论文部分内容阅读
为确定《广东省餐饮业食品卫生质量评价表》在现场评价工作中符合实际的评价界值,利用整群抽样与单纯随机抽样相结合的方法,选定60家餐饮企业进行现场评价,计算出每一企业的评分;综合频数分布和聚类分析方法对评分的研究结果,确定评价的界值。经现场对60家已获得“食品卫生许可证”的餐饮企业评分,最高分为99·3,最低分为42·6,评分的极差为56·7,说明《评价表》对企业的评价敏感度是较高的,从另一侧面说明本次现场所选定的被评分对象基本覆盖了食品卫生状况各异的企业,代表性较好。从企业在各分数段的得分分布看,被评分企业的得分主要集中在70~90分段,占所有被评分企业的56·7%(34/60),而得分70以下和90以上的企业分别占23·3%(14/60)和20·0%(12/60),呈正态分布。参考本次评价对象在各分值段的分布情况,把评价的结论分为3类,“良好、一般、差”,相对应的界值为“90~100、70~、40~”,这样既比较符合实际操作的评价习惯,也符合实际的分布情况,较为合理。
In order to determine the evaluation table of food hygiene quality in Guangdong catering industry, it meets the actual evaluation threshold in the on-site evaluation. Using the combination of cluster sampling and simple random sampling, 60 food and beverage enterprises were selected for on-site evaluation to calculate The score of each enterprise; the results of the research on the scores by the comprehensive frequency distribution and clustering analysis methods, and the determination of the cut-off value of the evaluation. On the spot, 60 food and beverage companies that have obtained the “Food Hygiene Permit” were scored, with the highest score of 99.3 and the lowest score of 42.6, with a very poor score of 56.7, indicating that the “evaluation form” evaluates the enterprises The sensitivity is higher, indicating from the other side that the site selected by the scoring object basically covers the enterprises with different food hygiene status, and the representativeness is better. According to the score distribution of enterprises in each score segment, the scores of enterprises scored mainly concentrated in segments 70-90, accounting for 56.7% (34/60) of all the enterprises scored, while those with scores below 70 and above 90 Accounting for 23.3% (14/60) and 20.0% (12/60) respectively, showing a normal distribution. With reference to the distribution of the evaluation object in each score segment, the evaluation conclusion is divided into three categories, “good, fair and bad”, and the corresponding cutoff value is “90-100, 70-40,” Both more in line with the practice of evaluation habits, but also in line with the actual distribution, more reasonable.