Adoption of innovations and best management practices by goat farmers in eastern Mediterranean Region of Turkey

Goat farming has traditionally been a major livelihood for many rural families in the mountainous and uphill areas of the Eastern Mediterranean region of Turkey. In recent years, the increased demand for goat products raised the issue of developing a sustainable goat sector in the region. The primary purpose of this study was to determine best management practices and innovations in goat farming and their adoption levels in the Eastern Mediterranean Region of Turkey. For this purpose, 140 goat farmers were surveyed and their socioeconomic characteristics, management practices, problems encountered, levels of applying innovations and best management practices, as well as factors influencing their adoption were investigated. Results of the study revealed that goat farmers face problems with expensive feeds, low governmental subsidies, and cheap prices for goat products. Adoption level of innovations and best management practices were found quite low and it was influenced by farmers’ experience, income, travels, and contacts with extension service and private veterinarians.


INTRODUCTION
Goat farming has traditionally been the main livelihood of many rural families in the Eastern Mediterranean region of Turkey (Jordan et al., 2002).It represents a type of family farming in the mountainous areas of Adana, Osmaniye, Hatay, and Kahramanmaras provinces where farmland has environmental restrictions for cultivation.Using farm machineries such as tractors, cultivators, combines, and seeding machines is both uneconomical and erosion sensitive.As moving from low lands to uphills and mountainous areas the slope of landscape increases and farmlands turn in smaller and fragmented parcels where machinery use is more costly and in some cases technically impossible.If farming practices which require soil operations are insisted without taking adequate measures, they will trigger soil erosion and therefore weaken natural resource base which presently provides livelihood, even not self-sufficient for many rural families.
An agricultural system forcing farmers to cultivate lands in uphills and mountainous areas doesn't provide rural people with income which would make it possible for them and for their children to enjoy at least an averaged quality of rural life (Jordan et al., 2002).This farming system accompanied with lack of regular services of education, health care, and transportation in rural areas, force farmers to migrate big cities where more opportunities are to be found.
There have been at least two major driven forces which encourage goat producers in the uphill and mountainous landscape of the region.One of them is a gradually growing interest in goat meat and milk.According to many consumers in the region, goat is the most frequently consumed meat and is hardly subsidized for weal or muffon.Goat milk is also considered healthier as it is believed to have a lower level of fat (Coşkun and Öndül, 2004).
The second driven force is the continuously growing demand for goat milk which is the major input for internationally well-known "Kahramanmaras ice-cream".At least three large companies, namely, Mado, Edo, and Carpedo have exceeded regional even national borders and they started franchising activities in many cities around the Middle east and the Europe.Not long time ago they were purchasing goat milk solely from farmers around the province of Kahramanmaras.However, they have recently started to raise Saanen goats to produce their own milk, but due to the growing demand for Kahramanmaras Ice-cream, there have been shortages in goat milk and therefore mixing caw milk which inevitably reduces the quality of ice-cream.In order to overcome the goat milk shortage Kahramanmaras icecream companies have started to initiate contract farming with goat producers in mountainous areas.
The number of goats in Turkey and in the Eastern Mediterranean Region changes according to economic stability of the country and macroeconomic policies related to animal husbandry.In the last two decades, the sharpest decrease in the number of goats in Turkey and in the East Mediterranean region was in 2009 when the number of goats decreased to 4,9 million and 0,37 million, respectively (Turkish Statistical Institute -TSI, 2013).Later on governmental subsidies and project incentives for animal production gave their positive results and the number of goats for the year 2011 reached to 7,1 millions in Turkey, and 0,56 millions in the East Mediterranean Region (TSI, 2013).The increase rates were calculated as 45%, and 51%, respectively.
Literature review reveals that there have been studies on the adoption of innovations and best management practices among farmers in different parts of the world.Some of the studies concentrated on innovations regarding dairy farms (Jaisridhar et al., 2013;Barham et al., 2004;Foltz and Chang, 2002); forage crops (Lapar and Ehui, 2004); beef cattle farming (Johnson et al., 2010;Gillespie et al., 2007;Suppadit et al., 2006;Kim et al., 2005;Chaudhry et al., 1993), and sheep producers (Budak et al., 2011).However, there is no study which is directly focused on the adoption of innovations and best management practices in goat farming.Nevertheless, Smith (2010), Kaymakçi andDellal (2006), andKaymakçi (2002) gave basic principles and information about goat farming while Ozturk (1999) investigated the problems encountered by goat farmers in Kahramanmaras province of Turkey.
In order to achieve a sustainable goat production sector at large, it is important for goat farmers not only depend on the governmental subsidies and incentives but also the application of best management practices and innovations.Therefore, the overall purpose of the study was to determine the application levels of best management practices and innovations in goat farming and socioeconomic factors and information-seeking behavior influencing their adoption.The specific objectives are:

Area of study and sampling procedure
Basic material used for this study was information obtained by administering a questionnaire to 140 goat farmers in the Eastern Mediterranean Region of Turkey.Target population to which the findings of this research were to be generalized was all goat farmers operating in this region.In order to draw an accurate sample to represent this population, first of all an accessible population in which every goat farmers had an equal and independent chance of being included in the sample was determined.For this purpose, 36 villages from four provinces of the region (Adana, Hatay, Kahramanmaras, Osmaniye) were selected with the help of province directorates of the Ministry of Food Agriculture and Livestock.
Lists of goat farmers with their numbers of goats were obtained from district directorate offices.Because some district directorates did not have an updated list of statistics for every village, these villages were visited in advance to determine the goat farmers and the number of animals they owned.Lists of goat farmers from the 36 villages made the accessible population.Based on the number of goats owned by each farmer, the accessible population was divided in three strata.
Then Yamane (2009), stratified sample size determination formula was used accepting 5% error term from the mean and 95% confidence interval.
accepting 5% error term from the mean and 95% confidence interval.The number of sample size was determined as 140.This number was proportionally distributed to each of the three strata and respondents from each stratum were randomly selected.

Data collection procedure
A two section questionnaire was prepared to collect data; the first section included questions about goat farming, management practices, innovations, and animal care; the second section included questions about socioeconomic characteristics and information-seeking behavior of farmers.Kaymakçi (2002), Vincent (2005), Smith (2010), Tölü et al. (2011), Savaş et al. (2012), and Ceyhan (2012) were explicitly utilized for the questionnaires.Questionnaires were mostly filled in respondents' farms or houses.In several cases they were contacted while they were grazing their herd on mountains.Data were collected in January-June 2008 period.

Analitical procedure
Descriptive statistics including frequencies and percentages were used to analyze data regarding objectives 1, 2, and 4 while means and standard deviations were used for objective 3.For the fifth objective the ordered probit method was used to determine the extent to which selected socioeconomic characteristics and information-seeking behavior influenced the application of best management practices and innovations among goat farmers.The dependent variable of the model was constructed with three levels (0 = low level adoption, 1 = medium level adoption, and 3 = high level adoption).
For this purpose, 30 management practices and innovations regarding goat farming in the region were predetermined considering the literature reviewed and specific characteristics of the region.Considering the frequency distribution of the responses on the questions whether or not farmers applied these practices or innovations, three adoption categories were formed.Those who adopted at least 10 practices or innovations were assigned to the low level adoption category, those who adopted between 11 and 20 were assigned to the medium level adoption category, and finally those who adopted more than 20 were assigned to the high level adoption category.Two separate models (the first one for socioeconomic characteristics and the second one for informationseeking behavior) were run.The ordered probit model can be expressed as follows: (1) , where y* denotes the vector of unobserved dependent variable,  denotes a vector of coefficients, xi denotes a vector of explanatory variables, ε denotes a vector of error terms normally distributed N[0,1], y denotes the observed dependent variable with three adoption levels, and finally  denotes the threshold values which indicate the inclinations of adoption (Greene, 2012).Because the vector of error term is normally distributed, the likelihood of beef cattle farmers Boz 231 falling one of the three categories of the dependent variable can be expressed as: expressed as: where  denotes the cumulative (2) where  denotes the cumulative standard normal distribution and 1 is greater than zero.Empirically, this model was similarly used by Boz et al. (2011), Budak et al. (2011), Boz and Akbay (2005), Chen et al. (2002), Abdel-Aty ( 2001), and McLean-Meyinsse (1997).

Measurement of variables
Income level was asked in the question that "If all farmers at your village were to be divided in three income categories as low, medium, and high income levels which category you would likely to fall in".
For the third objective 23 predetermined items were listed and asked by goat farmers to respresent each item in a five point Likertscale with 1 = Strongly Disagree (SD), 2 = Disagree (D), 3 = Neutral (N), 4 = Agree (A), 5 = Strongly Agree (A).Respondents were also asked to feel free to express any other problems left out of the questionnaire.Means, standard deviations and response categories wee also calculated.To interpret the means of all the items listed in the table an interpretative scale (0.00-1.49= SD, 1.50-2.49= D, 2.50-3.49= N, 3.50-4.49= A, 4.50-5.00= SA) was developed.
For the fourth objective of the study 30 predetermined innovations and best management practices were determined by an extensive use of the literature, explicitly Kaymakçi (2002), Kaymakçi and Dellal (2006) and Veteriner cc. ( 2012).Respondents were asked if they applied these practices in their own cases.
For the fifth objective, initially respondents were divided in three categories according to the number of applied innovations or best management practices.Of the 30 innovations and best management practices respondents who applied at least 10 items were assigned to the low level category and coded as '0'; those who applied between 11 and 20 items were assigned to the medium level category with a code of '1'; and finally those who applied more than 20 items were assigned to the high level category with a code of '2'.
Several models were run and the one presented was significant at an alpha level of 0.01 or better (Model chi square for socioeconomic factors = 28.313,Degrees of freedom = 7).The estimated threshold value (µ1 = 1.088) was positive and significant at 0.01 alpha level indicating that there was a natural ordering in the three levels of innovations and best management practices among goat farmers in the region.
Model chi square for information seeking behaviour = 38.637,Degrees of freedom = 7.The estimated threshold value (µ1 = 1.162) was positive and significant at 0.01 alpha level indicating that there was a natural ordering in the three levels of the dependent variable.
The independent variables entered the model as dummies with the following codes: Reading newspaper (At least once a week = 1, 0 otherwise), listening to radio (Every day = 1, 0 otherwise), Watching television (Every day = 1, 0 otherwise), Use of the Internet (Several times a month = 1, 0 otherwise), traveling to city (At least once a week = 1, 0 otherwise), contacts with extension service (At least once a month = 1, 0 otherwise), contacts with veterinarians (At least once a month = 1, 0 otherwise).

Socioeconomic characteristics of the respondents
Socioeconomic characteristics of the goat farmers were presented in Table 1.From the table it can be seen that 47.9% of the respondents were 35-50 age gap, 21.4% younger than 35 and 30.7% older than 50 years of age.The average age of respondents was 45.67.More than one-third of the respondents (36.4%) had less than 10 years' experience with goat farming while 34.3% had an experience between 10 and 20 years, and 29.3% more than 20 years.The mean of experience with goat farming was calculated as 18.68 years.In terms of education 63.6% held at least an elementary school degree while 20% received an education beyond elementary and 16.4% were illiterate.
Similarly, 39.3% of the respondents were in the medium income category while 30.7% and 30.0% in highy and low categories, respectively.The percentage of those who were members of cooperatives was 18.6% while 16.4% participated in village administration, 35.7% invested on their farms in the last three years buying live animals, land, and/or farm equipment.
The percentage of farmers owning farm land of smaller than 25 decares was 48.6% while the percentages of landless farmers was 27.1% and those who owned more than 25 decares made 24.3%.Average farmland of the respondents was 16.73 decares.Those who owned improved breeds of goats were 16.4% while those who owned native breeds were 87.9%.The average goat keeper in the region owned 16.4 improved breeds, and 64.7 native goats.

Selected management practices
The results on goat management practices are presented in Table 2. Based on the sales of milk and its products, it was established that 42.9% of the respondents sold teir milk daily while, 17.9% every 2-3 days and 12.1% weekly.Twenty-seven percent reported that they process goat milk into cheese or butter and sold it later.Majority of the farmers (51.4%) sold their milk to milkmen and 37.9% to traders.Similarly, majority of the respondents (76%) sold their animals to dealers and 51% sold goat hair to traders.
Almost half of the respondents (49.3%) carried out disease surveillance in their herds and monitored other livestock management problems while 27.1% practiced this duty yearly.Majority of the farmers (71.4%) thought that pasture lands around their village had been grazed properly.They stressed that lack of strict law and regulations cause early and excessive grazing which lowers the quality of pasture and degrades the fragile landscape leading to environmental problems such as erosion.Twenty-one percent reported that there was no publicly owned pasture land around their villages.Thus, they had to graze their herds in the pastures and bush areas owned by neighbor villages.
In addition cotton and grain harvested fields provide a good grazing opportunity especially around Kirikhan and Hassa districts of Hatay.However, it was qualitatively noted that respondents from these districts reported that large land owners in the area cannot stand any longer their harvested cotton and grain land to be grazed for free by goat farmers.Especially close to cotton harvest time they use some chemicals to drop the leaves of the crops due to a more convenient and economical harvest.Because many goat and sheep deaths have been reported in harvested cotton fields, herders abstain from grazing their animals in these fields.
Another restriction reported was that as long as farmers harvest their crops they burn residues to prepare soil for the upcoming crops leaving no grazing opportunity for herders.Although, there have been a strict regulation for residue fires, little success has been achieved so far.
Manure is considered very important fertilizer.More than half of the respondents (52%) sell their goat manure while 29.3% use it on their farm and 18.6% use it for heating their houses during the winter season.Majority of respondents (75.7%) used natural breeding methods and more than half of the respondents (57.1%) reared goats with after half-milking the mother.
Although, there have been improved breeding and rearing methods which enable goat farmers to increase their income, they seem not familiar with these methods.

Problems encountered by goat farmers
Problems encountered by goat farmers are presented in Table 3.According to the interpretative scale given in the analytical procedure and measurement of variables section of this study, respondents agreed with five items, remained neutral with ten items and disagreed with eight items.There were no items in SA and SD response categories.
The most seriously encountered problems in goat farming were expensive feeds, lack of governmental subsidies, lack of concentrated feeds, and lack of fodder crops, and low milk yield.From the farmers' point of view these can be considered as the immediate measures which will affect their cash income and therefore wellbeing.They offered simple solution for their problems which they focused on cheaper and abundant inputs, higher governmental subsidies, and higher prices for goat products.
It seems interesting that the respondents disagreed with the problems that lack of training and extension activities, care of the doe at kidding, rearing of baby goats, mating, animal insurance, keeping farm records, hygiene of barns, and trimming and keeping goat hair.
During the interviews, it was figured out that almost no training and extension programs regarding small ruminants had been implemented in the region.Also, no animal insurance was observed.In general they indicated that they are conventionally knowledgeable about the items with which they disagreed and therefore they do not see them as problems.It was observed that except the immediate income generating measures such as lower input prices, higher governmental subsidies and output prices; goat farmers paid less attention and saw little benefits in other management practices which must be considered crucial to increase competitiveness and keep it up with changing market conditions.

Respondents' application of management practices and innovations
Results of the management practices and innovation applied by the respondents are presentad in Table 4. Majority of the respondents (91.4%) indicated that they practice goat pen proper cleaning and disinfection.Similarly, 83.6% indicated supplementary feeding before and during mating and 82.6% practice vacination against foot and mouth diseases.More than half of the respondents took proper care of udders and nails, and trimmed goat hooves.They had hay feeders, separate pens, and feed storages in their shelters.In addition they vaccinated their animals against foot and mouth diseases, contagious caprine pleuropneumonia, goat prox, anthrax, and rabies (vaccine for dogs).Of the 30 selected innovations or best management practices only 3 were applied by more than half of the respondents while the remaining 16 items had quite lower application levels.Among the all respondents 65.7% had no information about 'deodorizing bucks', 62.9% about 'neutering males not planned on breeding', and 52.1% about 'tie-up the umbilical cord with dental floss'.On the other hand, 4.3% of the respondents were unaware of vaccine against foot mouth disease while 45% were unaware of vaccine against pseudo tuberculosis.

Socio-economic factors ınfluencing the adoption of ınnovations and management practices
The results obtained through ordered probit procedure are presented in Table 5.It can be seen from the results that 52.1% of the respondents were in low category, 32.1% in medium and 15.7% in high category of the innovations adoption.Of seven socioeconomic characteristics entered in the ordered probit model, 'experience' (p = 0.064) at 0.1 alpha level, and 'income' (p < 0.01) at 0.01 alpha level was found significant.Both variables had positive signs indicating that as experience and income level of farmers go up they tend to adopt more innovation and best management practices.However, adoption level was not affected by education, cooperative membership, investments, and farm size.The marginal effects for the significant socioeconomic variables showed that as farmers have more experience their likelihood of being in the low level adoption category decreases by 0.0057 percentage point while the likelihood of being in the medium and high level adoption categories increases by 0.0028 and 0.0029 percentage points, respectively.As farmers have higher income their likelihood of being in the low level adoption category

Information seeking behavior ınfluencing the adoption of ınnovations and management practices
The model for information-seeking behavior is presented in Table 6.Of the seven explanatory variables entered the model 'traveling to city' and 'contacts with extension service' were significant at 0.01 alpha level while 'contacts with veterinarians was significant at 0.05 alpha level.From this finding we can conclude that as goad farmers have more travels to cities, and have more contacts with extension service and veterinarians they tend adopt management practices and innovations.The marginal effects for the significant variables showed that as farmers have more travels to cities their likelihood of being in the low level adoption category decreases by 0.359 percentage point while the likelihood of being in the medium and high level adoption categories increases by 0.1204 and 0.2387 percentage points, respectively.As farmers have more contacts with extension service their a. Determine socioeconomic characteristics of goat farmers; b.Acquire basic information about goat farming in the region; c.Determine what problems are being encountered by goat farmers; d.Determine awareness and application levels of selected innovations and management practices e. Determine the extent to which socioeconomic factors and information-seeking behavior influencing the adoption of management practices and innovations.
size, N = Number of farmers in accessible population, Nh = Number of farmers in each stratum, Sh = Standard deviation within each stratum, D 2 = Desired variance, e = Permitted error from the mean of accessible population, t = t-table value of accepted confidence interval.

Table 1 .
Socioeconomic characteristics of goat farmers.

Table 2 .
Selected management practices regarding goat farming.

Table 3 .
Problems encountered with goat farming.

Table 4 .
Awareness and application level of selected innovations and management practices.

Table 5 .
Socioeconomic factors influencing the adoption of innovations and best management practices and their marginal effects.

Table 6 .
Information-seeking behavior influencing the adoption of innovations and best management practices and their marginal effect.