JP2005149034A - Method, apparatus and program for presenting condition for guiding individual to success - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To calculate a rate of adaptation that a staff has evaluated as a non-success is at least partially adapted to a success model and to lead conditions necessary for completely adapting the staff to the success model by mathematically connecting staff's evaluation information and objective data such as the determination result (points) of a staff's aptitude test. <P>SOLUTION: A method for presenting conditions for guiding individuals to successes comprises a determination tree analysis step for guiding the success model by a determination tree analysis which is a statistical analysis method by using evaluation information for sorting respective individuals in a group to be analyzed into one of groups of successes, non-successes and intermediates and the determination result information of aptitude tests performed for the individuals and an adaptation degree calculation step for calculating the rate of adaptation of each individual to a success. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for presenting conditions for leading a member to success, an apparatus therefor, and a program therefor, and more particularly, a transition to become a successful person in a desired state with respect to a measurement object such as an employee. The present invention relates to a method for presenting a condition that can present a condition and a condition for leading a member to success, an apparatus therefor, and a program.

  With the recent globalization of the economy, there is an increasing demand for companies to increase their competitiveness through the effective use of human resources.

  In order to make effective use of such human resources, managers, human resources departments, superiors, etc. in each company observe the human resources that are evaluated as successful and experience the conditions that are deemed necessary for those human resources. Based on various conditions prepared by consulting firms, etc. in advance or by intuition, some of them were selected subjectively, and based on these, they were regarded as successful companies. In addition, based on the image of successful people, he has also empirically created training methods for non-successful human resources.

  In the above-described conventional technology, the conditions for becoming a successful person in the company can be derived subjectively to some extent, but this is not a result derived based on objective data, but often includes errors. In addition, the specific method for a successful person to become a successful person relies on someone's experience and cannot be directly derived from objective data. Therefore, there is a problem that effective utilization of human resources must be based on uncertain information that depends on intuition and experience.

  An object of the present invention is to solve the above-described problems. For example, the results of aptitude tests (tests) that companies generally perform on employees are used. In other words, whether it is performance data based on figures such as sales performance of a certain human resource or subjective judgment data given by an evaluator such as a senior manager, employee evaluation information and employee aptitude test determination result (score) By mathematically linking objective data such as, the ratio of those who are evaluated as not successful to the extent that they are partially compatible with the success model at that point in time is calculated. The above objective is achieved by deriving the necessary conditions to meet the requirements.

  As means for solving the above problems, the present invention has the following features.

  The first aspect of the present invention is a method for presenting a transition condition for obtaining a desired standard variable value (for example, successful person, best-selling product) for a plurality of measurement objects (for example, employees, products, etc.). Proposed.

  This method includes a step of obtaining a plurality of explanatory variable values (for example, an item “accepted outside the company” and an evaluation such as “strong, medium, weak”) for each measurement object, and a reference variable for each measurement object. A value of (for example, an evaluation of “successful person, non-successful person, intermediate person”) and a value of a desired reference variable (for example, “successful person” by decision tree analysis based on the value of explanatory variable One or more first groups (e.g., "success group") and one or more second groups (e.g., "non-success groups" that do not have the desired criterion variable value). ”,“ Middleman group ”), the step of calculating the fitness for each of the first groups for the measurement objects belonging to the second group, and the first having the highest fitness The step of selecting a group as the optimal first group of the measurement target, and the measurement target belonging to the second group based on the explanatory variables used for the classification of the optimal first group, And determining a transition condition for shifting to a group.

The second aspect of the present invention is proposed as a method for presenting conditions for leading members to success.
In this method, evaluation information that classifies each member of the group to be analyzed as either a successful person, an unsuccessful person, or an intermediate person, and determination result information of aptitude tests performed by these members are included. And a success model extraction step for deriving a success model by decision tree analysis, which is a statistical analysis method, and a success model suitability calculation step for calculating a ratio of each person's fit to the successful person. Yes.

  The third aspect of the present invention is proposed as an apparatus for presenting a transition condition for obtaining a desired reference variable value for a plurality of measurement objects.

  The apparatus includes explanatory variable value storage means for storing a plurality of explanatory variable values for each measurement object, reference variable value storage means for storing a reference variable value for each measurement object, explanatory variable value storage means, and reference variable Based on the value stored in the value storage means, one or more first groups having a desired reference variable value by decision tree analysis, and one or more second groups not having the desired reference variable value Group to be measured, and decision tree analysis means for outputting the result of grouping and the explanatory variable used for grouping, and the output of the decision tree analysis means, and for the measurement objects belonging to the second group , A fitness level calculating means for calculating a fitness level for each of the first groups, and a fitness level for each of the first groups from the fitness level calculating means, and based on the fitness levels An optimum group selecting means for selecting the first group having the highest matching degree as the optimum first group of the measuring object, the optimum first group selected by the optimum group selecting means, and the classification of the optimum first group Based on the described explanatory variables, the measurement object belonging to the second group has a transition condition determining means for determining a transition condition for the measurement object to transition to the optimum first group.

  The present invention is also realized as a program for causing a computer to execute any of the above methods. Furthermore, the present invention is also realized as a program for causing a computer to function as the above-described device.

  According to the present invention, regardless of objective data such as sales results and subjective data such as evaluation given by an evaluator such as a senior manager, evaluation information of a measurement object and determination of aptitude test of the measurement object By mathematically linking results (scores) and other objective data, we can derive the necessary transition conditions to perfectly match the measurement target with the desired attributes, characteristics, and features, regardless of intuition or experience. Make it possible.

Embodiments of the present invention will be described below with reference to the drawings.
[Device configuration]
FIG. 1 is a functional block diagram showing a configuration example of an apparatus (hereinafter referred to as “transition condition presentation apparatus”) for presenting transition conditions for obtaining desired reference variable values for a plurality of measurement objects. .

  The transition condition presentation device 10 is an information processing device such as a computer or a workstation, and includes an arithmetic processing unit (CPU), a main memory (RAM), a read-only memory (ROM), an input / output device (I / O), And it is comprised with the apparatus which comprises external storage apparatuses, such as a hard disk drive, as needed. A program for causing an information processing device to function as the transition condition presentation device 10 in the ROM or an external storage device, or a method for presenting transition conditions for obtaining a desired reference variable value for a plurality of measurement objects (“ A program for causing a computer to execute “transition condition presentation method” is stored, the program is loaded on the main memory, and the CPU executes the program to realize the transition condition presentation apparatus 10. The condition presentation method is executed. Further, the program need not necessarily be stored in a storage device in the information processing apparatus, and a computer-readable program recording such as a magnetic disk, an optical disk, a magneto-optical disk, a CD (Compact Disc), a DVD (Digital Video Disc), etc. The configuration may be such that it is provided from a medium or an external device (for example, an ASP (Application Service Provider) server) and is loaded on the main memory.

  The transition condition presentation device 10 includes an explanatory variable value storage unit 101 that stores a plurality of explanatory variable values for each measurement object, a reference variable value storage unit 102 that stores a reference variable value for each measurement object, and an explanatory variable value. Based on the values stored in the storage means 101 and the reference variable value storage means 102, the decision tree analysis has one or more first groups having a desired reference variable value and the desired reference variable value. A decision tree analysis unit 103 that groups measurement objects into one or a plurality of second groups that are not present, outputs a result of grouping and explanatory variables used for the grouping, and receives an output of the decision tree analysis unit 103; For the measurement objects belonging to the second group, the fitness calculation means 104 for calculating the fitness for each of the first groups, and the fitness calculation means 104 to the first group An optimum group selecting means 105 for receiving the matching degree for each of the groups, and selecting the first group having the highest matching degree as the optimum first group of the measurement object based on the matching degree; Based on the selected optimal first group and the explanatory variables used for classification of the optimal first group, the transition condition for the measurement target to move to the optimal first group for the measurement target belonging to the second group Transition condition determining means 106 for determining the desired condition and output means 107 for presenting the user with a favorable condition desired to be determined by display or printing.

  The measurement target (also referred to as an individual or a sample) handled by the transition condition presentation device 10 may be any object that can be evaluated, and is not limited to a person. Further, even if it is a person, it is not limited to an individual, and a group (for example, a school class, a company, a branch office, a department, a section, etc.) can be a measurement target.

  Each measurement object needs to have a value of the explanatory variable. `` Explanatory variable '' is an explanatory variable for classifying the measurement object. For example, if the measurement object is an employee, psychological factors that are difficult to grasp as numerical values such as `` sociality '' and `` aggressiveness '' Any element that can be displayed as a numerical value such as “numerical processing capability” or “sales” can be used as an explanatory variable.

  Each measurement object needs to be measured for the value of the reference variable. “Standard variable” is information indicating the result of evaluating a measurement object against a certain evaluation criterion, and is not classified as a successful person, an unsuccessful person, or an intermediate person (a successful person or an unsuccessful person) in a certain company, for example. And the sales amount of the salesperson.

  The transition condition presentation device 10 is intended to present a transition condition for obtaining a desired standard variable value. “Transition condition for obtaining a desired standard variable value” means that if a measurement object does not have a desired standard variable value (for example, a successful person), what condition should be provided This is information indicating whether the measurement object can have (or is estimated to have) the value of the desired reference variable. Note that the “value of the desired standard variable” does not necessarily have positive, positive, or positive value, but has negative, negative, or negative value, such as a product that fails, cannot be sold, etc. It doesn't matter.

[Device operation (transition condition presentation method)]
Next, an operation example of the transition condition presentation device 10 will be described. The transition condition presentation device 10 operates to execute the transition condition presentation method shown in FIG.

  First, in order to execute the transition condition presenting method, the transition condition presenting apparatus 10 acquires a plurality of explanatory variable values for each measurement target (step S201). This is performed, for example, as an input to the computer of the result data of the aptitude test of the measurement object.

  Further, the transition condition presentation device 10 acquires the value of the reference variable for each measurement target (step S202). This is performed, for example, as input to the computer of the upper evaluation data of the measurement target.

  The transition condition presentation device 10 has one or more first groups having a desired reference variable value by decision tree analysis based on the value of the explanatory variable, and one or more having no desired reference variable value The measurement target is classified into the second group (step S203).

  Next, the transition condition presentation device 10 calculates the fitness for each of the first groups for the measurement objects belonging to the second group (step S204).

  Next, the transition condition presentation device 10 selects the first group having the highest fitness level as the optimal first group to be measured based on the fitness level obtained in step S204 (step S205).

  Next, the transition condition presentation device 10 uses the measurement target belonging to the second group to shift the measurement target to the optimal first group based on the explanatory variables used for the classification of the optimal first group. Transition conditions are determined (step S206).

  Finally, the transition condition presentation device 10 outputs the determined transition condition for each measurement target (step S207). At this time, together with the transition condition, the display of the fitness level for each first group, the display of the optimal first group, and the like may be performed together.

An embodiment of the present invention will be described with reference to FIGS.
The embodiment described here is, for example, a person who has been evaluated as a successful person in each company, that is, has achieved numerically high performance such as sales performance, or has received sales force and management from an evaluator such as a senior manager. A person who is subjectively evaluated as having excellent power derives the conditions (success model) that are unique to the company that the company has, and a person who is evaluated as not successful A method for calculating the proportion of the success model that is even partially adapted at that time, and further deriving the conditions necessary to fully conform to the success model, and a transition condition presentation device for deriving such conditions, It has been proposed as a transition condition presentation method.

  Explained by exemplifying the logic to extract the success model corresponding to the first group from the data of the employee to be measured based on statistical mathematical methods in a company, and calculate the success model suitability of each employee To do.

  In this case, all employees who want to calculate the success model suitability corresponding to the suitability of each first group are subjected to a predetermined aptitude test (test) that is generally available on the market. Obtain test result data corresponding to the value. In addition, for each employee, items that serve as requirements for the successful person in the company corresponding to the value of the standard variable are assigned. In this embodiment, the requirement measure is a successful person at the present time of the company, an unsuccessful person who is not, and an intermediate person between them. Therefore, the employee to be measured is given one of “successful / non-successful / intermediate” as the value of the reference variable.

[Decision tree analysis processing]
First, the success model extraction process corresponding to the decision tree analysis step will be described.

  As described above, it is shown in the employee's successful / unsuccessful / intermediate sorting information corresponding to the value of the standard variable, and the result of the aptitude test performed by each employee as the value of the explanatory variable. A statistical mathematical analysis method called decision tree analysis is used by using the scores of the diagnostic item elements.

  There are various methods such as CART, C4.5, and CHAID for decision tree analysis. In this embodiment, for example, a method called CHAID (chain) analysis included in AnswerTree, which is a program product of SPSS, is used. However, the present invention is established even if a decision tree analysis method other than the CHAID analysis is used, and the present invention is not limited to the CHAID analysis.

  The CHAID internal algorithm uses a widely used statistical mathematics method called chi-square test to determine the success / non-success / intermediate distribution of sample employees and the results of aptitude tests performed by individual employees. The relationship between the score of the diagnostic item element shown in the above is analyzed in detail.

  That is, for the decision tree analysis program such as CHAID analysis, the distribution information of successful / unsuccessful / intermediate employees of sample employees is a standard variable, and the diagnosis items shown in the result of aptitude test performed by individual employees By inputting the element score as an explanatory variable, the success model of the company is output.

  At this time, the score of the diagnostic item element, which is the value of the explanatory variable, is normally given as a deviation value (which can be converted into a deviation value if it is not a deviation value), but is converted into three types of values as follows. That is, a value greater than the deviation value 54.31 is classified as “strong”, a value between the deviation values 45.69 and 54.31 is “medium”, and a value less than the deviation value 45.69 is classified as “weak”. Here, the deviation values “45.69” and “54.31” are deviation values corresponding to three equal divisions when the sample follows a normal distribution. Further, the condition values for these divisions may be changed according to options.

  A success model output from a decision tree analysis program such as CHAID analysis is a condition of a successful person peculiar to the company, and indicates a combination of diagnostic item elements of aptitude tests which are explanatory variables. The combination is called a success pattern, and is usually output in one or more ways. A set of success patterns is called a success model.

  The diagram shown in FIG. 3 is a diagram showing an example of the output of CHAID analysis corresponding to the decision tree analysis step. Nodes (N0 to N13) indicating the set of employees to be measured and explanatory variables adopted when classifying the upper level nodes into a plurality of nodes (evaluation items shown in the results of aptitude test; C1 to C5) Is displayed. Further, in the boxes indicating the nodes N0 to N13, the standard variables (successor / non-successor / intermediate distribution information) of the employees belonging to the node are shown. In this example, when the ratio of successful persons included in the node is equal to or greater than a predetermined value, the node is determined to be a successful pattern.

  In the example shown in this figure, of the success patterns A (node N6), B (node N13), C (node N11), and D (node N4) among the nodes N1 to 13 branched from the node N0 which is the population. Four success patterns are output. For example, the success pattern A branches in the order of node 2 (N2) and node 6 (N6) from a box written as node 0 in the top row, and the combination of the branch conditions is “power to face difficulty” ( The deviation value of C1) is “medium or strong”, and the deviation value of “visual idea power” (C3) is “strong”. Based on this branch condition, a transition condition to a success pattern A corresponding to one of the first groups having a desired reference variable value is generated.

  The other successful patterns B, C, and D are also output with combinations of branch conditions.

  In addition, the information written in each box called a node indicates how many employees are assigned to successful / unsuccessful / middlemen and what percentage of the total. In the corresponding node, the number of successful persons is larger than the number of intermediate persons and unsuccessful persons.

  In addition to this figure, information regarding which node an individual employee enters may be output separately.

[Relevance calculation step]
Next, an example of the success model fitness level calculation process for a certain employee corresponding to the fitness level calculation step will be described.

A certain success pattern P is composed of a combination of n diagnostic item elements (corresponding to explanatory variables), that is, a combination of n branch conditions.
(1) The deviation value of the diagnostic item element X1 is Y1, and
(2) The deviation value of the diagnostic item element X2 is Y2, and
...
(N) The deviation value of the diagnostic item element Xn is Yn
Suppose that

  Here, the deviation values Y1, Y2,..., Yn are either “strong”, “medium”, “weak”, “weak or medium”, or “medium or strong”, and numerically as described above. Further, a value greater than the deviation value of 54.31 is “strong”, a value between 45.69 and 54.31 is “medium”, and a value less than 45.69 is “weak”.

  On the other hand, the deviation values (corresponding to the values of explanatory variables) of the diagnostic item elements X1, X2,..., Xn indicated in the determination result of the aptitude test of a certain employee are Z1, Z2,. Suppose that it is Zn.

  If the value of the deviation value Z1 of the diagnostic item element X1 matches the deviation value range of Y1, the fitness W1 of the diagnostic item element X1 is assumed to be 100%. However, if the value of Z1 does not match the deviation value range of Y1, W1 is calculated by the following formula.

If Y1 is "weak", W1 = (120-Z1) ÷ (120-45.69) x 100 (%)
If Y1 is "weak or medium", W1 = (120-Z1) ÷ (120-54.31) x 100 (%)
If Y1 is "medium" and Z1 is greater than 54.31, W1 = (120-Z1) ÷ (120-54.31) x 100 (%)
If Y1 is "medium" and Z1 is less than 45.69, W1 = (Z1-(-20)) ÷ (45.69-(-20)) x 100 (%)
If Y1 is "medium or strong", W1 = (Z1-(-20)) ÷ (45.69-(-20)) x 100 (%)
If Y1 is “strong”, W1 = (Z1-(-20)) ÷ (54.31-(-20)) x 100 (%)
In the above calculation formula, “120” and “−20” are deviation values that can be considered that the fitness is 0% when the sample follows a normal distribution. Similarly, the fitness W2,..., Wn of each diagnostic item element X2,..., Xn is calculated using the above formula.

The overall fitness WP of the employee's success pattern P is given by the following formula: WP = W1 ÷ 100 × W2 ÷ 100 × ・ ・ ・ × Wn ÷ 100 × 100 (%)
Is calculated by

  The fitness calculation step is thus completed.

[Optimum group selection step]
When the fitness level calculation step is completed, an optimal success pattern corresponding to the optimal first group is selected based on the fitness level.

  In the above example, if the success model of the company has a success pattern other than the success pattern P, the fitness value of the employee to all the success patterns is calculated, and then the highest value among them is calculated. By selecting a success pattern having a large fitness level, it is possible to obtain an optimal success pattern for the employee and a corresponding fitness level.

[Transition condition determination step]
Finally, a method for deriving the conditions for transition to an optimal success pattern for an employee is shown.

  In the above example, if the overall fitness WP of a certain employee's success pattern P is 100%, the employee is identified as a person who has already matched the success pattern. If WP is less than 100%, it is possible to indicate the transition condition to the optimum success pattern of the employee by the following logic.

  To that end, in the success pattern optimized for the employee, the diagnostic item elements X1, X2,..., Xn conditions Y1, Y2,. Deviation values Z1, Z2,..., Zn, and the fitness levels W1, W2,.

  First, Z1, Z2,..., Zn are classified into “strong”, “medium”, and “weak” on the basis of the same deviation value as Y1, Y2,.

  For example, if the employee's conformity W1 to the diagnostic item element X1 is 100%, the transition condition of the diagnostic element item is “Z1 → as is”. When W1 is less than 100%, the transition condition is determined by the following method.

(1) If Y1 is “weak”, the transition condition is “Z1 → weak”.
(2) If Y1 is either “weak or medium”, “medium”, or “medium or strong”, the transition condition is “Z1 → medium”.
(3) If Y1 is “strong”, the transition condition is “Z1 → strong”.
Similarly, the transition conditions for the diagnostic element items X2,.

  The table shown in FIG. 4 is a list of the suitability and transition conditions of each employee for the optimum success pattern, and displays the conformance corresponding to the optimum success pattern and the transition conditions for the unsuccessful person to the optimum success pattern. doing.

  This table has one record 400 for each employee. Each record has a personal ID field 401 for storing an employee ID for identifying an individual, a suitability field for each success pattern 402 for storing a fitness value for each success pattern, and an optimum for storing information for identifying an optimal success pattern. It has a success pattern field 403 and a transition condition field 404 indicating a transition condition for the employee of the record. In this example, transition conditions are not shown for employees who are successful.

[advantage]
This embodiment has the following advantages.

  (1) By examining the diagnostic item elements indicated in the results of aptitude tests performed by individual employees and the successful / unsuccessful / intermediate assignment of employees as described above, Necessary diagnostic item elements can be found out. In other words, by associating the requirement scale required for different successful persons in each company with the results of the aptitude test of the sample employees existing in the company, it is possible to obtain the conditions as a successful person unique to that company. It becomes possible. Moreover, such conditions can be obtained by using the results of general aptitude tests. In particular, it is time-consuming to create different test questions for each company for new criteria. Not necessarily required.

  (2) In this embodiment, by using the following statistical mathematical method and calculation logic with respect to the above-described method, the human resource success model and the calculation standard of each employee's fitness to it are made more accurate. Yes.

  (3) If the measurement target is a new employee, the possibility of success of the new employee can be calculated.

  (4) If the target of measurement is an existing employee, it will be possible to provide guidance by presenting a model of success to be aimed at and issues to be overcome.

Functional block diagram showing a configuration example of the transition condition presentation device The flowchart which shows the process example of a transfer condition presentation method The figure which shows the example of the result of the decision tree analysis used by this invention The figure which shows the example of the result output of a transition condition presentation apparatus or the transition condition presentation method

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Transition condition presentation apparatus 101 ... Explanation variable value storage means 102 ... Standard variable value storage means 103 ... Decision tree analysis means 104 ... Conformity calculation means 105 ... Optimal group selection means 106 ... Transition condition determination means 107 ... Output means N0- N13 ... Nodes C1 to C5 ... Branch condition (explanatory variable)

Claims (5)

In a method for presenting transition conditions for obtaining a desired standard variable value for a plurality of measurement objects,
Obtaining a plurality of explanatory variable values for each measurement object;
Obtaining a reference variable value for each measurement object;
Based on the value of the explanatory variable, one or more first groups having a desired reference variable value by decision tree analysis, and one or more second groups not having the desired reference variable value Classifying the measurement object into
Calculating the fitness for each of the first groups for the measurement object belonging to the second group;
Selecting the first group having the highest fitness as the optimum first group of the measurement object based on the fitness;
Determining a transition condition for the measurement target to shift to the optimal first group based on the explanatory variable used for the classification of the optimal first group for the measurement target belonging to the second group. A method characterized by that. Using the evaluation information that classifies each member of the group to be analyzed as either a successful person, an unsuccessful person, or an intermediate person, and the judgment result information of aptitude tests conducted by these members, statistics Success model extraction step to derive a success model by decision tree analysis which is a dynamic analysis method,
A success model goodness-of-fit calculating step for calculating the proportion of each person's fit to the successful person;
A method for presenting conditions for leading a member to success. In a device for presenting transition conditions for obtaining a desired reference variable value for a plurality of measurement objects,
Explanatory variable value storage means for storing a plurality of explanatory variable values for each measurement object;
Reference variable value storage means for storing the value of the reference variable for each measurement object;
Based on the values stored in the explanatory variable value storage means and the reference variable value storage means, one or a plurality of first groups having a desired reference variable value by decision tree analysis, and a desired reference variable value A decision tree analyzing means for grouping measurement objects into one or a plurality of second groups not having, and outputting a result of grouping and explanatory variables used for the grouping;
A degree-of-fit calculation means for receiving the output of the decision tree analysis means and calculating the degree of fit for each of the first groups for the measurement objects belonging to the second group;
Optimal group selection means for receiving the fitness for each first group from the fitness calculation means, and selecting the first group having the highest fitness based on the fitness as the optimal first group of the measurement target; ,
Based on the optimum first group selected by the optimum group selection means and the explanatory variables used for classification of the optimum first group, the measurement object belonging to the second group is the optimum first group. And a transition condition determining means for determining a transition condition for shifting to the apparatus.   The program for making a computer perform the method of Claim 1 or 2.   The program for functioning a computer as an apparatus of Claim 3.

Images