 450C) salt brine. The recommended salt concentration is four
percent (four cups salt per nine gallons water). The soak time should be
no less than 30 minutes. Occasional, gentle stirring will assist the soak-
ing process. After brining, the fillets should be washed with a light
rinse of clean water. The soaked fillets should be air dried on racks held
in refrigeration until a glaze-like pellicle develops on the surface. A
drying time of approximately 30 minutes has been found adequate.
Racks of fillets should be placed in a preheated smokehouse. The
fillets should be smoked for 1 hours at 1.0 F (490C) in moderate smoke.
If humidity controls are available maintain a relative humidity of
approximately 60 percent. Smoking temperatures between 80 to 1200F
(270 to 490C) may be effective depending on the characteristics of
different smokehouses and fillet thicknesses. Air flow rates can be
adjusted to control dehydration of the product. The finished product is
not cooked, but has a pale yellow, damp appearance and the surface
flesh becomes firm.
Packaging and Storage
The LTS fillets must be store frozen at 00F (-200C) or below.
The fillets should be packaged layered with freezer paper and wrapped
in plastic bags. Avoid bulk packaging to permit a more rapid freeze.
Thaw frozen fillets in refrigeration overnight; then fry, bake, or
broil as desired. Deep fat frying at 3500F (1770C) until golden brown
is an excellent cook method. Frying does not require batter or breading.
PRODUCTION WITH VARIOUS SPECIES
Application of the basic production process was determined for four
different fish species. These were grouper (Epinephelus sp.), snapper
(Lutjanus sp.), cod tails (Gadus morhua), and Spanish mackerel
(Scomberomorus maculatus). All fish fillets had been previously frozen
for at least one month. Only boneless, skinless fillets were.used from
each species. All raw fillet sizes were within a six to ten ounce range.
The fillets were thawed, then processed by the recommended basic process.
The finished products were fried by standard procedures then presented
to a 20 member preference panel for evaluations of aroma, color, texture,
flavor, and overall preference. Remaining fillets were packaged and
frozen for storage studies.
The panel results indicated there was essentially no difference in
the product characteristics for LTS fillets made from grouper, snapper,
or cod tails. No statistical tests were used to compare means, but there
was a higher overall preference for LTS mackerel fillets (Table 4). The
mackerel may have provided a better flavor and texture due to the higher
fat content. It is interesting to note that the overall preference for
these species was similar to that recorded for fried mullet fillets.
Table 4.--Sensory evaluations of LTS fillets prepared from various
Fish Product Characteristics Overall
Aroma Color Texture Flavor
Mackerel 7.8 7.8 7.6 8.1 8.2
Cod 7.1 7.6 7.1 6.9 7.0
Grouper 7.2 7.8 7.1 7.0 7.0
Snapper 6.9 7.3 6.8 7.2 6.9
Mulletb 7.1 7.2 6.6 7.4 7.1
aRatings are based on a 9.0 scale where 9 = excellent
poor. Mean ratings are based upon 20 observations.
tests were used to compare means.
and 1 = extremely
bThe sensory evaluations for fried mullet are taken from Table 3
for the purpose of comparison.
Storage studies were conducted to determine the influence of frozen
storage on the quality of LTS fillets. Panelists rated general preference
for LTS fillets prepared from whole, fresh mullet and whole mullet which
Shad been frozen (-200F or -290C) for six weeks. All LTS fillets were
fried at 350F (1770C) for two minutes prior to presentation to the
panelists. Half of the LTS fillets prepared from the whole, fresh mullet
were frozen (-200F or -290C) for two weeks prior to preference rating,
and the other half were cooked and rated immediately after production
or unfrozen. Likewise, the LTS fillets prepared from whole mullet which
had been previously frozen for six weeks prior to smoking were divided
into two groups, LTS fillets not frozen and LTS fillets frozen two weeks
before cooking and rating. This design allowed evaluations of the final
cooked product as influenced by previous frozen storage of the initial
raw carcass prior to smoking or frozen storage of the LTS fillets, and
combinations of the two storage treatments.
The results indicated six weeks frozen storage of whole mullet prior
to smoking did not have any apparent detrimental effect on final product
preference (Table 5). Likewise, two weeks frozen storage of the finished
product did not decrease product preference. Average product ratings in
all sensory categories were higher for the frozen products, but there was
no significant difference in overall general preference for the LTS fillets
which had never been frozen versus the frozen LTS fillets. These results
tend to indicate frozen storage of the initial raw carcasses and of the
finished product is not detrimental to production of the LTS fillets.
Table 5.--Effect of frozen storage on panel preference for fried LTS
Source of Fillets
Fresh Mullet 6 Weeks
LTS Fillets LTS Fillets
Product Not Frozen Not Frozen
Characteristics Frozen 2 Weeks Refrozen 2 Weeks
Aroma 7.7 8.3 7.8 7.9
Color 7.6 8.4 8.3 8.2
Texture 8.0 8.1 7.9 7.6
Flavor 7.8 8.4 8.1 7.9
Preference 8.0* 8.3* 8.2* 8.1*
aRatings are based upon a scale where 9 = excellent and 1 = extremely
poor. Mean ratings are based on 20 preference panelists ratings ob-
tained in two replications per fillet type, frozen or unfrozen.
According to Duncan's Multiple Range test, mean ratings for overall
preference followed by the same symbol are not different at the
0.05 level of significance.
A second series of extended storage studies were conducted to deter-
mine the influence of long term frozen storage on LTS fillets. The fillets
were prepared as previously noted from grouper, snapper, cod tails, and
Spanish mackerel. The fillets were prepared by the recommended basic
process, then packaged and frozen for 30 and 90 days at -200F (-290C).
After storage, the fillets were thawed in refrigeration, then fried for
presentation to the preference panel. Panelists rated various sensory
characteristics on a 9.0 scale (9 = excellent; 1 = extremely poor).
There was a slight decrease in average overall preference after three
months of frozen storage, but the ratings were still judged to be accept-
able (Table 6). This slight decrease was due to lower ratings for texture
and flavor. Prolonged frozen storage had caused a slight increase in
product toughness and produced a slightly detectable off-flavor described
as initial rancidity. Product aroma and color was stable during frozen
storage. If the original smoked fillets had been prepared from fresher
fish fillets, the frozen storage life could have been extended. The
research results indicate that a frozen storage period of three to four
months is acceptable.
Table 6.--Sensory evaluations of influence of frozen storage on LTS
fillets prepared from various fish species
Storage Time (days)
Characteristics Speciesa 0 30 90
aSM = Spanish mackerel; C = Cod tails;
Ratings are based on a
G = Grouper; S = Snapper
9.0 scale where 9 = excellent; 1 = extremely
A comprehensive market survey was designed to determine the market
potential for the LTS fillets. The basic objective was to assess con-
sumers' reactions to this new product form and to determine any necessary
product improvements. These results may encourage industry commerciali-
ation of the product, thereby expanding the processing and consumption
of Florida seafoods. A more complete description and analysis of the
market survey i. reported elsewhere3
Briefly, consumer test samples were LTS mullet fillets prepared
by the recommended basic process. Each cold smoked fillet had been
previously frozen for approximately one month. The fillets were thawed
at 40F (40C) and deep fat fried in peanut oil at 3500F (1770C) for
two minutes. The six ounce cooked fillets were cut into two ounce sam-
ples and presented warm to consumers. The fish species was not identi-
fied to the consumers. Face-to-face interviews were conducted immediately
after the consumers had sampled the test product. The interviews were
conducted by two independent marketing firms located in relatively large
shopping malls in Tampa and Jacksonville, Florida. At each location, the
firms supplied test kitchens for sample preparation and professionally
The principle investigators designed the questionnaires and monitored
the interview process. All consumers were prescreened to select those
that eat fish and those that were over 18 years of age. Consumers were
selected to assure an equal representation of sex, but there were no
3Degner, R. L., W. Steven Otwell, and John A. Koburger. 1980. Consumer
Acceptance of Low Temperature Smoked Fish Fillets, FAMRC Industry
Report 80-3. Food and Resource Economic Department, IFAS.
additional screening criteria. A total of 402 consumers were interviewed,
200 in Jacksonville and 202 in Tampa. The interview included twenty ques-
tions concerning product characteristics, suggested improvements, buying
intentions at various price levels, preferred package size, and species
identification. The entire interview required approximately five minutes.
Results from the consumer surveys indicated a very positive consumer
reaction to the test product. There were nc significant differences in
consumer responses from the two cities which enabled the combining of
responses for most analyses. Consumer ratings for the product character-
istics indicated that the test product had E. very high appeal (Table 7).
Consumers generally agreed that the product would be acceptable as a
family meal, as a special meal for friends, or as a restaurant item. The
overall rating for the LTS mullet fillets was significantly higher (7.9)
than that recorded for typical hot smoked fish (6.9) that the respondents
had previously eaten.
The mean ratings for the product color, flavors, and texture indicated
the consumer felt the final product was nearly 'just right' (Table 8).
Ninety-one percent of the consumers, identified as the 'primary family food
shopper', said they would buy the product if available (Table 9). The
preferred package size would be 12 to 16 ounces (Table 10). Over 50
percent of the primary food shoppers would be willing to substitute the
cold smoked fillets for currently available frozen fish fillets at least
1/3 to 1/2 of the time (Table 11).
Table 7.--Consumer ratings of characteristics of the test product and
previously eaten smoked fish.
Characteristics Both Citiesa Tampa Jacksonville
Smell 7.8 --- ----
Overall Taste 8.2 --- ---
Overall Appeal 7.9 --- ---
As a Family Meal ---- 7.4 8.0
As a Special Meal for Friends ---- 6.4 7.0
As a Restaurant Meal 6.6 --- ---
Rating of Previcusly Eater
Smoked Fishb 6.9 --- ---
aMeans are based on a rating scale where 10 = excellent and 0 = ex-
tremely pcort. Where only one mean is reported for both cities, a
t-test indicated that differences between cities were not statis-
tically significant at the 0.05 level.
A paired t-test indicates the difference between the overall appeal
rating and the rating given previously eaten smoked fish is statis-
tically significant at the 0.01 probability level, t = 5.49 with
319 degrees of freedom.
Table 8.--Consumer ratings of physical attributes of the smoked fish fillets.
Meana Standard Percent of b
Attribute value deviation respondents
Exterior color 2.6 0.6 ---
Much too dark ---- --- 2.2
Slightly too dark ----- ---- 33.1
Just right --- --- 63.2
Slightly too light ---- ---- 1.5
Much too light --- ----- 0.0
Interior color 2.7 0.5 ---
Much too dark ----- ----- 1.2
Slightly too dark ----- ---- 30.9
Just right ---- ---- 66.7
Slightly too light ----- ----- 1.0
Much too light ----- ----- 0.3
Smoked flavor 3.0 0.6 ---
Much too Smokeyy --- ---- 2.2
Slightly too Smokeyy" ---- ---- 7.7
Just right .---- -.-- 78.4
Not quite enough smoked flavor ---- ---- 11.7
Need much more smoked flavor ----- ----- 0.0
Texture 2.8 0.5 ---
Much too tough ---- ----- 0.5
Slightly too tough ---- ----- 28.1
Just right --- ----- 69.2
Not quite tough enough --- --- 2.2
Need to be much tougher --- ---- 0.0
Salt 2.9 0.6 ---
Much too salty ---- ---- 2.0
Slightly too salty --- ----- 16.9
Just right ---- ---- 65.9
Not quite enough salt ---- --- 14.9
Needs much more salt ----- ----- 0.3
aMeans were calculated by assigning numerical values of 1-5, respectively,
to the semantic differential scales in the order listed. Thus, a mean of 3.0
would indicate a "just right" rating on each attribute. According to Chi-square
tests for each attribute there were no statistically significant differences in
ratings between the two cities. There was a total of 402 observations.
bTotal percentages may not sum to 100 percent due tc rounding.
Table 9.--Primary food shoppers' purchase intentions
Purchase intentionsa Number Percent
Yes, would buy i- available 171 91.0
No, would not buy 17 9.0
Totals 188 100.0
aChi-sqiare Enalysis indicates no statistically significant differ-
ence in purchase intentions by city, X2 = 0.90, with 1 degree of freedom.
Chi-square analysts for purchase intentions by age, income, race, or
household size were not statistically valid because of sparse numbers
Table 10.--Primary food shoppers' preferred package sizes for frozen
package size Number Percent
8 28 16.4
12 38 22.2
16 63 36.8
32 21 12.3
48 10 5.9
Various 11 6.4
Totals 171 100.0
aWhen the 32 and 48 ounce package classifications are combined and
the "various" category eliminated, chi-square analysis indicates no
statistically significant difference in package size preferences between
cities, X2 : 2.73, with 3 degrees of freedom.
bThe "various" size category includes responses that ranged from
0.33 to 6 pounds.
for the test product.
Table 11.--Primary food shoppers' indicated substitution of the test
product for currently available frozen fish fillets.
Substitution Tampa Jacksonville Both Cities
Percent Number Percent Number Percent Number Percent
100 7 9.1 23 25.3 30 17.9
75 8 10.4 11 12.1 19 11.3
33-50 39 50.7 46 50.6 85 50.6
20-25 23 29.9 11 12.1 34 20.2
Totals 77 100.0 91 100.0 168 100.0
aChi-square analysis indicates a statistically
between cities at the 0.01 probability level.
degrees of freedom.
X = 12.74, with 3
bTotal percentages may not sum to 100 percent due to rounding.
The LTS fillets were also perceived as a desirable restaurant item
(Table 12). Over 68.5 percent of all the respondents expressed a will-
ingness to order the LTS fillets if available in restaurants. Responses
were categorized by their typical frequency for ordering regular fish
fillets. The responses indicated 78.3 percent of the frequent purchasers
of regular fish would order the LTS fillets, and more impressively, 55.7
percent of the respondents that were 'never' purchasers of regular fish
in restaurants would order the LTS fillets if available.
Table 12.--Respondents' intentions to order the test product if available
in restaurants, by current frequency of fish fillet orders.
Order intentions for test product
Current frequency of Number of Do not
fish fillet nrr-rs respondents Yes No know Totals
Never 61 55.7 36.1 8.2 100.0
Infrequently, less than
once per month 108 63.9 29.6 6.5 100.0
Frequently, one to three
times per month 175 78.3 16.6 5.1 100.0
Very frequently, once
per week or more 56 60.7 30.4 8.9 100.0
All respondents 400 68.5 25.0 6.5 100.0
Since mullet is often considered a low value fish, it was interesting
to note that 76.0 percent of all consumers could not identify the fish
species prepared as the test product (Table 13). In Tampa, where mullet
is a more commonly eaten fish, a significantly larger percentage of
consumers could identify the test product as mullet fillets. However,
the identification of the fish did not influence the overall product
Table 13.--Respondents' ability tc correctly identify species from which
test products were made.
Response Tampa Jacksonville Both cities
Number Percent Number Percent Number Percent
Did not know
species 77 38.7 83 41.1 160 39.9
species 65 32.7 29 14.4 94 23.4
species 57 28.6 90 44.6 147 36.7
Totalsb 199 100.0 202 100.0 401 100.0
Total percentages may not sam to 100 percent due to rounding.
aChi-square analysis indicates a statistically significant
in responses between cities at the 0.01 probability level, X2 =
2 degrees of freedom.
This report has provided the basic production process, expected
frozen storage life, and consumer evaluations of a new product form which
could increase the use of traditional and underutilized fish harvested
in Florida. Results indicate that the product could be produced with
existing processing facilities and the product form suits the basic re-
quirements for retail outlets, including most restaurants. The very
positive consumer acceptance should encourage industry exploitation.
Industry production of LTS fillets will require necessary process
modifications to assure volume production with quality control. This
report supplies the basic starting methods. Refinements in brining,
spice formulations, smoking, and packaging may be required to assure
company standards. At present, there is no one best method for produc-
tion of LTS fish fillets. Future production of these products will
depend on industry innovation.