IJFS#1448_bozza


	

Ital. J. Food Sci., vol. 31, 2019 - 416 

SURVEY 
 
 
 
 
 

HOUSEHOLD CHARACTERISTICS INFLUENCING 
FISH CONSUMPTION IN VAN PROVINCE, TURKEY 

 
 
 
 
 
 

M. TERİN* 
Van Yuzuncu Yil University, Faculty of Agriculture, Department of Agricultural Economics, 

65080 Van, Turkey 
*Corresponding author: Tel.: +905356478210 

E-mail address: mustafaterin@yyu.edu.tr 
 
 
 
 

ABSTRACT 
 
This study investigates the relationship between households’ fish consumption frequency 
and their socio-demographic characteristics and attitudes. Using Chi-square test of 
independence, the study compares households’ fish consumption frequencies of never, 
once a month, twice a month, once a week and more than once a week. The empirical 
model was estimated using an ordered probit model to obtain the coefficients applied to 
the calculation of marginal effects and probabilities. The results indicate that, households’ 
income, children per households, working households’ head, households’ consumption of 
aquaculture products other than fish and the surveyed being households’ head 
significantly influence the frequency of fish consumption. 
 
 
 

Keywords: fish consumption, socio-demographic characteristics, ordered probit model, Turkey 
 



	

Ital. J. Food Sci., vol. 31, 2019 - 417 

1. INTRODUCTION 
 
In recent years, healthy nutrition has increasingly been encouraged, as a result, different 
healthy food consumption tendencies are emerged (GILBERT, 2000; LEEK et al., 2000). Sea 
food is an important part of healthy nutrition (TRONDSEN et al., 2003). Regular fish 
consumption reduces the likelihood of many chronic diseases including cardiovascular 
disease (KORNITZER, 2001; PLEADIN et al., 2017) and contributes significantly to healthy 
living (VERBEKE and VACKIER, 2005).  
Sea food is regarded as one of the most valuable nutrients in terms of the nutrients it 
contains. Sea food products contribute greatly to human nutrition because of high protein 
ratio, richness in omega-3 fatty acids, and minerals and vitamins they contain 
(GÜLYAVUZ and ÜNLÜSAYIN, 1999). Especially, fish has many benefits to human 
nutrition. Fish meat is easy to digest, contains high protein and is excellent in fat content. 
In addition, the vitamins and minerals and the low energy of the dietary supplement 
increase its importance (TATAR, 1995; TURAN et al., 2006; SAYGI et al., 2015). These 
factors are main causes that led consumers to change their consumption preferences from 
red meat to chicken meat and fish meat (RICKERTSEN, 1996; MANGEN and BURRELL, 
2001).  
Aquaculture plays an important role in ensuring nutritional needs and global food 
security in both developed and developing countries. In the past 50 years, global average 
supply of fishery products has increased by 3.2% per year on average, and world 
population has increased by 1.6%, resulting in an increase in average per capita 
consumption of aquatic products. In the world, the average per capita annual 
consumption of aquatic products is estimated to be 20.5 kg in 2017, while it was 9.0 kg, 
17.0 kg and 20.2 kg in 1961, 2000 and 2015, respectively. This impressive increase in 
average fish consumption per capita was mainly due to the increase in production, 
income, population and urbanization, as well as the development of modern distribution 
channels (FAO, 2018).  
As a country surrounded by sea, Turkey has a significant potential for aquatic products 
with its lakes, dams, streams and spring waters. Fishing in Turkey is an important field of 
activity in terms of being one of the basic livelihood resources in the coastal regions and 
human nutrition (ANONYMOUS, 2014). Aquaculture production in Turkey has shown 
significant fluctuations over the years. In 2017, the production of aquaculture products in 
Turkey increased by 7.15% to 630 thousand tons compared to the previous year. Out of 
this production 354 thousand tones (56.2%) was obtained through hunting, and 276 
thousand tones (43.8%) were obtained through aquaculture. Van Province with the biggest 
Lake (Van Lake) accounts for 23.0% (8310 tons) of inland water fish production in Turkey 
with its Pearl Mullet (Tarek) fish unique for Van Lake (TURKSTAT, 2018). In addition, 
there are 24 trout farms in the province and approximately 200 tons of trout is produced 
yearly (GÜNGÖR, 2014). 
In Turkey, per capita consumption of aquatic products has ranged from 6.3 to 8.6 kg/year 
in the last 18 years and has been 5.5 kg/year as of 2017 (TURKSTAT, 2018). The amount of 
consumption of aquatic products per capita differs significantly between regions in 
Turkey. While per capita consumption was high at regions by seas in Giresun and Trabzon 
28.08 kg/year (AYDIN and KARADURMUŞ, 2013), in Mersin 25.8 kg/year (ŞEN, 2011) in 
Hatay 21.5 kg/year (DEMIRTAŞ et al., 2014) and in İzmir 15 kg/year (ÇAYLAK, 2013), 
domestic, Eastern and Southeastern regions were below the world average being 13 
kg/year in Tokat (ERDAL and ESENGÜN, 2008), 12.4 kg/year in Isparta (HATIRLI et al., 
2004), 6.5 kg/year in Erzurum (UZUNDUMLU, 2017), 4.13 kg/year in Kahramanmaraş 
(ERCAN and ŞAHIN, 2016), 3.8 kg/year in Niğde (BASHIMOV, 2017) and 3.4 kg/year in 
Ankara (GÜL YAVUZ et al., 2015). 



	

Ital. J. Food Sci., vol. 31, 2019 - 418 

There are many factors affecting fish consumption, including socioeconomic structure, 
general food consumption structure, personal health status and maritime nature of the 
living area (MYRLAND et al., 2000; TRONDSEN et al., 2004; VERBEKE and VACKIER, 
2005). But, the most determining factor for purchasing fish is nutrition (ADELI et al., 2011). 
The aim of the study in this context was to determine the socio demographic and 
behavioral characteristics that affect the frequency of fish consumption of households in 
urban areas in Van. 
 
 
2. MATERIAL AND METHODS  
 
The main material of the study is the original data collected through questionnaires from 
260 households living in the urban area of Van. Survey was conducted between December 
2015 and January 2016. The sample size was determined by ungrouped one stage random 
likelihood sampling method based on households (COLLINS, 1986; AKBAY et al., 2007). 
 
 𝑛 = 𝑡^2 [1 + (0.02)(𝑏 − 1)] ∗ 𝑝𝑞/𝐸^2   (1) 
 
The statistical relationship between the frequency of fish consumption of households and 
their socio-demographic and behavioral characteristics was estimated using the Chi 
square test. On the other hand, the effects of the socio-demographic and behavioral 
characteristics of the habits on the fish consumption frequency was estimated using 
“Ordered Probit Model" method. Statistical Package for Social Science (SPPSS 17.0) and 
LIMDEP 10 programs were used in the analysis of the data. 
The ordered probit model is based on the McFadden (1973) utility maximization theory. 
The utility function in the research indicates the utility of the consumer in terms of the 
frequency of fish consumption. However, the level of utility provided here cannot be 
observed. Behind the observable, intermittent and ordered categories (y) in the ordered 
probit model is assumed to be a continuous, but unobservable, hidden dependent 
variable. The unobserved, latent dependent variable (y*) is explained by the vector of 
explanatory variables and the error term. The term error is assumed to have normal 
distribution (GREENE, 2012). 
 
 Y* = x'β + ε ε ~ N [0, 1] (2) 
 
In the study, households chose one of the five alternatives for fish consumption, the 
dependent variable was classified according to its size (y = 0, 1, 2, 3, 4). Thus, the 
relationship between the model dependent variable (y) and the unobserved dependent 
variable (y*) is as follows (CHEN et al., 2002; GREENE, 2012). 
 
 if y*≤0, y=0 
 if 0<y*≤μ1, y=1 
 if μ1<y*≤μ2 y=2 
 if μ2<y*≤μ3 y=3 
 if μ3 ≤ y* y=4 
   
   
 
The dependent variable used in the model is one of the following categories: “y=0”, “y=1”, 
“y=2”, “y=3” and “y=4” which represents households’ non-fish consumers, households 
consumes fish once per month, households consume fish once per fifteen days, 



	

Ital. J. Food Sci., vol. 31, 2019 - 419 

households consumes fish per week and households consumes fish more than once per 
week, respectively. In the ordered probit model, the likelihood of the producers selecting 
one of five alternatives is as follows (GREENE, 2012). 
 
 Prob (y = 0|x) = Φ (− x’β),  (3) 
 Prob (y = 1|x) = Φ (µ1 − x’β) − Φ (− x’β), (4) 
 Prob (y = 2|x) = Φ (µ2 − x’β) − Φ (µ1− x’β), (5) 
 Prob (y = 3|x) = Φ (µ3 − x’β) − Φ (µ2− x’β), (6) 
 Prob (y = 4|x) = 1 −Φ (µ3 − x’β) (7) 
 
For all probabilities to be positive, 0<µ1<µ2<...<µj-1 Φ shows the cumulative normal 
distribution function. The solution of the model can be realized by "logarithmic maximum 
likelihood" method. The marginal effects of the variables are calculated as follows for each 
probability (GREENE, 2012). 
 
 (𝜕𝑃𝑟𝑜𝑏(𝑦 = 0|𝒙))/𝜕𝒙 = −∅(𝒙^′ 𝛽)𝛽,    (8) 
 (𝜕𝑃𝑟𝑜𝑏(𝑦 = 1|𝒙))/𝜕𝒙 = [∅(−𝒙^′ 𝛽) − ∅(µ1 −  𝒙^′ 𝛽)]𝛽,       (9) 
 (𝜕𝑃𝑟𝑜𝑏(𝑦 = 2|𝒙))/𝜕𝒙 = [∅(µ1 −  𝒙^′ 𝛽) − ∅(µ2 −  𝒙^′ 𝛽)]𝛽,         (10) 
 (𝜕𝑃𝑟𝑜𝑏(𝑦 = 3|𝒙))/𝜕𝒙 = [∅(µ2 −  𝒙^′ 𝛽) − ∅(µ3 −  𝒙^′ 𝛽)]𝛽,     (11) 
 (𝜕𝑃𝑟𝑜𝑏(𝑦 = 4|𝒙))/𝜕𝒙 = ∅(µ3 −  𝒙^′ 𝛽)]𝛽        (12) 
 
 
3. RESULTS AND DISCUSSION 
 
Out of households surveyed 89.2% consumed fish while the remaining 10.8% did not 
consume fish. In similar studies conducted in Van province, 88.2% and 78.53% of the 
consumers consumed fish, respectively (SARI et al., 2000, CEYLAN, 2006). In some other 
studies, the fish consumption rate of consumers was 95.8% (GÜRGÜN, 2006), 88.3% 
(ORHAN and YÜKSEL, 2010), 90.6% (BALIK et al., 2013), 98.8% (ONURLUBAŞ, 2013), 
84.0% (OLGUNOGLU et al., 2014), 95.0% (CICEK et al., 2014) and 96.52% (DJORDJEVIC et 
al., 2015). Out of the households who didn’t consume fish 35.7% stated the reason as fish 
odor followed by dislike fish, having no habit of fish consumption, and insufficient 
purchasing power with 25.0 21.4 and 17.9%, respectively. ORHAN and YÜKSEL (2010) 
stated that 60.50% of consumers didn’t consume fish due to odor followed by 12.12% 
having no habit of fish consumption in Burdur Province, Turkey.  
Out of surveyed households 10.8% didn’t consume fish at all. On the other hand, 27.3% of 
households consumed fish once per two months followed by once per week, once per 
month and more than once per week with 26.9, 22.7 and 12.3%, respectively (Table 1). In 
the study conducted by GÜNGÖR (2014), 36.2% of the consumers consumed fish once a 
month, followed by those who consumed fish once per fifteen days and once per week 
with 23.0 and 17.8%, respectively. CICEK et al. (2014) stated that 28.0% of consumers 
consumed fish once per fifteen days followed by those who consumed fish once per week 
and once per month with 25.0 and 23.0%, respectively in Elazığ Province, Turkey. SAYGI 
et al. (2015) pointed out that 25% of consumers in İzmir, Turkey consumed fish once per 
week while consumers in Ankara generally consumed fish once per fifteen days. In a 
study conducted in Serbia more than half of consumer at school age consumed fish once 
per week (52.24%) and 34.33% once per month (DJORDJEVIC et al., 2015). In Mexico one 
third of consumers (32.51%) consumed fish more than once per week followed by 25.62% 
once per week, 25.62% once per week, 24.24% once per fifteen days and 17.63% once per 
month (PEREZ-RAMIREZ et al., 2015). HICKS et al. (2008) pointed out in American 22.0% 
of consumers consumed fish more than once, 24.0% of consumers consumed fish once per 



	

Ital. J. Food Sci., vol. 31, 2019 - 420 

week, 29.02% of consumers consumed fish once per two-three months and 12.0% of 
consumers consumed fish once per month.  
The average monthly income of the households varied between TL 750 and TL 7500 and 
the monthly average household income was TL 2776.77. While the average monthly 
income of 7.7% of households was less than 1000 TL, 66.9% of households had a monthly 
income varied between TL 1001 and TL 3000 and 25.4% of consumers had income above 
TL 3001. In a study conducted in Van Province, 14.9% of the households had monthly 
income between TL 0-1000, 43.8% between TL 1001-3000 and 41.3% had income above TL 
3001 (GÜNGÖR, 2014). 
The average number of children per household was 2.61 (Table 1). Out of surveyed 
households, 26.9% had one child followed by 23.8, 16.9, 12.0 and 11.9% of households with 
three, one, five and four child, respectively. In a study conducted by CEYLAN (2006), 
55.21% of households in Van, Turkey had more than one child, 12.5% had one child and 
2.08% had no child. GÜRGÜN (2006) pointed out that 7.3% of households had no child, 
14.6% of households had two and 5.0% had one child in Bitlis, Turkey. In a study by 
VERBEKE and VACKIER (2005) in Belgium, it was found that 57.3% of the households 
consisted of families with children. 
 
 
Table 1. Descriptive statistics of the sample. 
 

Variables Values 
Dependent variable  
Fish consumption frequency  
Never                                 Y= 0 %10.8 
Once a month                   Y= 1 %22.7 
Twice a month                  Y= 2 %27.3 
Once a week                    Y= 3 %26.9 
More than once a week    Y= 4 %12.3 
Continuous explanatory variables Means 
Income (TL/month) 2776.77 (1453.18) 
Child number 2.61 (1.29) 
Binary explanatory variables Values 
Household head  (if 1; otherwise 0) %75.8 
Household head woman  (if 1; otherwise 0) %19.6 
House wife working (if 1; otherwise 0) %13.8 
Consumption other seafood except fish (if 1; otherwise 0)   %9.6 
Resides in rental house (if 1; otherwise 0) %33.1 
Fish prices high (if 1; otherwise 0) %62.3 
Public spots effect fish consume. positive (if 1; otherwise 0) %65.0 
Household head working (if 1; otherwise 0) %94.6 

 
Standard deviations are given in brackets. 
 
 
The average monthly fish consumption quantity per household and per capita was 6.3 kg 
and 1.4 kg, respectively. Average yearly fish consumption quantity per capita in Turkey is 
7 kg (TURKSTAT, 2018). In this case, yearly fish consumption quantity per capita in Van 
province is more than that of Turkey. Pearl mullet (only caught from Van Lake) consisted 
30.0% of the Turkey's inland fish production in 2017 (TURKSTAT, 2018). Considering 



	

Ital. J. Food Sci., vol. 31, 2019 - 421 

factors such as the price of pearl mullet, which is cheaper than the other fish types and 
being suitable for the taste of the local people, it is expected that the quantity of fish 
consumption per capita in the province of Van is relatively high. This is also confirmed by 
the research results. 
The statistical relationship between the socio-demographic and behavioral characteristics 
of the households and the fish consumption frequency is given in Table 2. A statistically 
significant relationship was found between socio-demographic characteristics such as 
households’ income, working household head, the households’ consuming aquatic 
products other than fish, the high prices of fish, positive effect of public spots and fish 
consumption frequency of households. It can be said that the results obtained are in 
accordance with the expectations. 
 
 
Table 2. Socio-demographic characteristics and several attitudes of the sample and fish consumption 
frequencies. 
 

Variables Never Once a month 
Twice 

a month 
Once 

a week 
More than 

once a week X
2 

Household Head       
Yes  10.2 20.8 26.4 27.9 14.7 

5.95 
No  12.7 28.6 30.2 23.8   4.8 
Household head woman      
Yes  13.7 21.6 27.5 25.5 11.8 

0.61 
No  10.0 23.0 27.3 27.3 12. 
House wife working       
Yes    2.8 22.2 27.8 33.3 13.9 

3.21 
No  12.1 22.8 27.2 25.9 12.1 
Consumption other seafood except fish     
Yes    0.0 16.0 8.0 48.0 28.0 

17.40*** 
No  11.9 23.4 29.4 24.7 10.6 
Renter       
Yes  10.5 24.4 26.7 27.9 10.5 

0.60 
No  10.9 21.8 27.6 26.4 13.2 
Fish prices high        
Yes  17.3 25.3 28.4 19.8   9.3 

 29.87*** 
No    0.0 18.4 25.5 38.8 17.3 
Public spots effects fish consumption positive    
Yes   0.0 24.3 32.5 30.2 13.0 

 59.47*** 
No  30.8 19.8 17.6 20.9 11.0 
Household head working      
Yes    9.3 21.5 27.6 28.5 13.0 

17.19*** 
No  35.7 42.9 21.4   0.0   0.0 
Income       
Less than 1000 TL 30.0 35.0 30.0   5.0   0.0 

 23.57*** 1001-3000 TL 10.3 24.1 28.2 27.0 10.3 
3001 TL and over   6.1 15.2 24.2 33.3 21.2 

 
***: 0,01 significant level. 
 
 



	

Ital. J. Food Sci., vol. 31, 2019 - 422 

The ordered probit model results of the socio-demographic and behavioral characteristics 
that affect the frequency of fish consumption of the households are given in Table 3. The 
ordered probit model was found to be totally statistically significant using likelihood 
method (p <0.000). The coefficients of the model were tested using z-rate and standard 
error. Estimated threshold values in the model indicate the numerical relationship 
between the utility function of the consumer and the consumption frequency (AKBAY et 
al., 2007; GUNDUZ and EMIR, 2010). In view of MADDALA (1983), the threshold values 
should be positive and μ1 <μ2 <μ3. The threshold values of the model were positive and 
statistically significant at 0.01 level. This shows that the consumption frequency categories 
households are arranged appropriately and the socio-demographic and behavioral 
characteristics of the households are influential on fish consumption.  
The marginal effects of the socio-demographic and behavioral characteristics that affect 
the frequency of fish consumption of the households are given in Table 4. While model 
results are interpreted, marginal effects and coefficients are discussed together. In the 
study, a positive and statistically significant relationship was found between the 
household income and the fish consumption frequency (Table 3). As a result, the increase 
in household income will increase the frequency of fish consumption of the households. A 
TL 1,000 increase in household income will result in reducing the likelihood of no 
consumption (y=0) and monthly consumption (y=1) by 2.0% and 3.6%, respectively while 
increase the likelihood of consuming in a weekly (y=3) and more than once a week (y=1), 
by 3.5% and 2.5%, respectively (Table 4). A positive relationship between household 
income and the frequency and amount of fish consumption has been found in previous 
studies (AKINBODE and DIPEOLU, 2012; CAN et al., 2015; DAUDA et al., 2016). 
In the study, a positive and statistically significant relationship was found between the 
number of children of households and the frequency of fish consumption (Table 3). As a 
result, one more child in the households will reduce the likelihood of no-consumption 
(y=0) and consumption per month (y=1) by 1.08 and 1.92%, respectively, while it will 
increase the likelihood of consumption once per week (y=3) and consumption more than 
once per week (y=4) by 1.85 and 1.31%, respectively (Table 4). In a study by MYRLAND et 
al. (2000), it was found that the frequency of fish consumption increases by having 
children between 0-7 years and 8-12 years, having children between 0-7 age group reduce 
the likelihood of no-consumption and consuming per month by 1.9 and 1.4%, respectively 
while increasing the likelihood of consuming once per week and once per two weeks by 
2.9 and 0.9%.  
In the study, a positive and statistically significant relationship was found between the 
heads of households and the frequency of fish consumption (Table 3). As a result, the head 
of households will reduce the likelihood of no-consumption (y=0) and consumption per 
month(y=1) by 6.36 and 9.26%, respectively, while it will increase the likelihood of 
consumption once per week (y=3) and consumption more than once per week (y=4) by 
9.39 and 5.68%, respectively. 
A positive and statistically significant relationship was found between the consumption of 
other aquaculture products except fish and the frequency of fish consumption (Table 3). 
As a result, the households’ consumption of aquaculture products except fish will reduce 
the likelihood of no-consumption (y=0) and consumption per month (y=1) by 5.55 and 
13.58%, respectively, while it will increase the likelihood of consumption once per week 
(y=3) and consumption more than once per week (y=4) by 11.09 and 13.52%, respectively 
(Table 4). MYRLAND et al. (2000) pointed out that households’ consumption of 
aquaculture products would increase the fish consumption frequency, namely would 
decrease the likelihood of non-consumption and consumption per month frequency by 
16.2 and 23.7%, respectively while would increase the likelihood of consumption 
frequency of once per week and twice per week by 32.7 and 24.9%, respectively.  



	

Ital. J. Food Sci., vol. 31, 2019 - 423 

In the study, a positive and statistically significant relationship was found between the 
households’ finding fish prices high and the frequency of fish consumption (Table 3). As a 
result, the households’ considering the fish prices high will reduce the households’ fish 
consumption frequency. The households’ considering the fish prices high will increase the 
likelihood of no-consumption (y=0) and consumption per month (y=1) by 6.90 and 13.15%, 
respectively, while it will reduce the likelihood of consumption once per week (y=3) and 
consumption more than once per week (y=4) by 12.14 and 10.24%, respectively (Table 4).  
In the study, a positive and statistically significant relationship was found between the 
households’ thought that public spots had positive effects on fish consumption and the 
frequency of fish consumption (Table 3). The households’ thought that public spots had 
positive effects on fish consumption will reduce the likelihood of no-consumption (y=0) 
and consumption per month (y=1) by 9.71 and 13.97%, respectively, while it will increase 
the likelihood of consumption once per week (y=3) and consumption more than once per 
week (y=4) by 14.11 and 8.97%, respectively (Table 4).  
A positive and statistically significant relationship was found between the working 
households’ head and the frequency of fish consumption (Table 3). As a result, 
households’ head working at any place will increase fish consumption frequency. The 
working households’ head will reduce the likelihood of no-consumption (y=0) and 
consumption per month (y=1) by 15.28 and 13.84%, respectively, while it will increase the 
likelihood of consumption once per week (y=3) and consumption more than once per 
week (y=4) by 16.19 and 7.31%, respectively (Table 4).  
 
 
Table 3. Estimates ordered probit model for fish consumption frequencies. 
 

Variables Coefficient St. Error z-statistic p-value 
Constant   -0.2056 0.402  -0.51 0.6089 
Income       0.00016        0.0000546   2.91    0.0036*** 

Child number  0.084 0.043   1.94  0.0528* 

Household head  0.422 0.168   2.51    0.0120*** 

Household head woman -0.158 0.189  -0.83 0.4040 
House wife working -0.105 0.233  -0.45 0.6518 
Other seafood consumption  0.630 0.242   2.61   0.0091** 

Resides in rental house  -0.147 0.153  -0.96 0.3352 
Fish prices high -0.591 0.142  -4.15    0.0000*** 

Public spots  0.648 0.143   4.54    0.0000*** 

Household head working  0.754 0.333   2.26   0.0238** 

Threshold parameters      
μ (1)  1.031     0.08262 12.48    0.0000*** 

μ (2)  1.856     0.08267 22.45    0.0000*** 

μ (3)  2.879     0.11400 25.26    0.0000*** 

Log likelihood = -357.910 Restricted log likelihood = -400.952 
Likelihood ratio statistic (LR) = 86.084 Chi-square(10;0.05)=18.307 LR>Chi-square=0.000 

 
*:0.1, **:0.05 and ***: 0,01 significant level. 
 
 
 
 
 



	

Ital. J. Food Sci., vol. 31, 2019 - 424 

Table 4. The marginal effects of factors on the probability of relative frequencies for fish consumption. 
 

Variables Y(0) Y(1) Y(2) Y(3) Y(4) 
Income       -0.000020***     -0.000036***       -0.0000031       0.000035***       0.000025*** 

Child number  -0.01078* -0.01921*  -0.00164   0.01850*   0.01313* 

Household head   -0.06366**   -0.09256***   0.00552    0.09392**     0.05678*** 

Household head woman 0.2171 0.03566   0.00079 -0.03514 -0.02303 
House wife working   0.01426 0.02384   0.00084 -0.02338 -0.01556 
Other seafood consumption     -0.05501***   -0.13582*** -0.05534     0.11094***    0.13523** 

Resides in rental house   0.01955 0.03337   0.00176 -0.03252 -0.02216 
Fish prices high      0.06900***    0.13151***   0.02331    -0.12144***    -0.10239*** 

Public spots     -0.09714***   -0.13973***   0.00609     0.14111***     0.08967*** 

Household head working  -0.15276   -0.13839***   0.05612    0.16191**     0.07312*** 
 

*:0.1, **:0.05 and ***: 0,01 significant level. 
 
 
4. CONCLUSIONS 
 
Result of this study showed that various socio-economic and demographic factors of 
households and households’ heads significantly influenced the likelihood of consuming 
fish. There was a positive relationship between the socio-economic and demographic 
characteristics of households such as households’ income, children per households, 
working households’ head, households’ consumption of aquaculture products other than 
fish, households’ head and the behavioral variables such as households’ thought that 
public spots affected the fish consumption positively.  
Based on the findings of the study, the following recommendations were made; public or 
private organization should continue to educate the households’ heads (parents) on the 
importance of fish on their health. Price of fish should be reduced so as to increase the fish 
consumption in the area since it was observed that price of fish and fish consumption are 
inversely related. As the income of the households' increases, fish consumption also 
increases, therefore government should provide income opportunities by creating jobs to 
enhance the household’s purchasing power. In this way, it is possible to provide a 
healthier life for the society by encouraging the households to consume more fish. 
Educational programs regarding healthy and balanced nutrition in the region, should 
intended on the development of fish consumption habits of households in the region. 
 
 
REFERENCES 
 
Adeli A., Hasangholipour T., Hossaini A., Salehi H. and Shabanpour B. 2011. Status of fish consumption per capita of 
Tehran citizens. Iran J. Fish Sci. 10(4):546-556. 
 
Akbay C., Tiryaki, G.Y. and Gül A. 2007. Consumer characteristics influencing fast food consumption in Turkey. Food 
Control 18(8):904-913.  
 
Akinbode S.O. and Dipeolu A.O. 2012. Double-Hurdle model of fresh fish consumption among urban households in 
South-West Nigeria. Current Research Journal of Social Sciences 4(6):431-439. 
 
Anonymous 2014. Aquaculture private commission report. Publication No: 2871, Ankara. 
 
Aydın M. and Karadurmuş U. 2013. Consumer behaviors for seafood in Giresun and Trabzon province. The Black Sea 
Journal of Sciences 3(9):57-71.  



	

Ital. J. Food Sci., vol. 31, 2019 - 425 

Balık I., Yardımcı C. and Turhan O. 2013. Comparatively investigation of fish consumption habits of people in Fatsa and 
Aybastı districts of Ordu province. Ordu Univ. J. Sci. Tech. 3(2):18-28 
 
Bashimov G. 2017. Determining fish meat consumption habits in Nigde Province. Turkish Journal of Agricultural and 
Natural Sciences 4(2):196-204. 
 
Can M.F., Günlü A. and Can H.Y. 2015. Fish consumption preferences and factors influencing it. Food Science and 
Technology 35(2):339-346.  
 
Çaylak B. 2013. Fishery product consumption and consumer preferences survey in İzmir province Msc.Thesis, Çanakkle 
Onsekiz Mart University, Institute of Natural and Applied Sciences, Çanakkale, Turkey.  
 
Ceylan M. 2006. The consumption structure of meat and meat products in urban and rural area of van province and the 
purchase behavior of consumers Msc.Thesis, Yuzuncu Yıl University, Institute of Natural and Applied Sciences, Van, 
Turkey.  
 
Chen K.Z., Ali M., Veeman M.M., Unterschultz J. and Le T. 2002. Relative importance rankings for pork attributes by 
Asian-origin consumers in California:Applying an ordered probit model to a choice-based sample. Journal of 
Agricultural and Applied Economics 34(1):67-79.  
 
Çiçek E., Akgün H. and İlhan S. 2014. Determination of fish meat consumption habits and preferences in Elazığ province. 
Yunus Research Bulletin 1:3-11.  
 
Collins M. 1986. Sampling (Editors: Worcester R.M. and Downham J.) Consumer Market Research Handbook, 
London:McGraw-Hill 
 
Dauda A.B., Ojoko E.A. and Fawole B.E. 2016. Economic analysis of frozen fish demand in Katsina metropolis, Katsina 
State, Nigeria. Journal of Fisheries and Aquatic Science 11(1):93-99.  
 
Demirtaş B., Dağistan E., Akpınar M.G. and Sayın C. 2014. Fish consumption patterns of urban households in Hatay, 
Turkey. Journal of Academic Documents for Fisheries and Aquaculture 2:69-77.  
 
Djordjevic V., Sarcevic D. and Petronijevic R. 2015. The attitudes and habits of Serbian school children to consumption of 
fish. Procedia Food Science 5:73-76.  
 
Ercan O. and Şahin A. 2016. Analysis of Fish Meat Consumption at Kahramanmaraş City Centre. KSU J. Nat. Sci. 
19(1):51-65.  
 
Erdal G. and Esengün K. 2008. Analysis of factor affecting fish consumption with logit model in Tokat. E.U. Journal of 
Fisheries & Aquatic Sciences 25(3):203-209. 
 
FAO 2018. The state of World fisheries and aquaculture. ISBN 978-92-5-130562-1 www.fao.org/3/I9540EN/i9540en.pdf 
(28.08.2018). 
 
Gilbert L.C. 2000. The functional food trend: What’s next and what Americans think about eggs. Journal of the American 
College of Nutrition 19(5):507-512. 
 
Greene W.H. 2012. Econometric Analysis. Seventh Edition. Pearson Prentice-Hall, Upper Saddle River, New Jersey, 
07458. ISBN 10:0-13-139538-6 
 
Gül Yavuz G., Yasan Ataseven Z., Gül U. and Gülaç Z.N. 2015. Factor affecting consumer preferences on seafood 
consumption:The case of Ankara. Yunus Research Bulletin 1:73-82.  
 
Gülyavuz H. and Ünlüsayın M. 1999. Fisheries processing technology. Sahin printing press, Isparta, 366 
 
Gündüz O. and Emir M. 2010. Analysis of factors influencing frozen food consumption: A case study of Samsun, Turkey. 
J. Agric.Fac.HR.U. 14(3):15-24.  
 
Güngör E.S. 2014. A survey on fish consumption and consumer preference in Erzurum and van provinces Msc.Thesis, 
Atatürk University, Institute of Natural and Applied Sciences, Erzurum, Turkey.  
 
Gürgün H. 2006. An investigation into the fish consumption in some towns on the shores of Lake Van Msc.Thesis, Ege 
University, Institute of Natural and Applied Sciences, İzmir, Turkey.  
 
Hatırlı S.A., Demircan V. and Aktaş A.R. 2004. An analysis of households’ fish consumption in Isparta province. 
Süleyman Demirel University Journal of Faculty of Economics and Administrative Sciences 9(1):245-256. 
 



	

Ital. J. Food Sci., vol. 31, 2019 - 426 

Hicks D., Pivarnik L. and McDermott R. 2008. Consumer perceptions about seafood-an internet survey. Journal of 
Foodservice 19(4):213-226.  
 
Kornitzer M. 2001. Fish and health among adults. In: Descheemaeker K. and Provoost C. Impact of food on health-Recent 
developments, Antwerpen:Garant, 53-65. 
 
Leek S., Maddock S. and Foxall G. 2000. Situational determinants of fish consumption. British Food Journal 102(1):18-39.  
 
Maddala G.S. 1983. Limited-dependent and qualitative variables in econometrics. New York: Cambridge University 
Press. 
 
Mangen M.J. and Burrell A.M. 2001. Decomposing preference shifts for meat and fish in the Netherlands. Journal of 
Agricultural Economics 52(2):16-28.  
 
McFadden D. 1973. Conditional logit analysis of qualitative choice behaviour. In: P. Zarembka (Ed.), Frontiers in 
econometrics (pp. 105-142). New York: Academic Press. 
 
Myrland O., Trondsen T., Johnston R.S. and Lund E. 2000. Determinants of seafood consumption in Norway: lifestyle, 
revealed preference and barriers to consumption. Food Quality and Preference 11(3):169-188.  
 
Olgunoğlu İ.A., Bayhan Y.K., Olgunoğlu M.P., Artar E. and Ukav İ. 2014. Determination of habits of fish meat 
consumption in the province of Adıyaman. Electronic Journal of Food Technologies 9(1):21-25.  
 
Onurlubaş E. 2013. The factors affecting fish consumption of the consumers in Keşan Township in Edirne. Bulgarian 
Journal of Agricultural Science 19(6):1346-1350. 
 
Orhan H. and Yüksel O. 2010. Fishery product consumption survey in Burdur province. Süleyman Demirel University 
Journal of the Faculty of Agriculture 5(1):1-7. 
 
Perez-Ramirez M., Almendarez-Hernandez M.A., Aviles-Polanco G. and Beltran-Morales L.F. 2015. Consumer 
acceptance of eco-labeled fish:A Mexican case study. Sustainability 7:4625-4642.  
 
Pleadin J., Lesic T., Kresic G., Baric R., Bogdanovic T., Oraic D., Vulic A., Legac A. and Zrncic S. 2017. Nutritional quality 
of different fish species farmed in the Adriatic Sea. Ital. J. Food Sci. 29:537-549. 
 
Rickertsen K. 1996. Structural change and the demand for meat and fish in Norway. European Review of Agricultural 
Economics 23(3):316.330.  
 
Sarı M., Demirulus H. and Söğüt B. 2000. A research on determination of fish meat consumption habits of students in 
Van province. In Proceedings of the East Anatolian Region 4th Aquaculture Symposium, Erzurum, Turkey.  
 
Saygı H., Bayhan B. and Hekimoğlu M.A. 2015. Fishery products consumption in the cities of Ankara and İzmir in 
Turkey. Turkish Journal of Agriculture-Food Science and Technology 3(5):248-254. 
 
Şen A. 2011. Individuals living in Konya and Mersin provincial centers comparison of fish consumption on the habits 
and knowledge level Msc. Thesis, Selçuk University, Institute of Natural and Applied Sciences, Konya, Turkey.  
 
Tatar O. 1995. Nutritional properties of fish and healthy respect. E.U. Journal of Fisheries & Aquatic Sciences 12:169-170. 
 
Trondsen T., Braaten T., Lund E. and Eggen A.E. 2004. Health and seafood consumption patterns among women 45-69 
years. A Norwegian Fish Consumption Study 1996. Food Quality and Preference 15(2):117-128.  
 
Trondsen T., Scholdere J., Lund E. and Eggen A.E. 2003. Perceived barriers to consumption of fish among Norwegian 
women. Appetite 41(3):301-314. 
 
Turan H., Kaya Y. and Sönmez G. 2006. Fish meat value and place of human health. E.U. Journal of Fisheries & Aquatic 
Sciences 23:505-508. 
 
TURKSTAT 2018. Turkish Statistical Institute. Aquaculture statistics. www.tuik.gov.tr/PreTablo.do?alt_id=1005 
(28.08.2018).  
 
Uzundumlu A.S. 2017. Determining fish consumption behavior among households and the most suitable type of fish in 
Erzurum Province. Iran J. Fish Sci. 16(2):684-697. 
 
Verbeke W. and Vackier I. 2005. Individual determinants of fish consumption: application of the theory of planned 
behavior. Appetite 44(1):67-82.  
 

Paper Received November 29, 2018  Accepted February 24, 2019