ABSTRACT The objective of this study was to evaluate milk quality for somatic cell count (SCC) analysis and total bacterial count (TBC) in agroecological production systems and conventional systems, relating them to the frequency of milking practices and types of facilities. The study was conducted in 2013 and 2014, in the western region of Paraná State, Brazil. We used a semi-structured questionnaire to identify management practices of producers in the 17 dairy systems (DS) in the city of Diamante d'Oeste and the 27 DS in Toledo. Milk sampling was performed monthly. Data were analyzed using two statistical methods. These were then used to create cross tabulation of groups with production systems and practices that can influence product quality. Cluster classification analysis allowed grouping of the initial 44 production systems into three homogeneous groups: Group 1, with better quality regarding the normative instruction 62 (NI62), with average values of SCC 407.37 (cells × 1000/mL) and TBC 210.50 (cfu × 1000/mL); Group 2, with SCC 328.50 (cells × 1000/mL) and TBC 699.75 (cfu × 1000/mL), both outside the NI62 parameters; and Group 3, with SCC 680.25 (cells × 1000/mL), considered low- and high-value TBC 1775.25 (cfu × 1000/mL), out of NI62. The results indicate that management practices and some other preventions adopted during milking are the determining factors of milk quality. Therefore, SCC and TBC levels are not determined by agroecological or conventional production systems, but rather, these levels are determined by the practices differentially adopted by the production systems.

O objetivo definido neste trabalho foi verificar se produtores de leite que participam de arranjos horizontais (AH) possuem menor assimetria de informações, frente àqueles que não participam desses arranjos. Parte-se do pressuposto de que o maior acesso à informação resulta em menor risco para o produtor rural. Foram aplicados 120 questionários semiestruturados junto a produtores rurais de leite, localizados em quatro macrorregiões do Estado do Paraná. Foi realizada a análise fatorial exploratória (AFE) para definição de fatores relacionados à assimetria de informação em Sistemas Produtivos Leiteiros (SPL's). Os 120 produtores entrevistados foram divididos em dois grupos: aqueles que participavam de arranjos horizontais (AH) e aqueles que não participavam. Esses grupos foram confrontados com os fatores gerados na análise anterior. Foi encontrada diferença significativa (P<0,05) para os fatores "Informações relacionadas às transações" e "Informações globais para a atividade", sendo que produtores de leite que participavam de arranjos horizontais obtiveram melhores resultados. Por fim, conclui-se que produtores que participam de AH têm mais acesso à informação, o que pode criar um ambiente de menor assimetria de informações e consequentemente de menor risco, comparativamente a produtores que não participam de AH

An empirical investigation was conducted to study whether dairy farmers involved in horizontal arrangements (HA) have lower information asymmetry than those who do not participate in HA. Our assumption is that greater access to information results in fewer risky decisions in production systems. One hundred and twenty semi-structured questionnaires were applied to dairy farmers located in four different geographical regions in Paraná State, Brazil. Exploratory factor analysis was used to define factors related to information asymmetry in dairy agribusiness system (DAS) and four factors were defined. In a second step, the 120 dairy farmers were split into two groups: the first one involved in HA and the second one not involved in HA. Mean test (t-student) were performed to compare these groups between factors. Significant differences (P<0.05) were observed for factors related to transaction information and for general market information, and dairy farmers participating in HA achieved the greatest values. Finally, it can be concluded that dairy farmers who participate in HA have higher access to information, which can create an environment with lower information asymmetry and, consequently, be subject to lower risks than dairy farmers who do not participate in HA