Technical and Economic Factors Affecting Losses in Sweet Cherry Production: A Case Study from Turkey

ABSTRACT

The aim of this study was to determine the factors affecting losses in sweet cherry production. The research was carried out in the Kemalpaşa district of Izmir province, which is one of Turkey’s most important sweet cherry production centers. The research data were obtained from face-to-face questionnaires with 102 farmers. Losses in production derive from poor practices in three distinct stages of activity on the farm: pre-harvesting, harvesting, and post-harvest processes. Ordinal Logistic Regression Analysis was used to determine the technical and economic factors affecting these losses. Results indicate that “experience of sweet cherry production,” “experience in agriculture in general,” “quantity of production” and “certified production” had a significant negative effect on sweet cherry losses. “Technical problems in production” and “reasons for sweet cherry crop losses” had a significant positive effect on sweet cherry losses. It is thought that promoting certified production methods such as Good Agricultural Practices (GAP) and organic farming, increasing the effectiveness of public agricultural extension services and agricultural organizations will have a positive effect in reducing rates of loss and improving quality.

KEYWORDS:

• Sweet cherry
• crop losses
• ordinal logistic regression
• Turkey

Introduction

Food loss can be defined as any reduction in the quantity or quality of food in the supply chain produced for human consumption (FAO, 2013). Losses can occur at any stage of the supply chain, for example: at the agricultural production stage, during post-harvest operations and storage, in processing, in distribution (logistics and storage), or at final consumption (Demirbaş, 2018; Demirbaş et al., 2017; Gustavsson et al., 2011). These losses can occur for different reasons. In developed countries, food losses are mostly seen at the retail and consumption stages (Permanandh, 2011; Prusky, 2011). In contrast, in developing countries 95% of food losses are unintentional losses incurred in the early stages of the food supply chain (FAO, 2018). Food losses in Turkey are generally found more in the agricultural production stages (Tatlıdil et al., 2013).

Fresh fruit and vegetable are an important sub-component of the agricultural sector, but these are often spoiled and discarded for a variety of reasons before they reach the consumer. The extent of these losses varies according to the type and variety of the fruits and vegetables. It is stated that between 4% and 12% of fresh fruit and vegetable losses are incurred on the farms themselves, 2 to 8% occurs in transportation to the market and to wholesale markets, and 1 to 5% is lost in the final consumption phase (Tatlıdil et al., 2013; Ünlü, 2015). The extent of these losses at different stages along the supply chain can rise to 50 or 60% depending on the particular products.

In many countries, the actual degree of food loss, and food waste, is difficult to establish and effective methods of measurement are still under discussion (Delgado et al., 2017). Though questions may be asked regarding the reliability of measurement methods, the level of fruit produce loss in Turkey is estimated to be up to 12.7% of total fruit production (Öztürk et al., 2012). In developed countries, on the other hand, studies indicate that these losses are around 5% (Kader, 2005). The main causes of losses in fruit production can be listed as: inefficient practices in the production process, deficiencies in combatting diseases and pests, inappropriate harvesting techniques and lack of knowledge of the necessary applications (Batziakas et al., 2020; Dijksma, 2015; FAO, 2018; Food Drink Europe, 2013; Keding et al., 2013; Rallo et al., 2018; T. R. Ministry of Foreign Affairs, 2018; Tarabay et al., 2018). For example, low water levels and insufficient fertilizer management can result in poor product quality and high losses. Adverse weather conditions such as heavy rainfall can also seriously affect the health of trees (Batziakas et al., 2020; Thompson, 2007). In addition, sometimes fruits are left on the tree because they do not meet the quality standards (for shape, size, or weight) set by processors, retailers, or the target market (Stuart, 2009). For example, in 2009 in Italy, 17.7 million tons of agricultural crops were left in fields and gardens, which equates to 3.25% of total production. In the USA, it is estimated that 7% of cultivated areas are not harvested every year (Segre and Falasconi, 2011). The factors which affect the loss of fresh fruit can be summarized as: insufficient knowledge of the market and of product pricing, lack of training and R&D resources, low processing capacity, lack of access to investment and credit, deficiencies in quality control standards, the structure and type of farmers’ organizations, the extended length of the supply chain and the lack of specialist support services for farmers (Kumar et al., 2015).

The sweet cherry (Prunus avium L.) is a type of stone fruit that has substantial sales potential throughout the world due to its suitability for cultivation under various different ecological conditions. The sweet cherry has a shorter shelf life than soft seed fruits and is more susceptible to physiological and pathological disorders when stored and also throughout its shelf life (Çetin, 2010; Hasdemir, 2011). These disorders give rise to post-harvest quality losses. Cracking, pitting and wrinkling and darkening of stalk color are the main physiological disorders found in cherries (Mitcham et al., 2006). Darkening of the stalk color is seen as a serious negative factor for farmers in terms of the marketing of the product (Schick and Toivonen, 2000).

A major issue for sweet cherry growers is that regular yields cannot be obtained from the trees every year. Fruit set, flowering, and fertilization are all significant problems. Climatic conditions can also contribute to low efficiency (Hussein et al., 2018, 2020; Lang, 2019). Warm winter months cause the timing of flowering to be delayed, the length of the flowering period to be prolonged and the flowering itself to be irregular in sweet cherry trees which have not had a full winter’s rest. Some flower buds do not open while others are shed without opening (Engin and Akçal, 2013; Lang, 2019). Furthermore, bad weather may cause the pollination process to fail or cause discoloration due to nutrient deficiency (Creamer and Johnson, 2018).

Harvest time for cherries, like other fruit types, is extremely important with regard to storage and marketing (Botina et al., 2019; Crisosto et al., 2003; Kalajdzic et al., 2019). Marketing strategies too can have an influence on the level of losses. Lack of stability in sales after the harvest and price fluctuations can lead to a drop in production levels (Adanacıoğlu, 2016).

The aim of this study was to determine the factors affecting losses in sweet cherry production. The research was carried out in Kemalpaşa district, an important sweet cherry-producing region in Turkey. According to the 2016/2018 period average, Izmir’s share of Turkey’s total sweet cherry production was about 10% and the share of Kemalpaşa district in İzmir’s overall sweet cherry production was approximately 64% (TURKSTAT 2019a). From available statistics for sweet cherry production in Turkey, the average of the level of losses over the period 2012 to 2016 was around 5% (TURKSTAT 2019b).

Materials and Methods

Methods of Sampling

The research population for this study consisted of all sweet cherry farmers registered in the Farmer Registration System (FRS) for the Kemalpaşa district of İzmir province. In the preliminary study, the total number of sweet cherry farmers registered to FRS for 2019 was 2190 (Ministry of Agriculture and Forestry, 2019). The sample size was determined using the proportional sampling (Newbold, 1995).

In the formula;

n = Sample size

N = Total number of producers (2190)

p = Proportion of producers who have lost production

σpx² = Variance of the ratio

In order to achieve the maximum sample size, the percentage of farmers who lost produce was taken as p: 0.50 and (1-p): 0.50. From this, the sample size was found to be 102 with a 95% confidence interval and a 9.5% margin of error.

The research area consisted of Merkez, Ören, Bağyurdu, Yiğitler, and Armutlu villages where sweet cherry production is highly concentrated. In selecting the villages, information was obtained from Kemalpaşa Agriculture and Forestry District Directorate and the Chamber of Agriculture. The number of farmers to be interviewed in each selected village was found by taking into consideration the contribution of each village to the total number of farmers. The number in the research area was 50.41% of the total number of sweet cherry farmers in Kemalpaşa. The number of questionnaires per village was converted to integers (Table 1).

Table 1. Number of surveys conducted by villages (2019)

Villages	Number of Cherry Farmers	Proportional Contribution of Villages (%)	Number of Farmers Surveyed
Bağyurdu	294	27	28
Ören	265	24	25
Armutlu	251	23	23
Merkez	170	15	15
Yiğitler	124	11	11
Total	1.104	100	102

Methods of Data Analysis

In this study, an Ordinal Logistic Regression model was used to identify the factors (variables) which affect losses in sweet cherry production. The dependent variable of the research consisted of ordered structures. Ordinal Logistic Regression is a logistic regression method applied in cases where the dependent variable has more than two categories and is on an ordinal scale (Çokluk, 2010; Emeç, 2002; Gumirakiza and Daniels, 2016; Kassa, 2016; Tanko, 2016).

The Ordinal logistic regression model is generally shown as follows:

$link\left(y_j\right)={\tau }_j-{\sum }^{\beta }{ }_k^{\prime }{x}_k$

(2) $ln\left(Y_j\right)=ln\frac{P\left(Y\le j/X_1,X_2,\dots X_i\right)}{1-P\left(Y\le j/X_1,X_2,\dots X_i\right)}=a_j-\left({\beta }_1{X}_1+{\beta }_2{X}_2+\dots +{\beta }_i{X}_i\right)=a_j-\sum _{i=1}^n{\beta }_i{X}_i$(2)

For an ordinal variable with K categories, K − 1 cumulative logit functions are defined. Each cumulative logit function includes a unique intercept or ‘cut point,’ β0(k), but all share a common set of regression parameters for the p predictors, β = (β1, …,βp). Consequently, a cumulative logit model for an ordinal response variable with K categories and j = 1, …, p predictors requires the estimation of (K − 1) + p parameters. The concept of a ‘cumulative logit’ is certainly more complex than that of a baseline category logit which is employed in the parameterization of simple logistic regression or even multinomial logit regression models. In reality though, this is simply an alternative way of parameterizing the model in order to estimate the probability that a response will fall in ordinal category y = 1, …, K . (Albayrak, 2006; Çokluk, 2010).

The dependent variable used in the model was the loss levels that occur during sweet cherry production (sweet cherry production = pre-harvesting + harvesting +pre-sale processing on the farms) (Demirbaş, 2019). The frequency distribution of the loss rates was used to divide them into sub-groups (Table 2). Sixty nine farmers were identified in the first group (≤5%), five in the second group (6–15%) and five farmers in the third group (16–20%).

Table 2. Groups categorized by crop loss rate (%)

Loss Groups	Loss Rate Limits (%)	Number	%
Group 1	≤ 5	69	67.60
Group 2	6–15	5	4.90
Group 3	16–20	5	4.90
Total		79	100.00

The Results

General Characteristics of the Farms and Farmers

The largest proportion of the farms examined (43.13%) were those with an area of 1.1–2.9 hectares (Table 3). The average area of the farms was 2.7 hectares and the number of parcels was 5.21 on average.

Table 3. Cherry land size groups in the studied farms (ha)

Land size groups (ha)	Number	%
≤1.0	28	27.45
1.1–2.9	44	43.13
3.0–4.9	17	16.67
≥5.0	13	12.75
Total	102	100.00

The average age of the farmers was 49.45 years, and their average time spent in full-time education was 8.45 years. The average of their experience in agriculture was 20.63 years, and their experience in sweet cherry production averaged 18.92 years. The farmers with a farm size of 5.00 hectares or more had the highest level of previous training, the longest experience in the agricultural sector and also the longest experience in sweet cherry production (Table 4).

Table 4. General characteristics of the farmers in the studied farms

	Variables (Years)
Farm Size (ha)	Age	Education	The experience in the agricultural sector	The experience in cherry production
Group 1: ≤ 1.0	48.85	8.85	16.89	16.10
Group 2: 1.1–2.9	48.50	8.50	20.93	18.97
Group 3: 3.0–4.9	51.44	7.38	22.38	20.88
Group 4: ≥5.0	51.33	8.91	25.66	22.33
General	49.45	8.45	20.63	18.92

Farmers are generally registered with more than one agricultural institution or organization. Many farmers are members of the Chamber of Agriculture (36.51%). This is followed by membership of the Agricultural Sales Cooperative (23.81%), the Agricultural Credit Cooperative (20.11%), the Irrigation Cooperative (15.87%) and the Agricultural Development Cooperative (3.70%).

Rates of Sweet Cherry Loss

For each of the farms surveyed, the farmers were asked firstly if they had incurred any losses in sweet cherry production. It was found that more than three-quarters of farmers (77.50%) experienced loss in production, though at varying rates. The number who reported no losses in sweet cherry production at all was approximately 23%. The others were then asked at what rate the losses were incurred and at what phase of the production process. The responses were grouped according to rate of loss. About 87% of the farmers in the first group (≤5%) were found to have experienced losses mostly during pre-harvesting and harvesting. In general, farmers were found to experience the greatest losses (63.25%) during pre-harvesting (Table 5).

Table 5. Loss rates by phase of production

Loss Groups (%)	Pre-Harvesting		Harvesting		Transport and Storage (on the farm)		Processing and Packaging (on the farm)
	Number	%	Number	%	Number	%	Number	%
Group 1: ≤5	64	86.49	28	84.85	2	40.00	3	60.00
Group 2: 6–15	4	5.40	2	6.06	2	40.00	1	20.00
Group 3: 16–20	5	6.75	3	9.09	1	20.00	1	20.00
Cherry Loss Farms (General)	73	63.25	33	28.21	5	4.27	5	4.27

Reasons for Crop Losses

On the farms included in this study, farmers stated seasonal factors (43.10%) as the most significant reason for losses in sweet cherry production. This was followed by diseases and pests (14.70%), workers errors (9.80%), inability to determine appropriate rootstocks (4.90%), inability to evaluate the amount of spilled product (2.90%) and lack of cold storage (2.00%). The proportion of farmers who were not able to specify the cause of the loss was found to be 22.50%, which in itself was a significant finding of the research (Table 6).

Table 6. Reasons for crop losses

Reasons	Number	%
Seasonal factors	44	43.10
Farmers who were not able to specify the cause of the loss	23	22.50
Diseases and pests	15	14.70
Workers errors	10	9.80
Inability to determine appropriate rootstocks	5	4.90
Inability to evaluate the amount of spilled product	3	2.90
Lack of cold storage	2	2.00
Total	102	100.00

Model Results for Factors Affecting Losses in Sweet Cherry Production

The main goodness of fit tests used in the ordinal logistic regression model are the Pearson Chi-square Test, deviation statistic, the Likelihood Ratio Test, and the Hosmer & Lemeshow tests. The value (p) approaching zero indicates an incompliance (Çelik, 2019). In this study, deviation and Chi-square values were used to measure the goodness of fit of the model. The probability of the test statistics shows that goodness of fit was obtained (Table 7).

Table 7. Goodness of fit test results

	Chi-square	Degree of Freedom	Probability (p)
Pearson	109.752	37	0.541
Deviation	58.612	37	1.000

There are a total of nine independent variables in this model (Table 8). The parameter interpretation of Ordinal Logistic Regression Analysis is different from, and more complex than, Binary and Multinomial Logistic Regression Analysis. Here, the “exp” of these values is taken in order to interpret the estimated parameter values. In this way, the values obtained take their final form in order to make interpretations (Albayrak, 2006; Çokluk, 2010). Reference categories are also determined in order to make comment. In this model, the reference categories of the variables were the last categories in Table 9.

Table 8. Variables used in the model

Variables	Definition of Variables
Dependent Variable	Loss rates
	Y = 1, (≤5%)
	Y = 2, (6 − 15%)
	Y = 3, (16 − 20%)
Independent Variables
Exp. in cherry production	Experience in cherry production of farmers (1 = if farmer had 1–19 yr of exp., 0 = otherwise) (dummy)
Age	Age of farmer (1 = if age of farmer is≤39, 0 = otherwise) (dummy)
Education	Farmers’ level of education (1 = if farmer had 1–5 yr of education, 0 = otherwise) (dummy)
Agricultural experience	Exp. in agricultural production of farmers (1 = if farmer had 1–14 yr of exp., 0 = otherwise) (dummy)
Garden size	Garden size in hectares (1 = the garden size is ≤1.0 ha, 0 = otherwise) (dummy)
Cherry production volume	Cherry production (ton/year) (continuous)
Certified production	Certificate ownership in cherry production like GAP/Organic farming, 1 = Yes 2 = No (categorical)
Technical problems	Technical prob. in cherry production (1 = Fruit retention) (categorical)
Reasons for losses	Reasons for losses in cherry production (1 = Seasonal factors) (categorical)

Table 9. Ordinal logit regression model estimates evaluating factors affecting losses in sweet cherry production

	Variables	Estimate (β)	SE	Wald	Odds ratio-exp (β)	Sign.	%95 confidence interval
The dependent variable	Loss rate [≤5]	−6.002	0.224	3.166	0.002	0.025	(15.9:12.9)
	Loss rate [6 − 15]	−4.085	0.446	6.327	0.000	0.030	(−7.7:21.7)
	Loss rate [16–20]	1.492	0.200	9.006	4.446	0.025	(−6.4:23.1)
The independent variables	Exp. in cherry production	−1.200	0.245	3.558	0.301	0.010*	(−0.2:0.6)
	Age = 1	−0.405	1.900	1.557	0.667	0.730	(−5.6:1.9)
	Age = 2	−1.609	1.423	1.040	0.200	0.229	(−4.1:2.7)
	Age = 3	−1.050	0.630	1.500	0.350	0.220	(−0.7:6.7)
	Education = 1	1.400	1.123	0.218	4.055	0.423	-
	Education = 2	−15.002	0.020	-	0.000	-	-
	Education = 3	-	-	-	-	-	-
	Agricultural experience = 1	−1.002	0.504	4.500	0.367	0.010*	(−5.4:9.4)
	Agricultural experience = 2	−1.620	0.021	4.223	0.198	0.003*	(−2.5:6.9)
	Agricultural experience = 3	-	-	-	-	-	-
	Garden size = 1	−1.200	0.230	1.004	0.301	0.625	(−2.0:4.0)
	Garden size = 2	−1.324	0.310	1.030	0.266	0.564	(−4.9:2.2)
	Garden size = 3	-	-	-	-	-	-
	Cherry production volume = 1	−2.004	0.230	3.041	0.135	0.032*	(−4.8:1.6)
	Cherry production volume = 2	−2.034	0.310	3.010	0.131	0.024*	(−6.1:1.2)
	Cherry production volume = 3	-	-	-	-	-	-
	Certified production = 1	−1.325	0.204	4.034	0.266	0.015*	(−0.8:4.4)
	Certified production = 2	-	-	-	-	-	-
	Technical problems = 1	3.205	1.425	2.472	24.656	0.328	(−3.0:9.7)
	Technical problems = 2	3.225	0.100	4.646	25.154	0.025*	(−1.6:12.7)
	Technical problems = 3	1.600	1.207	2.578	4.953	0.226	(−5.7:9.3)
	Technical problems = 4	2.100	1.002	2.626	8.166	0.330	(−2.1:11.3)
	Technical problems = 5	1.995	1.400	2.424	7.352	0.407	(−3.0:14.1)
	Technical problems = 6	1.030	1.232	2.678	2.801	0.325	(−2.1:14.3)
	Technical problems = 7	1.050	1.321	2.521	2.858	0.703	(−5.0:9.0)
	Technical problems = 8	4.625	1.971	2.332	-	0.380	(−12.6:15.6)
	Technical problems = 9	1.978	1.913	2.330	7.228	0.003*	(0.3:20.1)
	Technical problems = 10	4.221	0.740	4.223	68.102	0.332	(−14.0:14.1)
	Technical problems = 11	0.228	1.102	2.378	1.256	0.772	(−6.7:6.6)
	Technical problems = 12	10.225	1.775	2.495	-	0.500	(−20.4:55.1)
	Technical problems = 13	8.420	1.897	2.921	-	0.323	(−26.0:52.8)
	Technical problems = 14	-	-	-	-	-	-
	Reasons for losses = 1	8.724	0.225	8.210	6.149	0.028*	(−14.2:0.5)
	Reasons for losses = 2	14.125	0.300	7.323	13.627	0.020*	(−6.3:2.4)
	Reasons for losses = 3	3.227	1.227	2.217	25.204	0.815	(−12.5:4.3)
	Reasons for losses = 4	9.041	1.425	2.925	-	0.313	(−41.6:41.6)
	Reasons for losses = 5	-	-	-	-	-	-

*significance at the 5%.

The results of the Wald statistics, which are widely used to test the significance of logistic regression coefficients, were used for each independent variable. According to Çokluk (2010), Wald statistics correspond to the significance test of the coefficient β in a non-standardized logistic regression. In order to interpret the coefficient negative values, it is necessary to correct the Odds ratio to 1/Odds ratio (Adanacıoğlu, 2016). This rule was followed in the interpretation of negative coefficients in the model. In addition, a 95% confidence interval was taken into account in the selection of variables. The confidence interval is a kind of interval estimation for a population parameter. It is an inferential statistics solution tool. Since a population parameter value can be estimated by a single number, there is a range of two (lower and upper limit) numbers that can cover this parameter value. Thus, confidence intervals show how reliable a prediction is (Tavşancıl, 2014).

The six explanatory variables included in the model (farmers’ experience in cherry production, agricultural experience, cherry production volume, certified production, technical problems in production, and reasons for cherry losses) were all statistically significant factors that affect the loss rate groups. Therefore, six variables were interpreted for the significant categories, and other variables which were not found to be significant were not interpreted (Table 9).

Farmers’ Experience in Cherry Production

The effect of experience in cherry production on the dependent variable was significant and negative. As experience in cherry production increases by one unit, cherry losses decrease by a factor of 3.32 (1/0.301).

Agricultural Experience

This variable had a significant negative impact on the dependent variable. The reference group of the variable was 25–30 years of agricultural experience. It was concluded that farmers who had 15–25 years of agricultural experience incurred 2.72 times less loss than those who had 25–30 years of experience. Also, farmers who had 14 years and less of agricultural experience were found to have incurred 5.05 times fewer losses than the farmers with 25–30 years of experience.

Sweet Cherry Production Volume

The groups of production volume had a significant negative effect on the dependent variable. The reference group of the variable was the group that produces 30 tons/ha or more. It was concluded that farmers who produce between 21 and 29 tons/ha of cherries have 7.40 times less losses than farmers who produce 30 tons/ha of cherries or more. Farmers whose sweet cherry production was 20 tons/ha and less, lose 7.63 times less cherries than the reference group. Therefore, there was a positive correlation between the volume of production and the amount of the losses incurred, and the amount of the loss decreases as production decreases. This may be due to the fact that farmers with fewer trees tend to carry out processes such as irrigation, spraying, pruning, and fruit picking more meticulously than those with a bigger number of trees. However, the extent of sweet cherry losses was a serious concern for farmers who produce smaller volumes.

Certified Production

This variable had a significant negative impact on the dependent variable. The reference group of the variable was non-certified production. Here, it was concluded that the certified farmers experience losses 3.76 times lower than the non-certified farmers.

Technical Problems in Production

Technical problems had a significant positive effect on the dependent variables. The reference group of the variable was farmers who do not have technical problems in sweet cherry production. According to the results of the model, it was revealed that the farmers who had fruit retention issues experienced 25.15 times more losses than the farmers who did not have this problem. In addition, it was determined that the farmers who had problems with disease and pests lost 7.22 times more of their crop than the reference group.

Reasons for Sweet Cherry Crop Losses

This variable had a significant positive effect on the dependent variables. The reference group of the variable was farmers who did not use cold storage. Farmers who experienced crop loss due to diseases and pests incurred 6.14 times more losses than the reference group. Similarly, farmers who experience sweet cherry loss due to seasonal factors were found to have 13.62 times more losses than the reference group.

Discussion

It was found that the highest loss of sweet cherries on farms occurred during the pre-harvesting period. The key to the prevention of product loss from diseases and pests is the effective application of the correct methods at the right time (Hussein et al., 2018). Also, in order to reduce product losses in the pre-harvesting period, farmers’ transition to certified production methods such as GAP should be strongly encouraged. The application of techniques such as GAP, Integrated Combat with Pests and Integrated Product Growing, to the production process is very likely to produce significant benefits (Demirbaş, 2019; FAO, 2012; Gül et al., 2016; Sinha et al., 2012; Tarabay et al., 2018). The results of the analysis show that the farmers who engaged in certified production experienced lower levels of crop loss. Agricultural practice control procedures involving registration, traceability, regular inspections and certification can all be effective in dealing with the causes of loss (Güneyli and Onursal, 2014; T.R.Ministry of Foreign Affairs, 2018). Compulsory training processes, which are an integral part of GAP, also have a positive effect in reducing losses (SKD, 2018).

The findings indicate that losses caused by diseases, pests, and seasonal factors (Hussein et al., 2018, 2020; Musacchi and Serra, 2018) are greater than those due to the lack of cold storage. This finding is also supported by current literature on the subject (Bayraktar, 2015) The İzmir and Kemalpaşa Directorates of Agriculture and Forestry can contribute to the awareness of farmers by organizing seminars or meetings to address technical issues which affect sweet cherry production and by providing practical training in all aspects of production (Başkaya, 2011). In addition, the installation of a mesh system in the gardens may provide a solution for farmers who experience quality and product losses due to seasonal factors (Creamer and Johnson, 2018; Doğan, 2016).

The critical marketing issue which emerged from the study was the lack of influence of the farmers themselves on the sales price of the product. A key reason for this is that the level of organization of farmers in the research area is not effective. Boosting production, producing a quality product, and achieving marketing success depends on effective organization (Terin and Ateş, 2010). It is evident that the farmers included in this research study have high expectations that farmers’ organizations will produce solutions to their current production and marketing problems. This same issue has been highlighted in other studies of sweet cherry marketing in the same region (Adanacıoğlu, 2016; Bayraktar, 2015; Doğan, 2016; Gül et al., 2020). Furthermore, it is anticipated that the re-activation of the Cherry Producers’ Union, which was long-established in the region but was subsequently closed for various reasons, combined with the efforts of the newly formed Development Cooperative to invest in cold storage and processing facilities, may have a significant effect in reducing the level of losses.

Conclusion

In this study, the factors affecting losses in sweet cherry production were determined using an Ordinal Logistic Regression Model. The results obtained are discussed within the framework of technical and economic problems affecting losses, after which recommendations are made.

In fact, it has been determined that the farmers are not sufficiently well informed or educated in the best production and harvesting methods within this area of research. Problems and deficiencies in relation to the organization of the farmers also negatively affect the technical and economic aspects of production. Promoting certified production methods and increasing the effectiveness of agricultural extension services and agricultural organizations will affect in reducing rates of loss and improving quality.

Declaration Of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Authorship

This work contains a section from the first author’s completed master thesis. I am supervisor of her. We worked together with M.K. in this study which I am responsible author. The questionnaire form was designed and analyzed together. Fieldwork was done by M.K. The Turkish language of this study was written together and translated by N.D. into English. It was send to a native speaker for proofreading by N.D., who can offer the paper showing that this work is edited by a native English speaker, if requested.

Ethics Of Human Subject Participation

This study was conducted according to guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Ege University.

Acknowledgments

This article was derived from the first author’s Master Thesis. The authors wish to thank the cherry farmers who contributed to the farm survey.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.