Breeding practices and trait preferences of smallholder farmers for indigenous sheep in the northwest highlands of Ethiopia: Inputs to design a breeding program.

The aim of this study was to identify breeding practices and trait preferences for indigenous sheep in three districts (Estie, Farta and Lay Gayient) located in the northwest highlands of Ethiopia. Questionnaire survey and choice experiment methods were used to collect data from 370 smallholder farmers. Respondents were selected randomly among smallholder farmers who own sheep in the aforementioned districts. A generalized multinomial logit model was employed to examine preferences for sheep attributes, while descriptive statistics and index values were computed to describe sheep breeding practices. Having the highest index value of 0.36, income generation was ranked as the primary reason for keeping sheep, followed by meat and manure sources. The average flock size per smallholder farmer was 10.21 sheep. The majority of the smallholder farmers (91%) have the experience of selecting breeding rams and ewes within their own flock using diverse criteria. Given the highest index value of 0.34, body size was ranked as a primary ram and ewe selection criteria, followed by coat color. Furthermore, choice modeling results revealed that tail type, body size, coat color, growth rate, horn and ear size have shown significant influences on smallholder farmers' preference for breeding rams (P<0.01). The part-worth utility coefficients were positive for all ram attributes except ear size. For breeding ewes, mothering ability, coat color, body size, lambing interval, growth rate, tail type and litter size have shown significant effects on choice preferences of smallholder farmers (P<0.05). Moreover, significant scale heterogeneity was observed among respondents for ewe attributes (P<0.001). Overall, the results implied that sheep breeding objectives suitable for the northwest highlands of the country can be derived from traits such as linear body measurement, weight and survival at different ages, and lambing intervals. However, selection decisions at the smallholder level should not only be based on estimated breeding values of traits included in the breeding objective but instead, incorporate ways to address farmers' preference for qualitative traits.

Introduction

Ethiopia is one of the few African countries endowed with huge sheep genetic resources. The estimated population size is about 31.3 million sheep, of which 99.8% are indigenous types [1]. Ecologically, sheep are found in diverse production environments that range from arid lowlands to extremely cool highlands. The northwest highlands are among the major sheep production areas of the country, where Farta sheep breed and their crosses with other indigenous sheep are widely distributed [2, 3]. For the smallholder farmers, sheep provide valuable contributions through income generation, direct food sources, non-food utilities and various sociocultural privileges [3, 4, 5]. Particularly when crop farming is less reliable due to drought or other factors, sheep are commonly used to mitigate adverse effects, for instance, related to food shortage at the smallholder level.

Despite the presence of a large number of sheep and their diverse functions, the average productivity of indigenous sheep is generally low. For example, Mekuriaw et al. [6] reported an average yearling weight of about 20 kg for Farta sheep, while Gulilat et al. [7] obtained an average carcass yield of 10 kg for the same breed. The causes for low performances of indigenous sheep are known to be multilaterally, but largely related to the lack of effective breeding programs. For decades, crossbreeding between exotic sheep (such as Awassi from Israel and Dorper from South Africa) and some indigenous sheep breeds have been performed [8, 9]. However, achievements are far below expectations due to the lack of effective crossbreeding strategies and poor adaptability of crossbreeds that have high exotic blood level [8]. Recently, community-based sheep breeding programs have been designed and implemented for Menz, Afar, Bonga and Horro indigenous sheep breeds [10, 11]. Nevertheless, the majority of the indigenous sheep breeds in different parts of the country are still managed in the traditional breeding system, without being supported by proven scientific methodologies and the state of art technologies in animal breeding.

Given the suitability of the area and adaptive potentials of the existing sheep, improving sheep productivity can be a pathway to put smallholders out of poverty in the northwest highlands of Ethiopia. However, to design and implement effective breeding programs, breeding objective traits specific to sheep breeds reared in the target areas have not yet been identified. In other words, a thorough analysis of sheep breeding experiences and trait preferences is required to sensibly define breeding objectives and design genetic improvement programs at smallholder level [4, 5].

Various methodological approaches have been used to identify breeding objective traits in sheep in Ethiopia, for instance choice experiments [12], ranking among a list of traits [13, 14], and live animal rankings [10, 15]. Recently, a choice experiment has been widely applied to investigate farmers’ preferences for animal traits. It provides a hypothetical depiction of attribute levels, giving adequate options for the respondents to reflect their interests. Although it requires higher cognitive efforts, a choice experiment is useful to identify preferences when the number of sheep per household is very small to conduct a live animal ranking method [16]. Overall, applying a combination of methods has been suggested to effectively identify breeding objective traits [17]. As such, this study aimed at identifying breeding practices and smallholder farmers’ trait preferences for indigenous sheep in the northwest highlands of Ethiopia, using a ranking of traits and choice experiment methods.

Materials and methods

Ethics statement

Prior to the study, data collection formats and procedures were reviewed and approved by the research ethics review committee of Debre Tabor University, Ethiopia (number DTU13/19). Respondents also provided their verbal informed consent to take part in this study. Furthermore, the data were analyzed anonymously and names, ethnicity and religious issues were not asked and recorded during data collection.

Description of the study areas

The study was conducted in Estie, Farta and Lay Gayient districts of South Gondar Zone of the Amhara Region, located in the northwest highlands of Ethiopia. These districts were selected because they have huge potentials for sheep production, compared to other eight districts of the zone, for instance, in terms of the availability of communal grazing land and large sheep population. Furthermore, a preliminary assessment prior to the present study has identified the three districts as niche areas for Farta sheep breed, in which designing and implementing a breeding program is under consideration. The agricultural practice in these areas is crop-livestock mixed farming, where livestock play invaluable roles for crop cultivation and the livelihood of smallholder farmers [3].

The detailed comparisons in terms of area coverage, human and livestock populations of the three districts with that of the country, Amhara Region and South Gondar Zone are displayed in Table 1. Concerning climate variables, the average annual rainfall for Estie, Farta and Lay Gayient districts are 1591, 1122 and 1200 mm, respectively. Estie and Lay Gayient districts have received similar minimum (9°C) and maximum (22°C) average daily temperatures, whereas the minimum and maximum average daily temperatures of Farta districts are 8°C and 18°C, respectively.

Table 1

Description of the study districts in comparison to the national, regional and zonal level.

	Ethiopia c	Amhara Region c	South Gondar Zone c	Study districts d
				Farta	Lay Gayient	Estie
Area a	426,400	59,733.46	5442.18	424.42	587.81	527.1
Human b	94,351,001	21,134,988	2,484,929	272,177	251,926	251,708
Cattle	60,392,019	16,148,390	1,808,185	213,188	120,579	190,853
Sheep	31,302,357	11,086,083	1,085,652	113,978	88,836	191,985
Goat	32,738,385	7,766,661	514,746	51,556	48,758	104,604

a area coverage is in square miles (Source: [18])

b human population projection of Ethiopia for 2017 (Source: [19])

c cattle, sheep and goat populations of Ethiopia, Amhara Region and South Gondar (Source: [1])

d cattle, sheep and goat populations of the three districts (Source: South Gondar Zone Livestock Department annual report for 2017, Unpublished)

Sample areas and focus-group discussion

Secondary data about livestock population and distribution and availability of infrastructures were obtained from the Agriculture Office of the respected districts. Based on sheep population size and accessibility for transport services, three kebeles (the lowest formal administrative units in the district) were selected from each district. For focus-group discussion, a list of smallholder farmers who are regarded as knowledgeable in sheep breeding was identified with the help of livestock development agents and administrative staffs of the kebele. Then, three participants per kebele were randomly selected from the list. For each district, a separate focus-group discussion was held by involving nine well-experienced farmers, a livestock expert and researcher, with the latter two played facilitation roles. During the discussion, a list of ram and ewe traits was identified taking into account their socio-cultural and economic importance, both at the local and national levels. To prioritize traits, each participant farmer was asked to rank the traits based on their indigenous knowledge. Finally, the ranks from the three districts were combined and analyzed to identify the most important traits of rams and ewes in which choice experiments were designed.

Study approach

The study followed two approaches to acquire the data. First, data about socioeconomic status and sheep breeding experience of the respondents were collected through in-person interviews, using a semi-structured questionnaire. The questionnaire was prepared following Haile et al. [16], who suggested the types of information that need to be collected regarding sheep breeding practices and breeding objectives for the purpose of designing a community-based breeding program. The questionnaire was translated into the local language (Amharic). It includes sheep breeding practices such as production objectives, flock size and composition, selection and culling criteria, ram use and mating system. The list of smallholder farmers who own sheep and dwell permanently in the sampled kebeles was obtained from the local authorities of the respected kebeles of each district. From the list, respondents were selected using a simple random sampling method. For the respondents, willingness and having sheep were the criteria required to involve in the study.

Second, sheep trait preferences were collected by means of a discrete choice experiment, where the respondents have chosen their preferred alternative from choice sets built through hypothetical trait levels combinations. Valuing of non-marketable traits is the typical usefulness of the choice experiment method in animal breeding. Furthermore, in low input production systems, where smallholders’ literacy level is low and performance recording is virtually absent, trait preference could be better elucidated using choice experiment method [10, 20]. However, when a choice is made among too many tasks, a choice experiment may not reveal true preferences due to biases allied with choice complexity [21]. Thus, the number of attributes and levels should be manageable in size to minimize the complexity of the choice experiment design.

Choice experiment design

Identifying attributes and their levels are the principal steps to design choice experiments [22]. In this study, sheep traits were identified and prioritized with focus-group discussions as described earlier. Such an approach has been implemented by Amadou et al. [23] and Siddo et al. [24] in sheep and cattle, respectively. Based on the results of the focus-group discussion, six traits for rams and seven traits for ewes were selected. Attributes with their descriptions and levels are shown in Table 2. Levels were effect-coded to minimize confounding between parameter estimates [22]. A JMP software version 14 [25] was used to construct a full factorial design containing 64 and 128 profiles for rams and ewes, respectively. However, making a choice among the full factorial could be cognitively complex for the smallholder farmers. To reduce the size of the design, a fractional factorial with resolution IV design was applied [26]. Subsequently, an orthogonal array with 16 profiles grouped into four blocks was generated. Furthermore, to prevent forced choices, an opt-out option was assigned in each block, making the number of choice sets five per block.

Table 2

Ram and ewe attributes and levels included in the choice experiment.

Attributes	Attribute descriptions	Levels with effect-coded
Body size	the physical appearances including the height and body length of rams and ewes	Large = 1, small = -1
Coat color	the type of color predominantly observed on the body of rams and ewes	brown = 1, white = -1
Growth	yearling live weight in which rams and ewes reach at breeding age	Rapid = yearling weight 30 kg = 1, slow = yearling weight 20 kg = -1
Lambing intervals *	the average lambing interval between two successive lambing of ewes	short = 3 lambing in 2 years = 1, long = 1 lambing per year = -1
Mothering ability *	the ability of ewes to nourish their lambs that could also be implicated on lamb growth and survival	Good = 1, poor = -1
Litter size *	the number of lambs born per ewe per lambing	twin = 1, single = -1
Ear size $	the size of the ears of rams	Large = 1, small = -1
Horn$	the presence and absence of horn in rams	Horned = 1, polled = -1
Tail type	the length and width of the tail in rams and ewes	length covering the testicular area with sufficient width for rams and extended halfway to the udder for ewes = good = 1, small and thin = bad = -1

*appeared only in ewes,

$ appeared only in rams, unmarked attributes appeared in both ewes and rams

Choice cards with picture representations of the different profiles of rams and ewes were prepared. Enumerators with a Bachelor’s Degree in Animal Science were selected and trained to collect data under the close supervision of the researchers. Prior to the collection of the actual data, the questionnaire and choice cards were pretested in the study sites. Based on the feedback, modifications were made on the questionnaire and illustrations of the profiles. A sample of final choice cards used for ewe attributes is shown in Fig 1.

Fig 1

Sample choice card for ewe attribute levels.

Sample size

To determine the minimum number of respondents required for the choice experiment, we used the method suggested by Johnson and Orme [27] and Orme [28] as: Where N is the sample size, c is the highest number of levels of any attribute for main effect design, t represents the number of choice tasks, a is the size of choice set per task.

In this study, the number of choice tasks per block was one with four alternatives, excluding the opt-out option and a maximum of two levels per attribute. Plugging in the information to the formula resulted in a minimum sample size of 250. However, we collected the data from a total of 385 respondents randomly selected among smallholder farmers who own sheep.

Data collection

Procedurally, the respondents were asked first about their socioeconomic characteristics and experiences about sheep breeding in a face-to-face interview. Subsequently, choice cards were introduced. Following a brief explanation about the choice cards, respondents were asked to choose the most preferred hypothetical sheep for breeding purposes. The field data were collected from March to June 2019. From the total respondents, 370 farmers (Estie = 117, Farta = 125 and Lay Gayient = 128) have successfully completed both the questionnaire and the choice experiments. Only 0.47% and 0.4% of the respondents have chosen the opt-out options for rams and ewes, respectively. Such a small proportion of the opt-out options may indicate that the available alternatives are plausible. Given the very small choices, the opt-out options were ignored during data analysis.

Choice modeling and statistical analysis

We started the empirical analysis of preference data using a standard multinomial logit model. Estimation of utility coefficients using a multinomial logit model requires the assumption of independence of irrelevant alternatives (IIA). However, the data could not satisfy the IIA assumption when tested using the Hausman-McFadden test. Fortunately, the generalized multinomial logit (G-MNL) model, derived from mixed and scaled multinomial logit models, does not require the assumption of IIA [29, 30]. Furthermore, the G-MNL model accounts for both taste and scale heterogeneity implying that utility estimates will not be affected by variations in the unobserved component of the model [30]. Mathematically, attribute utility using G-MNL model can be computed as: Where Uint is the utility associated with the i alternative chosen by the n respondent (n = 1, …, 370) at t choice scenario (t = 1 to 4)), X is a vector of observed attribute levels of ram and ewe, β is a vector of mean attribute utility weight, ƞ is s a random term associated with a person-specific deviation from the mean utility, ε is the idiosyncratic error term, σn represents the scale of the error term, γ is a parameter (value between 0 and 1).

Following Eq 2, let Ynit = 1 if the respondent n has chosen the i alternative at t choice situation, or 0 otherwise, the simulated probability choice in the G-MNL model takes the following form. Where , ƞd is a K-vector distributed with multivariate normal (0, ∑), τ is the standard deviation of the scale of the error term, whereas is a scalar with normal distribution N(0, 1). The simulation process requires d = 1… D draws for {ƞd} and {}.

Fiebig et al. [30] proposed different ways of implementing the G-MNL model by imposing restrictions on parameters. In the present study, we set the value γ to zero (γ = 0) to compute the utility coefficient. Given the difficulty of setting an appropriate initial value, most researchers restrict the value of γ to zero.

Maximum simulated likelihood estimates, the goodness of fit of the model and the odds ratio were estimated using the function mlogit and gmnl packages of R software version 3.6.1 [31]. Data for ram and ewe trait preferences were analyzed separately. Socioeconomic characteristics of the respondents, such as sex, educational status, location, sheep flock size, and crop and grazing land sizes, were evaluated for possible influences on preferences of smallholder farmers. However, none of them has exerted a significant effect on trait preferences, thus were omitted from G-MNL to keep models parsimony. Indices for ranking of sheep production objectives and ram and ewe selection criteria of the smallholder farmers were calculated for the first four ranks following König et al. [32] and Bett et al. [33]. Where a is the rank weight associated with trait or criteria m (a1 = 4, a2 = 3, a3 = 2, a4 = 1), X is the proportion of smallholder farmers who ranked the m trait or criteria in the n rank (n = 1 to 4 ranks), m represents sheep traits and the different purposes of sheep production.

Results

General characteristics of the respondents

From a total of 370 respondents, about 95.4% were male farmers. Such a large variation in gender participation was observed because the number of female-headed households who owned sheep and available for sampling during the study period was very small. All the respondents stated that their main occupation is agriculture, involving both crop and livestock productions. The mean crop and private grazing landholdings were 0.8 and 0.28 hectares, respectively. With respect to the literacy status of the respondents, 32.7% had attended informal and religious schools and were able to read and write. Likewise, about 29.2% of the respondents had attended primary school, while only 11.6% reached a post-primary school level. The remaining 26.5% of the respondents were illiterate.

Sheep flock size and composition

Mean flock size and standard deviations for each category of sheep and sampling districts are given in Table 3. The overall mean for sheep flock size of the study areas was 10.21, with a flock size range of 2 to 43 sheep. Across the study districts, sheep flock size was significantly larger in Lay Gayient than the two other districts (P<0.001). With the overall mean of 5.21 ewes, the proportion of breeding ewes accounted for about 51% of the total flock size. Moreover, a significant difference exists among the three districts in the mean number of breeding ewes (P<0.001). Given the overall mean of 0.48, the number of breeding rams was generally small across the study districts. The ratio of breeding ram to ewes was 1:11. Within the flock, ewe and ram lambs are known to be used as replacements for breeding flock or sold for income generation or otherwise slaughtered for consumption. Lambs account for about 26.6% of the total sheep flock.

Table 3

Sheep flock size and composition at the smallholder level by the study districts.

Sheep category$	Districts+						Overall (N = 370)*
	Estie (N = 117)		Farta (N = 125)		Lay Gayient (N = 128)
	Mean	Std.	Mean	Std.#	Mean	Std.	Mean	Std.
Breeding ewes	5.15a	1.34	3.86b	1.19	6.64c	3.2	5.23	2.43
Breeding rams	0.52a	0.57	0.46a	0.5	0.45a	0.6	0.48	0.56
Ewe lambs	0.45a	1.03	0.64a	0.87	1.87b	1.53	1.01	1.34
Ram lambs	0.62a	0.94	0.79a	0.92	0.68a	1.15	0.7	1.01
Lambs	2.83a	1.33	2.76a	1.75	2.58a	2.16	2.72	1.79
Castrated	0.07a	0.31	0.03a	0.18	0.14a	0.68	0.08	0.45
Total flock size	9.65a	2.76	8.54a	3.4	12.36b	6.39	10.21	4.79

*N is the number of respondents,

#Std. is the standard deviation,

+Means in a row with different letters are significantly different (P<0.001)

$Ewe and ram lambs represent those with age group between six months and one year while lambs are both male and female groups with age below six months

Sheep production objectives

The importance of sheep production for the smallholder farmers was examined based on the overall index calculated from the proportions of four ranks (Table 4). Given the highest index value of 0.36, income generation was ranked as the primary reason for keeping sheep. In addition, meat for consumption purpose and manure as organic fertilizer and local fuel sources were ranked as second and third objectives, respectively. Wool production was the least ranked objective of sheep production. However, nearly all respondents practiced the shearing of sheep at least once per year as part of controlling external parasites and have used the wool for traditional purposes.

Table 4

Rank proportions and index values for sheep production objectives at smallholder level.

Sheep production objectives	Rank proportions*				Index
	1	2	3	4
Income generation	0.65	0.33	0.01	0.01	0.36
Meat source	0.32	0.52	0.14	0.03	0.31
Saving/asset	0.02	0.08	0.18	0.09	0.08
Manure source	0.00	0.07	0.45	0.44	0.16
Sheep skin source	0.00	0.01	0.22	0.42	0.09
Wool production	0.00	0.00	0.01	0.03	0.00

*1–4 represent rank 1, rank 2, rank 3 and rank 4

Ram use practices and mating management

Across the study districts, about 46.2% of the smallholder farmers have their own breeding rams with sources born in the flock (36.8%), purchased with partners (7.8%) and bought privately (1.6%). On the other hand, more than half (53.8%) of the respondents were without breeding rams, but all had access to ram services from their neighbors and relatives. According to the view of the respondents, the average duration in which a particular ram stayed within the flock for breeding purposes is about 2.26 years. All the respondents stated that mating is uncontrolled with year-round lambing. Given the absence of pedigree records coupled with the practices of using rams born within the flock, the level of inbreeding within the flock is expected to be high. While discussing with smallholder farmers, their understanding of inbreeding appeared to be minimal.

Ram and ewe selection criteria

About 91% of the respondents have been practicing selection of breeding rams and ewes within their own flock using different criteria. Index values calculated for smallholder farmers’ selection criteria for breeding rams and ewes are given in Table 5. Given the highest index value of 0.34, body size was ranked as the primary selection criteria for breeding rams, followed by coat color and tail type. Similarly, body size (index = 0.34) and coat color (index = 0.25) were two of the most important breeding ewe selection criteria. However, records about sheep performance and pedigree were nonexistent in all respondents involved in the study, thus selection decisions are mainly based on physically observable attributes of sheep. Given the very low index values, adaptation and wool yield traits were the least considered selection criteria for both sexes. Across the study districts, there was no disparity in the order of major selection criteria for both rams and ewes.

Table 5

Rank proportions and index values for breeding ram and ewe selection criteria.

Selection criteria	Rank proportions and index (rams)*					Rank proportions and index (ewes)
	1	2	3	4	Index	1	2	3	4	Index
Body size	0.64	0.21	0.09	0.02	0.34	0.63	0.26	0.05	0.02	0.34
Litter size	-	-	-	-	-	0.02	0.02	0.09	0.12	0.04
Lamb survival	-	-	-	-	-	0.00	0.00	0.01	0.05	0.01
Lamb growth	-	-	-	-	-	0.02	0.04	0.03	0.08	0.04
Lambing interval	-	-	-	-	-	0.02	0.05	0.13	0.21	0.07
Ear size	0.03	0.05	0.13	0.19	0.07	0.02	0.04	0.10	0.07	0.05
Pedigree	0.03	0.01	0.05	0.08	0.03	0.05	0.01	0.03	0.05	0.03
Coat color	0.25	0.44	0.16	0.08	0.27	0.22	0.41	0.15	0.06	0.25
Growth rate	0.01	0.04	0.07	0.08	0.04	0.01	0.04	0.08	0.05	0.04
Sexual maturity	0.00	0.02	0.02	0.10	0.02	0.00	0.06	0.19	0.12	0.07
Libido	0.01	0.03	0.10	0.18	0.05	-	-	-	-	-
Tail type	0.03	0.14	0.18	0.17	0.11	0.01	0.06	0.08	0.11	0.05
Adaptation	0.00	0.00	0.01	0.01	0.00	0.00	0.00	0.02	0.02	0.01
Wool yield	0.00	0.01	0.02	0.01	0.01	0.00	0.01	0.02	0.02	0.01
Horn status	0.00	0.02	0.02	0.10	0.06	0.00	0.00	0.01	0.00	0.00

*1–4 represent rank 1, rank 2, rank 3 and rank 4

Culling criteria for rams and ewes

Nearly all smallholder farmers (97%) responded that both rams and ewes with small body sizes are not preferred for breeding purposes, thus excluded from the flock. Likewise, sheep with black coat color were not favored for breeding purposes in 94% of the smallholder farmers who practiced culling due to unwanted coat color. Furthermore, the majority of the respondents said that rams and ewes could be culled due to old age and fertility problems. Based on the view of the respondent, the average culling age for breeding ewes due to oldness was 9.78 years, while rams were excluded from the flock much earlier than becoming old. Moreover, poor mothering ability of ewes implicated largely on lambs’ performance was reported to be one of the main culling reasons for female sheep. Conversely, the majority of the respondents did not cull rams and ewes due to poor body condition. The respondents highlighted that body condition fluctuation due to seasonal variations in the availability of feed is very common in the study areas.

Choice preference for ram and ewe attributes

Utility coefficients estimated for ram and ewe attributes are presented in Table 6. Pseudo R-square (ρ2) estimates of the model were 0.2 and 0.27 for ram and ewes, respectively, implying that the overall fit of the model is good. According to McFadden [34], ρ2 values ranging from 0.2 to 0.4 are indicators for the excellent fit of a model. For each attribute, one of the levels was used as a reference with a value set to zero. Furthermore, body size and tail type for rams, and body size, growth rate and lambing intervals for ewes were fitted as random parameters. The selection of random parameters was based on the overall contribution to model fitness [29].

Table 6

Estimates of smallholder farmers’ preferences for breeding ram and ewe attributes.

	G-MNL model (ram)			G-MNL model (ewe)
	Estimates	SE	P-value	Estimates	SE	P-value
Fixed parameters
1: Intercept*	0.0	-	-	0.0	-	-
2: Intercept	0.2863	0.117	0.014	1.1342	.251	<0.001
3: Intercept	-0.9403	0.163	<0.001	0.6060	0.236	0.01
4: Intercept	0.9929	0.152	<0.001	1.0351	0.241	<0.001
Coat color (brown)	0.3237	0.039	<0.001	1.1517	0.312	<0.001
Growth rate (rapid)	0.2666	0.043	<0.001	-	-	-
Ear size (large)	-0.1869	0.060	0.002	-	-	-
Horn (horned)	0.2523	0.041	<0.001	-	-	-
Mothering ability (good)	-	-	-	1.3056	0.357	<0.001
Litter size (twin)	-	-	-	0.7737	0.327	0.018
Tail type (good)	-	-	-	0.8547	0.240	<0.001
Random parameters
Body size (large)	0.5132	0.048	<0.001	1.1287	0.333	<0.001
Tail type (good)	0.8986	0.051	<0.001	-	-	-
Growth rate (rapid)	-	-	-	0.9618	0.254	<0.001
lambing interval (short)	-	-	-	0.9645	0.23	<0.001
The standard deviation of random parameters
Body size (large)	0.001	0.072	0.99	0.0055	0.07	0.94
Tail type (good)	0.014	0.318	0.96	-	-	-
Growth rate (rapid)	-	-	-	0.2086	0.244	0.39
lambing interval (short)	-	-	-	0.0745	0.171	0.66
Tau (τ)	0.004	0.13	0.96	0.7565	0.172	<0.001
Gamma (γ)	0	-	fixed	0	-	fixed
Log likelihood null	-1864.1	-	-	-1894.4	-	-
Log likelihood function	-1493.1	-	-	-1376.3	-	-
Number of observations	1473			1474	-	-
McFadden’s ρ2	0.20	-	-	0.27	-	-

*the first choice alternative was used as a reference category hence intercept is zero

All attributes of breeding rams and ewes, included in the model as fixed and random parameters (Table 6), have shown significant influences on smallholder farmers’ preference (P<0.05). In other words, smallholder farmers preferred a ram with good tail type, large body size, predominantly brown coat color and rapid growth in the presence of horns. However, large ear size with an estimated coefficient of -0.1869 was not preferred over a ram with small ears. The magnitude of the utility coefficient shows that tail type of the ram is the most preferred attribute, while ear size is the least preferred. Regarding the choices for breeding ewes, all attributes resulted in positive utility coefficients as expected. Based on the magnitude of utility coefficients, good mothering ability was the most preferred ewe attributes, followed by predominantly brown coat color and large body size. Likewise, short lambing interval and rapid growth rate were the fourth and fifth preferred attributes, while good tail type and twinning were placed in the bottom as the least preferred ewe attributes.

Estimated coefficients for the standard deviation of random parameters for both rams and ewes were not significant (P>0.05), indicating that significant preference heterogeneity was not observed among respondents (Table 6). In contrast to the choice situations for ram attributes, the standard deviation of the scale parameter (τ = 0.7565) revealed significant scale heterogeneity for ewe attribute choice scenarios (P<0.001).

The odds ratio for choosing ram and ewe attribute levels

The odds of selecting a ram with good tail type is 2.46 times higher than a ram with bad tail type (Table 7). Similarly, a ram with large body size is 1.67 times more likely to be chosen by smallholder farmers than a ram with small body size. However, the choice of a ram having large ear size is less likely than a ram with small ear size. Concerning the odds of ewe attribute levels, ewes with good mothering ability is 3.69 times more likely to be picked up by smallholder farmers than ewes with poor mothering ability. Overall, all the preferred attribute levels of ewes have shown high odds ratios compared with the reference level of the corresponding attributes.

Table 7

The odds ratios for all levels of the different ram and ewe attributes.

Attributes	G-MNL model
	Ram odd ratio (95% CI)*	Ewe odd ratio (95% CI)
Body size (large vs small)	1.67 (1.52 to 1.84)	3.09 (1.61 to 5.94)
Coat color (brown vs white)	1.38 (1.28 to 1.49)	3.16 (1.72 to 5.83)
Growth rate (rapid vs slow)	1.31 (1.20 to 1.42)	2.62 (1.59 to 4.31)
Tail type (good vs bad)	2.46 (2.22 to 2.71)	2.35 (1.47 to 3.77)
Ear size (large vs small)	0.83 (.74 to .93)	-
Horn status (horned vs polled)	1.29 (1.19 to 1.40)	-
Lambing interval (short vs long)	-	2.62 (1.67 to 4.12)
Mothering ability (good vs poor)	-	3.69(1.83 to 7.42)
Twining (twin vs single)	-	2.17 (1.14 to 4.12)

*95% lower and upper confidence intervals in the parenthesis

Discussion

A good understanding of livestock production and the breeding system is fundamental to design a breeding program at the smallholder level [35, 36]. More importantly, for breeding programs to be operational at smallholder levels, the active involvement of smallholder farmers from planning to implementation is strongly advised by multiple scholars [13, 32, 37]. In this study, sheep breeding practices and trait preferences of smallholder farmers were identified using participatory approaches. The results of the present study can be used to design village-based breeding programs for indigenous sheep inhabiting the northwest highlands of Ethiopia.

Sheep production objectives and flock structure

In the study areas, income generation and meat consumption were the main objectives of keeping sheep by smallholder farmers. Such objectives imply that indigenous sheep can play huge roles in poverty reduction at the smallholder level if adequate efforts are made to improve their productivity. Similar sheep production objectives have been reported in different parts of Ethiopia, particularly in the crop-livestock mixed farming system [4, 5, 14, 38].

The present study revealed that the average flock size per smallholder farmer is generally small. In-country wise, sheep flock size per household is reported to vary across production systems depending on the availability of inputs and dependence on livestock. For instance, Edea et al. [14] reported sheep flock size of 8 to 11 in a crop-livestock mixed production system, while Nigussie et al. [39] reported an average flock size of about 97 and 72 sheep in Pastoral and agro-pastoral production systems, respectively. For breeding programs targeting genetic improvement through selection, the small number of sheep available at the individual farmer level could be problematic, for instance, from the perspectives of minimizing inbreeding and obtaining optimal genetic gain. In such scenarios, establishing strong collaboration among smallholder farmers to create large breeding flock is fundamental for the sustainable use of genetic resources.

Sheep breeding practices

Valuing indigenous knowledge is vital to ensure the sustainability of a breeding program intended to be implemented at the smallholder level [36, 37]. Interestingly, the majority of the smallholder farmers in the study areas have been practicing selection within their own sheep flock using diverse criteria. It was found that body size and coat color were two of the most important selection criteria for both breeding rams and ewes. Biologically, coat color has a qualitative nature, implying that it cannot be measured on a scale basis. Large body size, on the other hand, can be expressed in terms of linear body measurements such as body length, height, pelvic and girth circumferences. In addition, linearly measured body size traits are easy to measure even at smallholder level and are reported to be moderately heritable [40]. Such facts imply that smallholder farmers could indeed achieve some level of improvement when applying selection based on body size.

Furthermore, although the growth rate was not ranked among the top selection criteria in both sexes, it has been known to have moderate to high genetic and phenotypic correlations with body size traits that can be measured linearly [40, 41]. Thus, body size based selection practices of the smallholder farmers could have a positive impact on the growth rate. With regard to the other attributes, sheep breeds in the study areas are naturally characterized by a fatty tail type, horizontally orientated ear and males are often horned [2, 42].

Another important experience of smallholder farmers in the study areas is the practice of culling of sheep perceived as not suitable for breeding purposes within their own flock. Reasons for culling were due to small body size, unfavorable coat color, old age and fertility problems for both male and female sheep, and due to poor mothering ability of ewes. Such practices are also reported to be implemented by smallholder farmers in other parts of Ethiopia [43].

In the study areas, smallholder farmers often keep breeding ewes for longer periods, while breeding rams stay within the flock for a relatively short period. The practice of using rams for a short duration is assumed to be useful to minimize inbreeding within the flock. Yet, a large number of the smallholder farmers used rams born in the flock, that could result in mating between genetically related sheep, thereby increasing inbreeding. The issue of inbreeding at smallholder levels has been also indicated in other parts of Ethiopia [4, 14, 43] and elsewhere in West Africa [44]. Although pedigree-based records are lacking, the practice of random mating and the use of communal rams can be considered as encouraging breeding practices to reduce the effect of inbreeding in the study areas.

Choice preference for sheep traits

Choice modeling was applied to elucidate the preference of smallholder farmers for indigenous sheep traits. Among ram attributes, a good tail type was the most preferred attribute, followed by large body size and predominantly brown coat color. These attributes were also among the major ram selection criteria, although the order of importance is reversed. The choice of ram attributes seemed to be a reflection of the existing traditional breeding approach implemented in the study areas. Duguma et al. [12] performed a choice modeling in four Ethiopian indigenous sheep breeds (Horro, Menz, Afar and Bonga). These authors reported that sexual activity (libido) for Horro and Menz, good tail type for Bonga and color type for Afar sheep breeds were the most preferred ram attributes. Although libido attribute was not included in the choice experiment of the present study, it was ranked as the fifth ram selection criteria by the smallholder farmers (Table 5). Elsewhere in West Africa, Tindano et al. [45] have reported a high preference for disease resistance in rams. Such preference heterogeneity for ram attributes could likely be due to differences in sheep breed, production system and sociocultural characteristics of smallholder farmers.

Rapid growth rate and the presence of horns were also found to be important ram attributes given their significant influence on the preference of smallholder farmers. Growth related traits, such as weight at different ages, are easy to measure and have reasonably high heritability. Due to such characteristics, growth traits have been the main targets of breeding programs piloted for a few indigenous sheep breeds in Ethiopia [46, 47]. Furthermore, the earlier attempts of selection for body weight on nucleus herds have shown that Ethiopian indigenous sheep are reasonably responsive for selective breeding [48]. Unexpectedly, a ram with small ear size was preferred to a ram with larger ears, although ear size is the least important attribute in terms of the magnitude of the utility coefficient. Based on the index value in Table 5, ear size was ranked as the fourth ram selection criteria, implying that the emphasis of smallholder farmers toward ear size is not strong.

Regarding the preference for ewe attributes, good mothering ability, representing the nourishing potential of ewes for better growth and survival of lambs, was the most preferred attribute. This result is in agreement with Duguma et al. [12] who reported a high choice preference for good mothering ability of ewes in four indigenous sheep breeds of Ethiopia. The strong preference for mothering quality could be a good indicator of the mechanism by which smallholder farmers are trying to be profitable, even under low input sheep production systems. One reason could be is that smallholder farmers often sell lambs for income generation, thus well-nourished lambs are expected to fetch a better price.

Although the tail type and litter size of ewes had shown significant influences on the preference of smallholder farmers, both were less important compared to other ewe attributes. This could indicate that the main emphasis of smallholder farmers for ewe choice is lamb production while focused largely on observable characteristics for ram profiles. However, for ewe attributes other than good mothering ability, smallholder farmers’ preference obtained in the present study was not in agreement with the findings of Duguma et al. [12]. This clearly highlights the importance of evaluating sheep trait preferences on a breed basis prior to designing a breeding program.

One appealing feature of choice modeling using G-MNL is the possibility of accommodating choice heterogeneity among respondents. Unlike for the breeding ram, ewe preference analysis revealed significant scale heterogeneity. A scale heterogeneity could occur due to variations in the choice behavior of respondents [30]. For instance, for some respondents, the choice may be driven by one or a few attributes resulting in a very small estimate for the scale of the error term or a very large utility coefficient. Such choice characteristics could likely be the source of scale heterogeneity in the present study. Fortunately, the G-MNL model takes into account both taste and scale heterogeneity simultaneously. Thus, maximum simulated likelihood estimates of parameters will not be biased due to heterogeneity.

Potential traits for breeding objective

The present study revealed that smallholder farmers have given due emphasis for both quantitative and qualitative attributes of indigenous sheep. Given the easiness of measurement and heritable nature of the traits, body size and growth rate can be the main components of sheep breeding objectives in the study areas. Linear body measurements such as body length, chest girth, pelvic width and height at wither are proxy traits for body size, while growth can effectively be described in terms of weight at different ages. Furthermore, lambing interval, early growth and survival of lambs should be incorporated in the breeding objective. Although the latter two traits were not among the main ewe selection criteria, they are more likely to reflect the mothering ability of ewes that highly influenced smallholder farmers’ preference.

However, qualitative traits such as tail type and coat color cannot be measured on a scale basis, thus are difficult to incorporate directly in the breeding objective, despite their significant effects on preference. This implies selection decisions at smallholder level should not only be based on estimated breeding value of traits in the breeding objective but instead, additional preferences of farmers should be taken into account, for example, in addition to estimated breeding values, considering coat color and tail type as selection decision criteria. It has been said that more than any form of financial support, satisfying the interest of smallholder farmers is vital for the sustainability of genetic improvement programs [37]. Overall, production, reproduction and adaptive traits could be combined alternatively for optimal genetic gain without making the breeding objective more complex.

Conclusion

Although average flock size per head is small, the majority of the smallholder farmers have experiences in selecting and culling of rams and ewes within their own sheep flock. Given the absence of any form of sheep performance recording system at the smallholder level, the breeding practices and decisions mainly rely on observable characteristics of sheep. Based on the magnitude of utility coefficients, a good tail type was the most preferred ram attribute followed by large body size, predominantly brown coat color and rapid growth rate. Similarly, smallholder farmers have shown their highest preference for good mothering ability of ewes followed by predominantly brown coat color, large body size, short lambing interval, and rapid growth rate. The present results implied that breeding objectives incorporating production, reproduction and adaptive sheep traits can be derived by alternatively combining highly preferred attributes having measurable and heritable characteristics. However, selection decisions at the smallholder level should not only be based on the outcome of traits included in the breeding objective but instead, additional preferences of farmers need to be taken into account.

Raw data for respondents’ general characteristics and sheep breeding practices.

It is an SPSS file with all the variables explained fully in the variable view section.

(SAV)

Click here for additional data file.

Raw data for breeding ram and ewe choice preferences.

It is an Excel file with all ram and ewe variables explained fully in the third cell of the sheet.

(XLSX)

Click here for additional data file.

R script for breeding ram and ewe choice preference data analysis.

R script for ram and ewe choice preferences are available separately but contained in the same file.

(TXT)

Click here for additional data file.

The questionnaire used in the survey (English version).

The questionnaire developed to collected socioeconomic and sheep breeding practice data.

(PDF)

Click here for additional data file.

The questionnaire used in the survey (Amharic version).

The questionnaire translated from English into the local language (Amharic).

(PDF)

Click here for additional data file.

24 Feb 2020

PONE-D-20-02104

Breeding practices and trait preferences of smallholder farmers for indigenous sheep in the northwest highlands of Ethiopia: Inputs to design a breeding program

PLOS ONE

Dear Dr Abebe

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Many thanks for submitting your manuscript to PLOS One

It was reviewed by two expert reviewers who have recommended some minor revisions be made prior to its publication

If you could address the reviewers comments and write a response to reviewers, this will expedite second review

I wish you the best of luck with your revisions

Many thanks

Simon

We would appreciate receiving your revised manuscript by Apr 09 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Simon Russell Clegg, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

3. We note that Figure #1 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

1.    You may seek permission from the original copyright holder of Figure #1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2.    If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript aimed identifying breeding practices and smallholder farmers’ trait preferences for indigenous sheep in Ethiopia. This kind of study is important since smallholder practices of animal breeding requires different approaches and information are scarce in the literature. The manuscript are well written, but some points need better understanding (attached I provide some comments in the own text). The discussion needs more attention since repeat several times the results. I suggest review it.

Reviewer #2: The manuscript was well-written. Topic is very unique. I suggest to publish the manusript after minor revision.

Introduction

-I would suggest shortening the introduction in term of general context, and including more information about Breeding practices and breeding program for indigenous sheep in Ethiopia.

- I would suggest describing what kind of research work/studies has been done with various; methodological approaches are available to identify breeding objective traits in sheep in Ethiopia.

Materials and methods

- The survey area: It is good to have the map but also add some statistics. How many square miles, relative size to the whole of Ethiopia, what is the population of people, livestock species in the region of study versus the whole country etc.

- How were the sampled villages/ districts chosen? Give more details. What information was there before hand?

-The quality of Figure 2 is not good. Moreover, for the method choice scenario, it is advisable to use 2 choice scenarios and a third scenario called "no choice". How can you further explain your approach and justify its limitations?

-Explain further, how you conducted your focus group discussions.

-You can use the articles below to enrich the Materials & Methods section and the discussion section:

1. Traoré, B., Govoeyi, B., Hamadou, I. et al. Analysis of preferences of agro-pastoralists for the attributes of traction dromedaries in harness cultivation: A case study of the Koro district of Mali. Pastoralism 9, 19 (2019). https://doi.org/10.1186/s13570-019-0153-9

2. Tindano, K., Moula, N., Traoré, A. et al. Assessing the diversity of preferences of suburban smallholder sheep keepers for breeding rams in Ouagadougou, Burkina Faso. Trop Anim Health Prod 49, 1187–1193 (2017). https://doi.org/10.1007/s11250-017-1315-7

3. Siddo, S, Moula, N, Hamadou, I, Issa, M, Marichatou, H, Leroy, P and Antoine-Moussiaux, N 2015. Breeding criteria and willingness to pay for improved Azawak zebu sires in Niger. Archive of Animal Breeding 58, 251–259.

4. Issa Hamadou, Nassim Moula, Seyni Siddo, Moumouni Issa, Hamani Marichatou, Pascal Leroy, and Nicolas Antoine-Moussiaux. Valuing breeders' preferences in the conservation of the Koundoum sheep in Niger by multi-attribute analysis. https://www.arch-anim-breed.net/62/537/2019/aab-62-537-2019.html

-I suggested to include the survey questionnaire.

Results

- The chaper was very well described.

Discussion:

- The chaper was very well described.

-The conclusion should be developed on the outlook for your work.

-The paper is very interetsing, but it can still be improved by exploiting all the collected data.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-20-02104_reviewer.pdf

Click here for additional data file.

11 Apr 2020

Comments from the academic editor:

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Response: Thank you for reminding the very important issue. The authors have prepared the manuscript based on the guideline of PLOS ONE and tried all the best to meet all the PLOS ONE journal style requirements. In all tables footnotes, we changed the italic fonts to normal. We also corrected the file naming for Supporting information files. Ethics statement was added in line 97-102

2. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

Response: Thank for raising this issue. We have developed a questionnaire for our study and attached it as supporting information (both in English and local language (Amharic)). While preparing the questionnaire, we followed the suggestions by Haile et al. (2011) regarding what to collect when planning to study sheep breeding practice and identify breeding objectives that can ultimately be used to design sheep breeding programs at smallholder level in Ethiopia condition. The questionnaire doesn’t contain copyrighted material nor owned by someone.

Furthermore, we added the following text about the survey from line 145 -154

“The questionnaire was prepared following Haile et al. [16] who suggested the types of information that need to be collected regarding sheep breeding practice and breeding objective for the purpose of designing a community-based breeding program. The questionnaire was translated into the local language (Amharic). It includes sheep breeding practices such as production objectives, flock size and composition, selection and culling criteria, ram use and mating system. The list of smallholder farmers who own sheep and dwell permanently in the sampled kebeles was obtained from the local authorities of the respected kebeles of each district. From the list, respondents were selected using a simple random sampling method. For the respondents, willingness and having sheep were the criteria required to involve in the study.”

3. We note that Figure #1 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

You may seek permission from the original copyright holder of Figure #1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

Responses: Thank you for your concern. The authors agreed to remove figure #1 from the submission. However, what we would like to make clear is that we prepared the map using satellite image data obtained from the Ethiopian Mapping Agency. We contacted the Ethiopian Mapping Agency for possible written permission but told us that such enquire is not common. Otherwise, we haven’t found someone who published and owned the copyright for the map we submitted. Because of such confusion, we decided to remove the figure from the system. We provided detailed descriptions of the study areas in the text and table form (considering the comments given by one of the reviewers).

Comments from reviewers

Reviewer #1

Comments:

The manuscript aimed at identifying breeding practices and smallholder farmers’ trait preferences for indigenous sheep in Ethiopia. This kind of study is important since smallholder practices of animal breeding requires different approaches and information are scarce in the literature.

The manuscript are well written, but some points need better understanding (attached I provide some comments in the own text).

Responses: Thank you for highlighting the points and we addressed the feedback line by line as follow:

Line 26, remove the phrase “Choice modeling using”

Response: Line 26, we removed the phrase “Choice modeling using” and corrected the sentence.

Line 55-56, “risk” is a generic word, clarify the type of risk and “living banks” not an appropriate term, clarify this?

Response: the sentence from line 54-56 was rewritten as follow:

“Particularly when crop farming is less reliable due to drought or other factors, sheep are commonly used to mitigate adverse effects, for instance, related to food shortage at the smallholder level”

Line 57, what is abundances? You used many generic terms. Should be more specific

Responses: the authors acknowledged the reviewer for raising this issue. The term abundance is replaced with “the presence of a large number of sheep” in line 57 and the sentence is corrected accordingly.

Line 78 – 80, exclude and rearrange some words from the sentence

Response: the sentence from line 78-80 was corrected based on the suggestion as follow:

“Given the suitability of the area and adaptive potentials of the existing sheep, improving sheep productivity can be a pathway to put smallholders out of poverty in the northwest highlands of Ethiopia.”

Line 179, levels for coat colors, has the breed have brown and white colors only?

Responses: The authors thank the reviewer for raising this issue. Indeed, these are not the only coat colors for the sheep breed in the study areas but are the major representative colors. We defined the levels as “predominantly brown” (which includes brown color only, dominantly brown with white, dominantly brown with black color, etc.) and the same thing for the level “predominantly white”. Besides, the proportion of sheep with dominantly black coat color is very small in the sheep population of the study areas <4% (Bimerow et al. 2011). Moreover, having black colored sheep in the flock is considered by many smallholder farmers as a sign of misfortune implying that incorporating, for instance, black color as the third level could easily lead lexicographic behavior among farmers (choice makers).

Because of the aforementioned reasons, we set two levels for coat color i.e. the most preferred is “brown” and less preferred is “white”.

Line 182, replace the word “pictorial” with “picture”

Response: we replaced the word “pictorial” with “pictures” in line 182

Line 183, What be well educated enumerators?

Responses: the authors thank the reviewer for pointing out expressions that look vague. We replaced the phrase “well-educated enumerators” by “Enumerators with a Bachelor’s Science Degree in Animal Science …” in line 183.

Line 186, insert article “the” before the word “profile”

Response: replaced as suggested

Line 188, Figure #2 does not represent the two levels of choice

Reponses: the authors thank the reviewer for this important issue. We intended to show one example of the choice cards that contain one level of each attribute. Following the reviewer’s comment, we noted that the figure should contain one full choice set to show all levels of the attributes.

In our study, the number of choices in one choice set (block) is 5 (4 profiles plus 1 opt-out). Overall, we have 4 choice sets that contain 16 profiles and 4 opt-outs. This is the design we used. Our assumption to use 4 attribute profiles and opt-out was that participants can have more choice options than providing only two profile choices, and thus we can get a better understanding of which trait(level) is very important for the choice makers. Because the two levels of each attribute appeared twice but with different combinations.

We prepared and uploaded a figure containing the 4 profiles plus the opt-out option that represents one choice set

Line 252, delete the phrase “while the remaining were female farmers”

Response: we deleted the phrase in line 252 “while the remaining were female” and corrected the sentence accordingly.

Line 258, it is important to define what means spiritual educational system

Response: We replaced “spiritual education system” by religious schools. For the sake of ethics, we are not going to mention the type of religion but farmers were able to read and write because they were educated in such schools.

sub-topic “sheep flock size and composition” should come first than sub-topic “sheep production objective

Response: thank you for pointing out such inconsistencies, we rearranged the two sub-topics and their contents as suggested

Line 280, replace “a” with “the”

Response: replaced as suggested

Line 284, replace “using” with “used”

Response: replaced as suggested but we removed “been” to correct the grammar

Line 265, insert “other”

Response: inserted as suggested

This information about the age of ram and ewe lambs and lambs must be in footnote of the table. Remember that a table must have information independent of the text.

Responses: we placed the information about the age of ram and ewe lambs and lambs as a footnote below Table 3 (line275-276) and rewritten the sentences as follow in line 269-271:

“Within the flock, ewe and ram lambs are known to be used as replacements for breeding flock or sold for income generation or otherwise slaughtered for consumption. Lambs account for about 26.6% of the total sheep flock.”

Line 330, delete “both sexes given”

Response: we deleted as suggested and corrected the sentence by inserting “who” in line 331

Line 362, the value of scale parameter (τ) is 0.7565 based on the table not 0.7535

Response: Thanks a lot, the value of the scale parameter (τ) is 0.7565, we accepted the correction

Line 367-368, delete the first sentence

Response: deleted as suggested and the second sentence (line 367-368) was modified as follow:

“The odds of selecting a ram with good tail type is 2.46 times higher than a ram with bad tail type (Table 7).”

Line 509, What would your suggestion for this?

Response: The authors thank the reviewer for raising this point. We incorporated our suggestion in line 509-510: “For example, in addition to estimated breeding values, considering coat color and tail type as selection decision criteria.”

We know that coat color and tail type are qualitative traits that cannot be measured to estimate breeding value (unlike growth traits), this could only be addressed if selection decision is based on both subjective criteria and estimated breeding values. This, however, is at the expense of genetic gain but sustainability and acceptability of a breeding program at smallholder is expected if their need is addressed properly. Many scholars argued that the lack of success for many livestock breeding programs in developing countries is largely associated with the failure to account for the interest of smallholder farmers (Gizaw et al., 2011; Wurzinger et al., 2011; Sölkner et al., 1998).

The discussion needs more attention since repeat several times the results. I suggest review it.

Responses: the authors thank the reviewer for raising the very important points and agreed to revise the discussion part based on the suggestion.

We re-examined the discussion part and minimized the mentioning of the results without impairing the ideas/concepts of the remaining paragraph/sentences.

For example:

� Under the sub-topic “sheep breeding practices”:

Line 417-418, we removed the following sentence: “Next to body size and coat color, tail type and ear size were the third and fourth selection criteria for breeding rams.” And line 419-421 was rewritten.

Line 421-422, we removed the following sentence: “For breeding ewes, female reproductive traits such as lambing interval, age at first lambing and litter size were the selection criteria considered following body size and coat color.”

Line 427-432, we rewrote the following sentence: “In the study areas, the average production lifetime of breeding ewes is about 9.55 years implying that smallholder farmers often keep preferred ewes for a longer period. Breeding rams, on the other hand, stay within the flock relatively for short period, which is around 2.26 years on average indicating that rams provide breeding services for about one year considering a year of sexual maturity age” added in line 433-434 as “In the study areas, smallholder farmers often keep breeding ewes for longer periods while breeding rams stay within the flock relatively for a short period.”

Furthermore, line 435 -437 was rewritten as “Yet, a large number of the smallholder farmers used rams born in the flock that could result in mating between genetically related sheep thereby increase inbreeding.”

� under the sub-topic “choice preference for sheep traits”

Line 475-477, we deleted the following text:” Besides good mothering ability, predominantly brown coat color, large body size, short lambing interval, and rapid growth rate were the preferred attributes of breeding ewes by smallholder farmers”

And line 477-479 was rewritten as “Although the tail type and litter size of ewes had shown significant influences on the preference of smallholder farmers, both were less important compared to other ewe attributes.”

Line 487-488, we deleted the sentence “However, taste heterogeneity in both sexes was not statistically significant.”

Reviewer #2

Comments:

The manuscript was well-written. Topic is very unique. I suggest to publish the manusript after minor revision.

Introduction

-I would suggest shortening the introduction in term of general context, and including more information about Breeding practices and breeding program for indigenous sheep in Ethiopia

- I would suggest describing what kind of research work/studies has been done with various; methodological approaches are available to identify breeding objective traits in sheep in Ethiopia.

Responses: the authors thank the reviewer for the insightful comments and suggestions.

We have included the following texts (line 60-70) to describe the sheep genetic improvement efforts attempted so far in Ethiopia, though such efforts address only a few indigenous sheep breeds.

“The causes for low performances of indigenous sheep are known to be multilateral but largely related to the lack of effective breeding programs. For decades, crossbreeding between exotic sheep (such as Awassi from Israel and Dorper from South Africa) and some indigenous sheep breeds have been performed [8, 9]. However, achievements are far below expectations due to the lack of effective crossbreeding strategies and poor adaptability of crossbreeds having higher exotic genotypes [8]. Recently, community-based sheep breeding programs have been designed and implemented for Menz, Afar, Bonga and Horro indigenous sheep breeds [10, 11]. Nevertheless, the majority of the indigenous sheep breeds in different parts of the country are still managed in the traditional breeding system without being supported by proven scientific methodologies and the state of art technologies in animal breeding.”

We have excluded the following texts (line 70-77) which seems general expression.

“It has been known that indigenous sheep are not selected systematically for a specific purpose but instead, developed through natural selection and raised for millennia under the traditional management system. Through such episodes, indigenous sheep are assumed to be merited largely with adaptive traits that are useful to survive and reproduce in low-input and stressful production environments [8]. The problem is that the traditional breeding system implemented at the smallholder level has not been supported with proven scientific methodologies and the state of art technologies in animal breeding. As a result, indigenous sheep have been performing poorly year in and year out, which is a common scenario in the northwest highlands.”

Furthermore, we modified line 85-93 to describe the methodologies that have been used to identify sheep breeding objective traits for different sheep breeds in Ethiopia and chosen the methods that fit our objective.

“Various methodological approaches have been used to identify breeding objective traits in sheep in Ethiopia. For instance, choice experiments [12], ranking among a list of traits [13, 14], and live animal rankings [10, 15]. Recently, a choice experiment has been widely applied to investigate farmers’ preferences for animal traits. It provides a hypothetical depiction of attributes levels giving adequate options for the respondents to reflect their interests. Although it requires higher cognitive efforts, a choice experiment is useful to identify preferences when the number of sheep per household is very small to conduct live animal ranking method [16]. Overall, applying a combination of methods has been suggested to effectively identify breeding objective traits [17].”

Materials and methods

- The survey area: It is good to have the map but also add some statistics. How many square miles, relative size to the whole of Ethiopia, what is the population of people, livestock species in the region of study versus the whole country etc.

Responses: we thank the reviewer for the comments.

We included the following table (Table 1) that contains information about area coverage, human population, cattle, sheep and goat population based on the administrative hierarchy (Ethiopia, Amhara region, South Gondar Zone and the study districts). This was described in the text from line 113 -115.

Table 1. Description of the study districts in comparison to the national, regional and zonal level.

Ethiopia c Amhara Region c South Gondar Zone c Study districts d

Farta Lay Gayient Estie

Area a 426,400 59,733.46 5442.18 424.42 587.81 527.1

Human b 94,351,001 21,134,988 2,484,929 272,177 251,926 251,708

Cattle 60,392,019 16,148,390 1,808,185 213,188 120,579 190,853

Sheep 31,302,357 11,086,083 1,085,652 113,978 88,836 191,985

Goat 32,738,385 7,766,661 514,746 51,556 48,758 104,604

- How were the sampled villages/ districts chosen? Give more details. What information was there before hand? Explain further, how you conducted your focus group discussions.

Responses: the authors appreciate the comments and addressed them in the following ways.

In the sub-topic “description of the study areas”, we mentioned (modified the previous texts) why the three districts were selected.

For example, have huge potentials for sheep production compared to other eight districts of the zone, for instance, in terms of the availability of communal grazing land and large sheep population (line 106-108)

Furthermore, the three districts are known to be the main pure breed sources for one of the sheep breeds we intend to design breeding programs, thus incorporating these areas in the study was assumed to be essential (line 108-110 previous text).

We included a sub-topic “sample areas and focus-group discussion” (from line 127-140) to explain how the sample areas and focus-group participants were selected, how the discussion was conducted in each of the three districts.

“Sample areas and focus-group discussion

Secondary data about livestock population and distribution and availability of infrastructures were obtained from the Agriculture Office of the respected districts. Based on sheep population size and accessibility for transport services, three kebeles (the lowest formal administrative units in the district) were selected from each district. For focus-group discussion, a list of smallholder farmers who are regarded as knowledgeable in sheep breeding was identified with the help of livestock development agents and administrative staffs of the kebele. Then, three participants per kebele were randomly selected from the list. For each district, a separate focus-group discussion was held by involving nine well-experienced farmers, a livestock expert and researcher with the latter two played facilitation roles. During the discussion, a list of ram and ewe traits was identified taking into account their socio-cultural and economic importance both at the local and national levels. To prioritize traits, each participant farmer was asked to rank the traits based on their indigenous knowledge. Finally, the ranks from the three districts were combined and analyzed to identify the most important traits of rams and ewes in which choice experiments were designed. “

-The quality of Figure 2 is not good. Moreover, for the method choice scenario, it is advisable to use 2 choice scenarios and a third scenario called "no choice".

Reponses: the authors thank the reviewer for this important issue. Our intention while preparing this figure (now it is Fig1) was to show one example of the choice cards that contain one level of each attribute. Following the reviewer’s comment, we noted that the figure should contain one full choice set to show all levels of the attributes.

In our study, the number of choices in one choice set (block) is 5 (4 profiles plus 1 opt-out). Overall, we have 4 choice sets that contain 16 profiles and 4 opt-outs. This is the design we used.

Making a choice from 2 choice scenarios (otherwise opt-out) may be relatively easy for the smallholder farmers as their literacy status is low but it may also limit their interest because of the restricted choice freedom (as to our understanding). For example, choosing one item from 4 options and 2 options do not have equal choice freedom, which might also affect the reflection of the real interest of the choice maker.

Our assumption to use 4 attribute profiles and opt-out was that participants can have more choice options than providing only two profile choices, and thus we can get a better understanding of which trait(level) is very important for the choice makers. Because the two levels of each attribute appeared twice but with different combinations. Duguma et al. (2011) used six profile plus an opt-out to identify sheep trait preference in 4 Ethiopian indigenous sheep breeds. But making the number of choices too many may also pose problems such as choice dilemma or failure to clearly understand the difference between choice options. Thus, we assumed that a choice among 4 profiles (otherwise opt-out) is fair.

Regarding the figure, we prepared and uploaded a figure containing the 4 profiles plus the opt-out option that represents one choice set (each level for each attribute appeared twice with different combinations.

How can you further explain your approach and justify its limitations?

Responses: the authors thank the reviewer for raising such important issues.

We applied a questionnaire survey to identify sheep breeding practices, which is a common and appropriate method for such kind of study.

Regarding the choice experiment method, it has its own merits and limitations. The methods, in which their applicability for identifying breeding objective traits in sheep in Ethiopia have been tested, are choice experiment (Duguma et al., 2011), ranking among a group of traits (Gizaw et al., 2010; Edea et al., 2012) and live sheep ranking also known as own flock ranking or phenotypic experiment (Mirkina, 2010; Gebre 2018).

Among the methods: choice experiment and ranking of traits require a hypothetical representation of choices. Despite their hypothetical nature, these methods are very helpful for studies carried out at smallholder level, for instance in our study areas, when sheep flock size (breeding ram and ewe) per smallholder farmers is very few. Thus, when flock size is very small, using live animal ranking method to identify breeding objective traits is less appropriate. That is why we opted to use a choice experiment (for trait preference) and ranking of traits (for breeding practice).

Besides its hypothetical nature, a choice experiment has limitations (Complexity) when large number of traits and levels are studied at the same time. Thus, we used six traits for ram and seven traits for ewes with two levels each.

In addition to what has been stated previously, we have added the following text to explain and justify our research approach (line 157-163)

“Valuing of non-marketable traits is the typical usefulness of the choice experiment method in animal breeding. Furthermore, in low input production systems where smallholders’ literacy level is low and performance recording is virtually absent, trait preference could be better elucidated using choice experiment method [10, 20]. However, when a choice is made among too many tasks, a choice experiment may not reveal true preferences due to biases allied with choice complexity [21]. Thus, the number of attributes and levels should be manageable in size to minimize the complexity of the choice experiment design.”

-You can use the articles below to enrich the Materials & Methods section and the discussion section:

1. Traoré, B., Govoeyi, B., Hamadou, I. et al. Analysis of preferences of agro-pastoralists for the attributes of traction dromedaries in harness cultivation: A case study of the Koro district of Mali. Pastoralism 9, 19 (2019). https://doi.org/10.1186/s13570-019-0153-9

2. Tindano, K., Moula, N., Traoré, A. et al. Assessing the diversity of preferences of suburban smallholder sheep keepers for breeding rams in Ouagadougou, Burkina Faso. Trop Anim Health Prod 49, 1187–1193 (2017). https://doi.org/10.1007/s11250-017-1315-7

3. Siddo, S, Moula, N, Hamadou, I, Issa, M, Marichatou, H, Leroy, P and Antoine-Moussiaux, N 2015. Breeding criteria and willingness to pay for improved Azawak zebu sires in Niger. Archive of Animal Breeding 58, 251–259.

4. Issa Hamadou, Nassim Moula, Seyni Siddo, Moumouni Issa, Hamani Marichatou, Pascal Leroy, and Nicolas Antoine-Moussiaux. Valuing breeders' preferences in the conservation of the Koundoum sheep in Niger by multi-attribute analysis. https://www.arch-anim-breed.net/62/537/2019/aab-62-537-2019.html

Responses: the authors thank the reviewer for providing very useful articles. We used these articles (cited 3) to improve the methodology and discussion parts of our manuscript.

-I suggested to include the survey questionnaire.

Responses: We have included both the English and Amharic (local language) versions of the questionnaire as Supporting information.

Results

- The chaper was very well described.

Discussion:

- The chaper was very well described.

-The conclusion should be developed on the outlook for your work.

Responses: thank you for the nice comment about the conclusion and we noted that the text in line 517-518 may not be in the outlook of our work, thus replaced with the following texts (line 519 -520):

“Given the absence of any form of sheep performance recording system at the smallholder level, the breeding practices and decisions mainly rely on observable characteristics of sheep.”

-The paper is very interetsing, but it can still be improved by exploiting all the collected data.

Responses: we thank the reviewer for the comment. We have used all the data (secondary and primary data) collected for this work.

References

1. Sölkner J, Nakimbugwe H, Zárate VA. Analysis of determinants for success and failure of village breeding programs: Proceedings of the 6th World Congress on Genetics applied to livestock production, 1998 January 11-16, Armidale, New England: University of New England; 1998.

2. Wurzinger M, Sölknera J, Iniguez L. Important aspects and limitations in considering community-based breeding programs for low-input smallholder livestock systems. Small Ruminant Research, 2011; 98: 170–175.

3. Gizaw S, Getachew T, Tibbo M, Haile A, Dessie T. Congruence between selection on breeding values and farmers’ selection criteria in sheep breeding under conventional nucleus breeding schemes. Animal, 2011; 5(7): 995-1001.

4. Bimerow T, Yitayew A, Taye M, Mekuriaw S. Morphological characteristics of Farta sheep in Amhara Region, Ethiopia. Online Journal of Animal Feed Research, 2011; 1(6):299-305.

5. Haile A, Wurzinger M, Mueller J, Mirkena T, Duguma G, Mwai O, et al. Guidelines for setting up community-based sheep breeding programs in Ethiopia. ICARDA tools and guidelines No.1, International Center for Agricultural Research in the Dry Areas; 2011

6. Duguma G, Mirkena T, Haile A, Okeyo AM, Tibbo M, Rischkowsky B, et al. Identification of smallholder farmers and pastoralists’ preferences for sheep breeding traits: a choice model approach. Animal, 2011; 5(12):1984–1992.

7. Gizaw S, Komen H, van Arendonk JAM. Participatory definition of breeding objectives and selection indexes for sheep breeding in traditional systems. Livestock Science, 2010; 128:67–74.

8. Edea Z, Haile A, Tibbo M, Sharma AK, Sölkner J, Wurzinger M. Sheep production systems and breeding practices of smallholders in western and south-western Ethiopia: Implications for designing community-based breeding strategies. Livestock Research for Rural Development, 2012; 24 (7).

9. Mirkena M. Identifying Breeding Objectives of Smallholders/Pastoralists and Optimizing Community-Based Breeding Programs for Adapted Sheep Breeds in Ethiopia. PhD Thesis, University of Natural Resources and Life Sciences, Vienna, Austria. 2010.

10. Gebre KT, Yfter KA Teweldemedhn TG, Gebremariam T. Production objectives, selection criteria and breeding practices of afar sheep in Abaala, afar region, Ethiopia. Journal of the drylands, 2018; 8(2): 834-845.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

20 Apr 2020

PONE-D-20-02104R1

Breeding practices and trait preferences of smallholder farmers for indigenous sheep in the northwest highlands of Ethiopia: Inputs to design a breeding program

PLOS ONE

Dear Dr Abebe

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, your manuscript requires some very minor modifications prior to acceptance. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

Many thanks for resubmitting your manuscript to PLOS One

Your manuscript was reviewed by the expert reviewers and they are happy to accept this subject to a proof read

To aid with this, I have performed the proof read of the manuscript for you and made some minor suggestions.

If you can make these minor modifications, the manuscript will be accepted. Please dont feel that you need to write a response to reviewers comments.

I wish you the best of luck with the minor modifications

Hope you are keeping safe and well in this difficult time

Thanks

Simon

==============================

We would appreciate receiving your revised manuscript by Jun 04 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Simon Russell Clegg, PhD

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: As the expert reviewers recommended that the manuscript be reviewed by a native English speaker, I have carried out this review for you. The comments below are generally minor. If you could make the suggested changes (mainly just additions of commas), the manuscript can then be accepted. Please don’t write a detailed reviewer response, just a simple line saying that they were all done is more than sufficient.

Line 27- …sheep attributes, while descriptive ….

Line 29- …keeping sheep, followed by …

Line 33- …selection criteria, followed by …

Line 40- change ‘objective’ with ‘objectives’

Line 51- …of the country, where Farta sheep ….

Line 58- …for Farta sheep, while Gullilat ….

Line 61- …multilaterally, but largely related …

Line 65- ..crossbreeds, due to having higher exotic genotypes (I think that was what you were trying to say but please check)

Line 68- ..breeding system, without being …

Line 73- … have not yet been identified

Line 77- sheep in Ethiopia, for instance … (link the two sentences)

Line 79- change attributes to attribute

Line 80- …attribute levels, giving adequate ….

Line 82- …conduct a live animal ranking method …

Line 85- highlands of Ethiopia, using a ranking …

Line 93- should recoded be recorded?

Line 97 …production, compared to the other eight ….

Line 102- …mixed farming, where livestock ….

Line 102- …and the livelihood of small ….

Line 108- …average daily temperatures, whereas the minimum …. (link the two sentences )

Line 124- … and researcher, with the latter …

Line 126- ..importance, both at the ….

Line 132- ….interviews, using a ….

Line 133- …[16] , who ….

Line 167- …picture representations ….

Line 180- …alternatives, excluding …

Line 220- …setting an appropriate initial value….

Line 227- …preferences, thus were omitted ….

Line 243- …school, while ….

Line 251- delete ‘is’. (…of breeding ewes accounted for …)

Line 269- …controlling external parasites ….

Line 275- …breeding rams, but all …

Line 276- …had access to ram services ….

Line 277- replace ‘staying’ with ‘stayed’

Line 286- …breeding rams, followed by …

Line 302- …was 9.78 years, while rams were ….

Line 319- ….presence of horns ….

Line 320- ….preferred over a ram with small ears.

Line 321- ..tail type of the ram is the most …

Line 321- …preferred attribute, while ear size …

Line 324- … preferred ewe attribute, followed by predominantly

Line 326- …preferred attributes, while good tail ….

Line 326- is this twining or twinning (have 2 offspring)?

Line 340- …levels, ewes with good ….

Line 342- … than ewes with poor…

Line 362- ..in a crop-livestock ….

Line 362- …system, while ….

Line 372- …qualitative nature, implying that ….

Line 384- …characterized by a fatty tail type ….

Line 385- …orientated ears ….

Line 391- …longer periods, while breeding rams ….

Line 392- stay within the flock for a relatively short period. (reword)

Line 394- …in the flock, that could ….

Line 395- …related sheep, thereby increasing ….

Line 404- …selection criteria, although the ..

Line 420- Unexpectedly, a ram with small ear size was preferred to a ram with larger ears, although ear size is the least important attribute in terms of the magnitude of the utility coefficient.

Line 429- …to be profitable, even under ….

Line 429- One reason could be…. (reword)

Line 440- Unlike for the breeding ram, ewe preference analysis revealed significant scale heterogeneity …..

Line 443- …one or a few attributes….

Line 452- change ‘objective’ to ‘objectives’

Line 453- …for body size, while growth can …

Line 456- …ability of ewes that highly influenced …

Line 459- …thus are difficult to incorporate ….

Line 459- …breeding objective, despite their …

Line 460- change ‘decision’ to ‘decisions’

Line 462- …should be taken into account, for example, in addition to … (merge sentences into one)

Line 463- remove because. (It has been said that…)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: Yes: Simon Clegg

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Apr 2020

Dear PLOS ONE Academic Editor, thank you for the comments. We addressed all the comments based on the suggestion.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

28 Apr 2020

Breeding practices and trait preferences of smallholder farmers for indigenous sheep in the northwest highlands of Ethiopia: Inputs to design a breeding program

PONE-D-20-02104R2

Dear Dr. Abebe

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Simon Russell Clegg, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Many thanks for resubmitting your manuscript to PLOS One

I have reviewed your manuscript and wish to thank you for addressing the reviewer comments.

I have recommended your manuscript for publication, and you should hear from the Editorial Office soon

It was a pleasure working with you, and I wish you all the best for your future research

Hope you are keeping safe and well in these difficult times

Thanks

Simon

30 Apr 2020

PONE-D-20-02104R2

Breeding practices and trait preferences of smallholder farmers for indigenous sheep in the northwest highlands of Ethiopia: Inputs to design a breeding program

Dear Dr. Abebe:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Simon Russell Clegg

Academic Editor

PLOS ONE