On the flight ability of factor analysis in the comprehensive analysis and evaluation

[Paper Keywords] flight capacity, factor analysis, common factor
[Abstract] Using factor analysis, revealing strengths and weaknesses that affect the flight cadet major factor is the integrated response capability. At the same time, ensure that the original data loss in case of the smaller, the number of variables integrated into three common factors, simplifies the data structure, the weights determined objectively, so that the flight more accurate and comprehensive evaluation, to select the flight ability of the students provide a useful reference.

1 Introduction
The development of China's civil aviation industry, requires a lot of pilots. Pilots of civil aviation training quality is further sustained, rapid and healthy development of the important foundation of a major event related to aviation safety. Pilots of high-quality determined by many factors, among which a very important determinant is the pilot's own ability to fly.

Because differences in flight ability, every student pilot may not be able to become a pilot, even though some have become pilots in the flight driving technically uneven. This situation not only in training, bringing economic losses to the country, but also to the flight safety of civil aviation has brought a certain extent. Therefore, improving the quality of pilot training, is particularly important. I think: do the following two aspects are important: First, select the flight ability of the students, the second is training in the learning phase, continuously strengthen the participants flying abilities, improve their ability to fly. In this way, we can train for the Chinese civil aviation pilots qualified to meet the sustained, rapid and healthy development of China's civil aviation requirements for pilots.

Our teachers in "Forecast Flight capabilities" made the first attempt. They used stepwise regression selected six pilot flying ability evaluation index, evaluation of flying ability to establish a regression equation, and flight ability in flight make an assessment of the students, will be level, but the effect is not very satisfactory. The reason is that: �� the evaluation of selected indicators of ability to fly although to varying degrees reflect a student's ability to fly some of the information, but there are some among the indexes correlated to some extent reflected on the information overlap, �� evaluation to determine the boundaries of flight ability levels, people are more subjective factors are involved, affecting the objectivity of the evaluation level. Therefore, it is necessary to find relatively few indicators and design, to integrate the information carried by each index, these composite indicators are independent, the information is not represented by overlapping, but also contains most of the information of the original target. Factor analysis of this idea is reflected, therefore, factor analysis is to solve the problem of a good way.

In this paper, factor analysis, a comprehensive analysis of the ability to fly and evaluation, to achieve better evaluation results, the selection of civil aviation pilots to provide a scientific reference.

2 Factor Analysis
Factor analysis of variables can be observed by studying the correlation matrix or covariance matrix of internal dependencies, the number of variables is called integrated to a few main factors common factor or factors, and use these non-linear common factor observed function to describe the specific factors and variables of each original observation, only to find observable variables governed by the laws of common factors, which exist as a reasonable interpretation of the observable correlation between the original variables, and play a simplified variable dimension and the role of the structure.

Let there be a problem in the correlation between the p variables And the observed variables that p a n set of data, , ..., , All are subject to unpredictable variables m common factors Domination (m <p), at the same time, each variable is also affected by a special factor Constraints, so the original variables Common factor and specific factors can be linear Expressed that

Called factor analysis model. With matrix: , Where The factor loading matrix, Variable for the i-j common factors in the load. For special factors, the actual Application is often ignored. Therefore, in the factor analysis process, first find out the factor loading matrix, and then specific issues, combined with professional knowledge The common factor is given a reasonable explanation and a name. If you find it difficult to find a reasonable explanation of the common factor, and further as a factor rotation, the rotation of the common factors have a more obvious practical significance. If the study of the problems to be, but also the common factor expressed as a linear combination of variables, then each of the variables of the study to calculate the estimated value of the common factor, that factor score, and then use the observable variables for each value of the factor scores can make a reasonable assessment of the variable.

3 Application
Evaluation of ability to fly the six indicators are: Light (hand) reaction time (microseconds) (AA1), sound (foot) reaction time (microseconds) (BB2), the passive response to the optimal value (microseconds) (C1), The total error of reaction times (C2), integrated response to the average time (microseconds) (DD1), integrated response to the total wrong time (DD3). Technical training at the end of the flight, the flight driving students into upper level of assessment, in the upper, middle, and lower in the lower five levels. Now graduates of a senior driving class for the upper flight of 15 people (sort the first 15-bit) and upper 15 people (15 in the post-sorting), a total of 30 students enrolled during the flight test data capacity factor analysis, and a comprehensive evaluation of its ability to fly. As follows:
(I) the indicators reflect the ability to fly and flying ability is inversely proportional to the strength of the first data on the indicators to take the countdown, and then the countdown to the data after taking standardized, standardized data sheet by
(Ii) based on standardized data sheet to calculate the corresponding correlation matrix in Table 1.


(Iii) calculate the eigenvalues ​​of the correlation matrix , Eigenvalue contribution rate and the accumulative contribution rate


From Table 2 we can see: the first three eigenvalues ​​of the contribution rate has reached 76.283%, you can describe the original information has reached 76.283% variable, then the three to carry less information, it is negligible . Therefore, the first three factors extracted by the analysis of the problem can make a better explanation.

(Iv) into the decomposition of the Lord, and use rotation varimax factor rotation method may factor loading matrix (Table 3).

Links to Research Papers Download http://www.hi138.com by rotation, be more satisfactory factor loading matrix. As can be seen from Table 3: X5, X6 are two variables in the factor F1 is greater on the load. This is a typical factor of the integrated response capability, comprehensive response on behalf of the flight capacity of the size of the student, the student pilot action flexibility, agility and accuracy to determine the expression.

In the factor F2, X3 has a larger positive load, in other indicators, there is a relatively small load or load bearing. This shows that the factor F2 is a typical passive response factor, representing the coordinated action of flight students and how fast paced action.

In the factor F3, X1 is a great load, X2 is a large load, so F3 is a typical simple reaction factor, representing the student pilot received a simple light (sound) signal to make the necessary response action the time, that represents the degree of signal response speed.

By factor analysis, with three main factors F1, F2, F3 response on behalf of the six indicators of ability to fly, has a 76.283% certainty, and each factor has also been a clear explanation. F1 contains a variable X5 and the X6 carries most of the information, known as the integrated response capability factor, F2 contains most of the information carried by the variable X3 is called a passive response factor, F3 includes the variables X1, X2 carry most of the information, as a simple response factor.

(V) the use of SPSS, 30 students obtained scores, rankings and comprehensive rankings, see Table 4.


Using factor scores, composite score for each student the value of available factors F1, F2, F3's contribution rate as weight ai (i = 1,2,3) and the corresponding score of the product:
In this way, you can get 30 student pilot flying in the capacity of each composite score zj (j = 1,2, ..., 30) and its comprehensive ranking, that is, the level of the strength of flight ability sort. (Table 4)
4 Conclusion
Through the above analysis, we can draw the following conclusions:
First, the flight ability of the factor analysis, while maintaining the original index, based on large amounts of information, simplifying the observation structure. Reflection of the original capacity of the six indicators of flying down into three-dimensional common factor, with three common factors to describe the six indicators reflect the ability to fly, that affect the student pilot flight capability factor is the strength of their overall response to student capacity, reactive power and simple reaction capability. Table 4, the integrated response capacity factor F1, the score in the top 15 and the flight level is superior driving skills have l1 people, in a simple passive response factor response factor F2 and F3, the first flight 15 and the driving respectively for the upper grades 8 and 10, which show that the ability to affect the flight of the first factor is the cadet's comprehensive response capabilities, integrated response capability, the relative driving skills like flying. Therefore, to train high-quality civil flying personnel, in their develop learning phase, participants in the focus on physical training at the same time, you can set the relevant courses, continuously strengthen the response capacity of the students comprehensive training to improve students integrated response capability.

Second, the use of factor scores, by the weighted sum to obtain a student flying ability 3O composite score and ranking (Table 4). These comprehensive information and scores from the perspective of the system determine the effect of draw weight, and avoid the subjective determination of the weights, so that an objective and comprehensive evaluation of the flight, and fair. As can be seen from Table 4, the flight level of driving the first 15 people for the upper 11's flying ability composite score of 15 in the top flight and the flight capacity measured level of driving 76% compliance rate . There are also two composite scores among the flight capacity of 16 and 17, the flight level of driving skills among the upper 15's ability to fly only two composite scores on the list. Composite scores can be seen flying high capacity, the flight is relatively high level of driving skills. This indicates that: use of factor analysis, to calculate the flight ability composite score, ranking evaluation scores according to the strength of the flight capacity is feasible for the selection of the flight the flight ability of science students provide a useful reference. This will allow airline pilot training quality has been further improved, for China to train more high-quality civil aviation pilots.

Links to Research Papers Download http://www.hi138.com