2.2. Phenotypic Data Collection and Statistical Analysis

A total of 78 goats (comprising 9 bucks and 69 does) exhibiting the most typical phenotypic characteristics of the Comune di Sicilia (CCS) goat population were carefully selected from the two herds. Specifically, 25 does and 6 bucks were selected from Petralia Sottana farm, and 44 does and 3 bucks from Bolognetta farm.

Morphometric linear measurements were taken only on the 41 adult subjects (4 bucks and 37 does with an age comprised between 18 months and 5 years); specifically, their age was estimated through the evaluation of the dentition [15,16]. Data were scored on eleven morphometric traits following the descriptor list of FAO (2012) and Abd-Allah et al. (2019) [17,18] for the phenotypic characterizations of goats.

Accordingly, the following traits were recorded using a graduated stick and a measuring tape and expressed in centimeters (cm):

Regarding qualitative data, coat color pattern and type, presence or absence of horn, ears, wattles, and beard were also registered for all 78 subjects according to Sponenberg et al. (1998) and Henkel et al. (2019) [21,22].

Phenothypic data were analyzed with SAS software (release 9.4, SAS Institute Inc., Cary, NC, USA). The univariate procedure of SAS was used to determine the following descriptive statistics for the distribution of the morphometric values (n = 37 female goats) obtained: mean, first quartile, median, third quartile, standard deviation, 95% confidence interval of the mean, skewness, kurtosis, and Shapiro–Wilk test to assess if data were normally distributed.

All morphometric data were analyzed with ANOVA mixed models using the GLIMMIX procedure of SAS. The statistical models included the fixed effects of the farm (1: Bolognetta farm; 2: Petralia Sottana farm), horn (yes or no), and wattles (yes or no). Individual goats were included as random effect. In addition, for the ANOVA analysis, trunk length (TL), chest width (CW), and shin circumference (SC) were log-transformed and presented as back-transformed data due to their not-normal distribution. Pair-wise comparisons were performed using the least significant difference test. Statistical significance was declared at p ≤ 0.05.