2.1 Expert basic information
20 experts, including 14 males and 6 females;7 graduate students, 10 undergraduate students and 3 junior college students;13 senior titles, 5 intermediate titles and 2 junior titles.There were 2 employees with working years <5 years, 10 employees with 5-10 years and 8 employees with >10 years.8 people mainly engaged in professional scientific research and teaching, 8 people in administrative management, 4 primary GPs.For the familiarity degree of evaluation elements, 11 people were very familiar with them, 9 people were relatively familiar with them(Table. 2).
2.2 The degree of positivity and authority of the expert
The positive coefficient of experts is the positive degree of experts' participation in research, which is reflected in the recovery rate of the questionnaire [10].In the first round of this survey, 20 questionnaires were issued and 20 were collected, with a recovery rate of 100%.In the second round, 20 questionnaires were issued and 20 were collected, with a recovery rate of 100%.This indicates that the experts were highly motivated.
The coefficient of expert authority was mainly affected by the judgment basis of the indicator Ca and the familiarity Cs [11].The judgment basis of experts was divided into practical experience (assignment 0.5, 0.4, 0.3), theoretical analysis (assignment 0.3, 0.2, 0.1), literature understanding (assignment 0.1, 0.1, 0.05), subjective feeling (assignment 0.1, 0.1, 0.05).The three levels of expert familiarity were assigned 1.0, 0.5 and 0.0 respectively.The authority coefficient of experts was the arithmetic mean value of the familiarity coefficient and judgment coefficient, which was between 0 and 1. The higher the authority coefficient is, the higher the authority degree of experts is.In the first round, the coefficient of expert authority was 0.803 (Ca=0.830, Cs=0.775).In the second round, the coefficient of expert authority was 0.858 (Ca=0.845, Cs=0.870).
2.3 The degree of concentration and coordination of expert opinions
The Kendall coordination coefficient (W) of the two rounds of expert consultation were 0.506 and 0.619, respectively, with statistically significant differences (P<0.01)(Table. 3).
2.4 Results of the first round of expert consultation
According to the opinions of the consulting experts, the three first-level elements, 10 second-level elements and 38 third-level elements of the evaluation index system of general practice graduates' ability and quality were modified as follows:(1) delete indicators: delete secondary elements of a "system analysis", delete the three elements of the five "using health information to guide the community working ability", "analysis of the ability of health", "the skills of using evidence-based medicine", "the ability to apply the law to solve disputes", "ability", opinions from others suggested;(2) modified index: "basic pharmacological knowledge" changed to "Pharmacological knowledge and clinical rational drug use", "health management knowledge" changed to "grassroots health management knowledge", "good communication skills with the patient" changed to "Good communication skills with patients and family members", "health records management skills" changed to "establish health records, use and management", “carry out the work of teaching ability" and "carry out the work of scientific research capacity" into "scientific research”.
2.5 Results of the second round of expert consultation
On the basis of the first round of expert consultation, the evaluation index system of general practice graduates' ability and quality was modified to form 3 first-level elements, 9 second-level elements and 32 level 3 elements.In the second round of expert consultation, the second-level element "continuous self-directed learning" was revised to "educational learning ".According to the opinions of two rounds of expert consultation, and after discussion by the research group, the evaluation index system of general practice graduates' ability and quality was finally determined, including 3 first-level elements, 9 second-level elements and 32 level 3 elements.
2.6 Evaluate index weight results
Yaahp software was used to build the hierarchical structure model, with "general medical graduates' ability and quality evaluation index system" as the decision-making target, the first and second level as the middle layer, and 32 third level as the scheme layer.Yaahp software builds a judgment matrix of decision objectives and intermediate elements according to the model.Combining with the weight of experts, each judgment matrix was scored in group decision-making, and the weight results of each index element were obtained through calculation and analysis( table. 4).