200907876 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種特質測評系統及方法；且 明係關於一種可自動造題、選題並分析特質之'系統 【先前技術】 適性化的心理測驗成為企業甄選人才之未 供得以精準評估受測者之適性化測驗，且借 ^勢’而欲挺 不可或缺的重要構件。然而，目前的題庫;== 項簡單的工程。 俊盾的靖庫’則並不是^200907876 IX. INSTRUCTIONS: [Technical field of invention] The present invention relates to a trait evaluation system and method; and the system relates to a system that can automatically create questions, select topics, and analyze traits. The test has become an important component of the company's selection of talents that can be used to accurately assess the appetite test of the subject, and to use the potential. However, the current question bank; == simple project. Jundun’s Jingku’ is not ^
同日;取代魏纽輕題庫的缺點，並 質置， 時間，使得題庫的建置更有效;。大_人力成本，並可節省 以 目 ======，侧嫩-種得 ;根據與一系絲知 素材排列、、且&，以產生複數個測驗題 選出特質i驗題、目之坪估參數，從這複數個測驗題目中 計受測叙物败答資料，估 5 200907876 【發明内容】 包含本一一置面評系統。本發明之系統 =W裝置，係具有-題目產生模組，將至少 梦署=第二素材制組合’以產生複數個測驗題目；-儲存 二播ί、儲存該複數個測驗題目；以及—評估裝置，係且有- 自硬數個測驗題目中選出一特質測驗題目，此評裹估二二」 估計模組係根據對應特質測驗題目之一作;; f 統包本共一種特質測評系統。本發明之系 化通裝置，係具有一題目產生模組，將至少一笛—表 少一第二素材排列組合’以產生複數個測驗題目―： 模組以及-估計模組，選題模組係根據一 ίί』、具ί:選題 ;:驗:二中選出-特質測驗題目，此評估參數ii 一系 ^ 據對應特質測驗題目之一作答資料，估言ί 议、，使用者；丨面，供輸入特質測驗題目之一作夂資料· 及一分析模組，以利用作答資料即時分析出一數1口、七Μ =明之又一面向，係提供一種特質測評系統。本發 測驗題目，此評任夫^ , B 特質 作答質而!=^ 題裝本置發明本27之面造向題， 第-素材及至少目產f模組，將至少-目；以及一且合，以產生複數個測驗題 、杈組，供選出符合一預設標準之測驗題目。 6 200907876 本發明之再一面向，係提供 估裝置。本發明之評估裝置包含：題且系統之評 作答資料;以及一=二：二題目之— 答資料，估計特質。 揭特貝而驗靖目之一作 本發明之再一面向，係提供一種 理褒置。本發明之管縣置包含：3二特系統之管 答資料；以及-分析模组，供分析貝=庫’1、儲存複數個作 本發明之再-面肖，據。 ί二;π服器更包含:-題目產生模組，將i c 目；-選題池，雜據―坪mvtt複數個測驗題 驗題目之一作答資料，估計‘質;-客： 上透過-網路’提供對應特質.題目之作 法包本 下之各再牛一驟面向，ί提供一種特質測評方法。本發明之方 ΐί Hi 一造題步驟’係將至少一第一素材及至少 相棱一ΐίϊϋ組合’以產生複數個測驗題目；—選題步驟， 題目i一夂:質=聯二，入步驟，輸入對應特質測驗 特質。丨# 及—估計步驟，係根據作答資料，估計 半if.明之ίτ面向，係提供一種特質測評方法。本發明之方 '03.一以題步驟，係將至少一第一素材及至少一第-音絲 >數’自稷數個測驗題目中選出—特質測驗題目，此評估^ 7 200907876 關聯；—輪人步驟，供輸人對應特質測驗題目 =合貧料；以及-分析步驟’供分析作答雜以即供 一數據。 、去肖’係提供一種特質測評方法。本發明之方 二：步驟’係根據一評估參數，自複數個測驗題目 入測驗題目，此評估參數係與-特質相關聯：-輪 s作=對=驗:;之π r; 一估計步驟， 料即時分析^數據憤貝，以及1析步驟，以利用作答資 【實施方式】 定特質關聯之評估錢數個麟題目，根據與特 題目，並根ί針二爻二複數個測驗題目中選出特質測驗 特定特質 働1目的作㈣料’估計受測者之該 發明才之範例描述本 他特質測評之朗。 技藝者當知，本發明亦適用於其 0 一造題裝置102、一評估梦w 1Λ/Ι寺貝測吁糸、、先10可包含 儲存裝置110。其中，造題用者介面⑽以及-目；評估裳置104係負責針^所』負責產生複數個測驗題 出特質測驗題目，並^㈣從職題目中選 質；使用者介面⑽= 系^評估受測者之特定特 儲存裝置η"系負貴儲存手測f輸入資料；以及 8 200907876 式作電性連接、無線的方式作通訊、或熟此技藝者所知的其 連接方式與儲存裝置110相連接；造題裝置102與評估^置 104亦可以有線或無線的方式直接通訊、或僅透過儲存^ 110間接相連。而使用者介面108亦可根據系統管理者或^ 者所在位置，以本區或遠端連線之方式，與造題裝置1〇2^1 平 估裝置104及儲存裝置no作溝通互動。 °、 本發明之造題裝置102係可以電腦軟體、硬體、或其結合 者實施。在本發明之較佳實施例中，造題裝置1〇2可包 目產生模組1021、一篩選模組1022、以及一預試模組1〇23。 如上述，造題裝置1〇2係負責產生複數個測驗題目，而測驗 目之產生，係可透過題目產生模組1021將至少一第一^材及 至^一第二素材排列組合產生之；篩選模組1〇22可在題 生杈組1021所產生之所有的測驗題目中，經過質化與量化 ί序，選出符合一預設標準之測驗題目；預試模組1〇、23則可 2T^： 1022 測驗題目進行預先測試，以分析出測驗題目 此，造題裝置⑽得以產生對特定特質具意義 及一具第體而 組1021係透過排列組合一第-素材 中，ί一主複數個測驗題目。在-較佳實施例 ΐ弟二素材係可針對狀企業產生之。舉例而 άΐ後’其内容分析結果可為許多試題 問將此等試』型== )反應選項（即第二素材）。情境題型（即第 方’係可以描述在公司遇顺突情境的 據自己的ί 權值’供作答者根 疋1 丁併汴稭此，碭目產生模組1021可將情境 200907876 題型（即第-素材）及反應選項（即第 進而產生複數個測驗題目。 謂）進仃排列組合 反岸選項（即第it 情境（即第—素材）與 中于群組歸類的動作，亦即情境群組 谷了刀柄—類，而反應群組中的選項亦可為同類型。 產ίί戶=ίίί 1022可進一步對題目產生模組1021所 :產作錄合，仍有 ，」等;==事=質=麼中處;; 除，僅剩^或酬，將不合邏輯的試題刪 1023 -同描述。〃的又°式者反應數據，因此將與預示模組 中戶賊模組聰可糊篩選模組在質化程庠 出==預先測試，以產生作答 之料ϋ丁施測，以收集在施測過程中預試人員所產生 作°貝枓與例如反應時間等其他相關資料。 接種!_直 收集到的各種資料係可傳遞ί例2 庫，再由筛賴組擷取侧^ 在師顧組體之量化程序中，項目分析係可分別根據古 200907876 典測驗理論與適性測驗理論進行，以產生各種冑目參數， ίί信度、效ΐ、試題鑑別度、試題難度、描述性資訊等來數。 J ^之’ f據·ί典測驗理論的項目分析可分析預試題組的平 ^(Mean)、標準差⑽、項目—測驗的相關（⑺⑽㈣ )tem-total coir—)、偏態係數(skewness)、峰度 (Kurtosis)與因素負荷量（factQr丨祕呢)等參數，依據洛代 論析可分析試題題組的難度、鑑別度、訊:量 專f數。在本發明之一實施例中，選模！且1022亦可在量化 程序中判斷各題目參數之臨界值(eHteri 模組_使狀特質測驗題目。 仏t、選通 =列而言’假設創意職能特質經由題目產生模組聰 10^固測驗題目，第—次先透過篩選模組體找出内容邏輯 广的300個測驗題目（即預試題組），針對此300個測驗題 ，此專作答-貝料及反應時間，薛選模組進行項目分析產 ，例如信度之題目參數’根據此題目參數，，選模組1022更 I選擇性地從先前的_侧麟目（即職題組）_，選出符 二題目參數(例如信度超過0.7)的100個測驗題目。而 这loo個測驗題目就是可供選題模組1041選擇的正式題庫， ^稱為特貝測驗題目。此等特質測驗題目即可供評估裝置 使用，以估計出受測者精確的創意職能特質強度。 明之評估裝置1Q4係可喊腦軟體、硬體、或其結合 了 。在本發明之較佳實施例中，本發明之評估裝置1〇4係 ^貝針對所欲評估之特質，從測驗題目中選出-特質測驗題 社每^根據其作答資料，評估施測者之特質，已如上述。在較 3 中，評估裝4 104係選擇已經筛選並作過預試的題目 垃測驗題目。但在另一實施例中，評估裝置104亦可直 牛=擇^目產生模組1021所排列出之測驗題目、或者僅進一 _過筛選模組1〇22作過特定質化篩選的測驗題目作為特質 11 200907876 測驗題目’其亦不影響本發明之實施。 在本發明之較佳實施例中，評估裝置 1041以及—估計模組1042。其中，選題模组『了選題模組 估參數’自該複數個測驗題目帽 '質 係根據-評 目之一作答資料，估計施測者之特質。” +應測驗題 在本發明之一實施例中，例如人力 f定職缺所需之人袼特質，選擇欲進行評：之泉童官可先依照 數係與所需之人格特質相關聯。舉 此評估參 不限於常模參數、題庫參數、向度參數可包含但 =則程序她、文祕絲數ϋ=參數、 者所知的其他參數等等不同性質的參^謂出參數或熟此技藝 除了選題模組1〇41内所包含之預設參數 7包“列如造題裝置102中之筛選模租1〇22卜，參數亦 數及題目參數料。換句話說 之量化參 不僅包含評估裝置104内的參數，；的評估參數 ΐί:ίΐ之二數。ΐϊ每-個評估參數“ i〇!:i 質的：因響-特以 質所需之特質測驗題目。 更精丰的選擇評估特定特 在一較佳實施例中，選題模組1〇41 砰估參數，從經過造題裝置102中 ;據上述任何一種 組助的質化、預試、量化程1022及預試模 題庫中，選出適當的特質測驗題目=質測^目/正式 可以最大訊息法自複數個特質測驗題D二，選題模組1041 受測者之態度估計值具有庫中，選取對 1〇41亦可以其他熟此技藝者所二心5^然=，題模組 計模組1042評估特定特質強度的特Ϊ測驗題目選出仔以供估 12 200907876 在一較佳實施例中，選出一特 _可先自複數個特質測驗題目之前，選題模組 題目，並根據對應之作答資料 中’選出^個測驗 體而言，選題模組所、登山斷疋否發出一警告訊息。具 測量之特質的態度尺度的兩驗題目、，係對應到欲 選題模組1041可比對受測者名μ 喊，作為測驗起始點。 相同，即表示應徵者未專心作欠題之反應，若反應 組_可再從正式題庫中挑息外，選題模 若應徵者仍未專心回答，即線 ^極端试題進行施測， 題模組1041可提供一驗制二話說，本發明之選 有效度及可信度。 以確保文測者之作答資料之 本發明之估計模組1042，伤椒诚^ 答資料，估計受測者之特質，^應2測驗題目之-作 者輸入對應特質測驗題目的作又資t言，每當受測 =。然而，其他航賴麵知賴纽料料適用^ 測 標 在一較佳實施例中，估計模組1〇42 =終止條件之酬數據，例如 ς 炊 反之 :終S題模組再選出另一個特質= 估ί實施例巾’料模組1G42亦可僅貞責計算中 f並將估計標準差傳給一判斷模組(未圖示 :核、、且(未圖示)決定是否達到測驗終 由 r止的條件，則判斷模組(未圖示）即終止測^，^達測 目拉、、且(棚示)使選題模組_繼續選出下—個待_以 13 200907876 參照第一圖’本發明之特質測評裝置 硬體裝置’例如資料庫及/或硬碟機 驗題目之答荦。再者，桉六骷要⑴十了為又剩者回應特質測 料。舉“ί，存受測者之背景資 之資料。舉例而言，管理者資料除了管理去=存系統官理者 這些資訊包括但不限於反應時間 題傳輸時間、受測狀態等資料。儲存 、斤在位址、試 測驗題目。 置110亦可儲存複數個 在一較佳實施例令，館存裝置 模組1021所產生之測驗題目之外，^ ^儲存題目產生 質化與量化後_驗題目，=過篩選模組 賴分析的測驗題目及其題目參數。，。過預物組⑽3進行 然—實體的儲存裝置。 性及/或__互連接為，%^實/= 雜置，透過電 作有線或無線之連接。 /、、置102、砰估裝置1〇4 更可包含-管理 熟此技藝者所知的其他連齡式射訊連接、或 裝置106與造題裝置丨02及評估 、0相連接。管理 線的方式直接通訊、或僅透過儲有線或無 置1〇6與使用者介面108之間。管理裂 以本區或遠端連線之方式作溝通互^。’、、’’官理者所在位置， 14 200907876 在—較佳實施例中，管 腿，供分析作答資料 可為信度、效度、，此數據亦 質的評量；了某-Γ系統1G進行特定特 了解這套測評系統1〇目前最新的常二管J要 瞭解景^^進行分析，立即 與相關職能模型的參數·據的〃，振本身讀的信效度 =第-圖所示’本發明之管理I置⑽亦可 字ΐ^Γ ί尋模組可供搜尋施測日期、姓t 、ϊί、專業類別、卫作資歷、廠區、各項 2;料使得管理者得以更容易地使用儲存裝置 no 然而二需要說明的是’第-圖以虛線方式纷示管 106 ’係代表在另一實施例中，本發明之特質 ^ 不包含管JE里農置1〇6 ’而僅分別具有獨立運作之分析模組二 及/或搜尋模組1062。具體而言，特質測評系統1〇中、、’ 模tri1可直接與儲存裝置iig通訊，以取得進行I析二ΐ 之作合資料及受測者的其他背景資料，進而提供 而 好比常模術據。而搜尋模組1〇62亦可直接與儲存裝置η 以供搜尋儲存裝置110中的各種資料。在一較佳實施例 中，本發明之特質測評系統10亦可包含—使 供管理者輪入第-素材以及第二素材。再者使 15 200907876 I供管理者輸入欲供搜尋模組1〇62搜尋之作答資料及管理者 資料，例如施測曰期、姓名、身份証字號、姓別、年齡、專 類別、工作資歷、廠區、各項特質分數的百分比、各項特質又 ΪίΪ量分數、各項特質分數的作答信度，以及評量風格等刀， 在另一方面，使用者介面108亦可供受測者輸入 及/或其背景資料及/或供預試人員輸入預試資料 &據不 =用者’使用者介面108可作適當的調整 在=知 =是，本發明較佳適用於客戶端/飼服器之架構。 在實鉍例中，題目產生模組1021在伺服器 題目，呈現於你Li存裝0中，立即選出一特質測驗 108輸入作答資料測者並接著透過使用者介面 裝置110) ^估古_ 亦可儲存於資料料即儲存 作答資料，估計受^者之可t據^應特質測驗題目之 »受測者之作答資料，莫二進—步 準差是否達_齡μ心姑的特質、計算估計標 題模組1(Μ1選出另、—個:上牛:若結果未達終止標準，則使選 驗終止條件為止出 特貝測驗題目，直到作答資料達到測 估計標準差，並將 十模組1042亦可僅負責計算出 模组(未圖示由;傳f飼服器端（未圖示）之—判斷 的條件，若已達到測驗故ϋ f不）決定是否達到測驗終止 、驗，ς止的條件’則判斷模組(未圖示）即終 16 200907876 測Z模組(未圖示)使選題模組_繼續選 佳f施中，舰器端（未圖示）可具有一分析模組， 之作答資料，並即時提供一常模數據。然而，在 圖、ίίί 此數據亦可為信度、效度、模數及比較表On the same day; replacing the shortcomings of Wei New's light question bank, and setting and time, making the construction of the test bank more effective; Big _ labor cost, and can save the target ======, side tender-species; according to a series of silk material, and &,; to generate a number of test questions to select the characteristics of the test, the purpose The evaluation parameters of the pings, from the plurality of test questions, the data of the refusal of the analytic object, estimated 5 200907876 [invention content] Include the one-on-one evaluation system. The system of the present invention=W device has a -title generation module, which combines at least a dream system = a second material system to generate a plurality of test questions; - stores a second broadcast, stores the plurality of test questions; and - evaluates The device, and there is - one of the number of test questions selected from the hard test, the evaluation of the estimated two or two" estimated module based on one of the corresponding trait test questions;; f package of a total trait evaluation system. The device of the present invention has a title generating module for arranging at least one flute-table and one second material to generate a plurality of test questions: a module and an estimation module, and a selection module According to an ίί』, with ί: topic selection;: test: two of the selected - trait test questions, this evaluation parameter ii one of the corresponding trait test questions to answer the information, estimate lie, user; In addition, one of the input trait test questions is used as a data and an analysis module to analyze the data of one port, seven Μ = Ming, and provide a trait evaluation system. This test questions, this evaluation of the husband ^, B characteristics for answering quality! = ^ title installed this invention 27 face to create the title, the first material and at least the production of f module, will at least - the eye; And to generate a plurality of test questions and groups for selecting test questions that meet a predetermined standard. 6 200907876 A further aspect of the present invention is to provide an evaluation device. The evaluation device of the present invention comprises: a question and a system evaluation answering data; and a = 2: two questions - a data, an estimated trait. One of the aspects of the present invention is to provide a rationale. The tube county of the present invention comprises: a pipe of the 3D system; and an analysis module for analyzing the shell = library '1, storing a plurality of pieces of the re-surface of the invention.二2; π server contains: - title generation module, ic head; - selection of the pool, miscellaneous - ping mvtt a number of test questions to answer the information, estimated 'quality; - guest: on the through - The network 'provides the corresponding traits. The problem is based on the various methods of the sequel, ί provides a trait evaluation method. The method of the present invention is to combine at least one first material and at least one phase to generate a plurality of test questions; - a topic selection step, a title i: quality = two, step into , enter the corresponding trait test traits.丨# and - Estimation steps are based on the data of the answer, estimating the IF of the IF, and providing a trait evaluation method. The method of the present invention is to select at least one first material and at least one first-sound wire> number from a plurality of test questions - the trait test question, the evaluation ^ 7 200907876 association; - the round step, for the input trait test question = the poor material; and - the analysis step for the analysis to make a data. Go to Xiao's Department to provide a method of trait evaluation. The second aspect of the present invention: the step of selecting a test subject from a plurality of test questions according to an evaluation parameter, the evaluation parameter is associated with the trait: - wheel s = = pair = test: π r; , real-time analysis ^ data anger, and a step of analysis, in order to use the answer [implementation] to determine the traits of the evaluation of the number of money topics, according to the special topic, and the roots of the two test questions Select the trait test specific traits 目的 1 purpose (four) material 'estimate the subject of the invention's example description of the traits. The skilled artisan will appreciate that the present invention is also applicable to a storage device 102, an assessment dream, and a storage device 110. Among them, the user interface (10) and the head of the question-making; the evaluation of the 104-series responsible for the needle is responsible for generating a plurality of test questions, and ^ (4) selecting the quality of the job; user interface (10) = system ^ Evaluating the specific storage device of the subject η" is storing the manual input f input data; and 8 200907876 is electrically connected, wirelessly communicated, or connected and stored by a person skilled in the art. The 110-phase connection; the question-making device 102 and the evaluation device 104 can also be directly communicated by wire or wirelessly, or directly connected through the storage device 110. The user interface 108 can also communicate with the problem-solving device 1〇2^1 evaluation device 104 and the storage device no according to the location of the system administrator or the user in the local area or the remote connection. The subject-solving device 102 of the present invention can be implemented by computer software, hardware, or a combination thereof. In the preferred embodiment of the present invention, the questioning device 1〇2 may include an application module 1021, a screening module 1022, and a pre-test module 1〇23. As described above, the problem-solving device 1〇2 is responsible for generating a plurality of test questions, and the test object is generated by combining the at least one first material and the second material through the title generating module 1021; The module 1〇22 can select the test questions that meet the preset criteria in all the test questions generated by the test group 1021; the pre-test modules 1〇 and 23 can be 2T. ^: 1022 The test questions are pre-tested to analyze the test questions. The question-making device (10) is able to produce a meaning for a particular trait and a first body. The group 1021 is a combination of a first-material, ί a primary number Test questions. In the preferred embodiment, the two materials can be produced for the enterprise. For example, the results of the content analysis can be a lot of questions for the test type == ) reaction option (ie the second material). The situational type (ie, the first party's system can describe the company's situation in which the company's situation is based on its own weight value' is for the respondent's roots.) The project generation module 1021 can set the situation 200907876 ( That is, the first material and the reaction option (that is, the first to generate a plurality of test questions.) is to arrange the combination of the reverse shore option (ie, the first context (ie, the material) and the action in the group classification, that is, The situation group has a handle-class, and the options in the reaction group can be the same type. Product ίί户= ίίί 1022 can further generate the module 1021 for the title: production, still, "etc; ==事=质=么中处;; In addition, only ^ or remuneration, the illogical test will be deleted 1023 - the same description. 〃 又 ° ° response data, so will be with the predictive module in the thief module The Congbu paste screening module is tested in the quality process == pre-test to generate the answering material, to collect the other factors related to the reaction time, such as reaction time, produced by the pre-test personnel during the test. Information. Inoculation! _ Collection of various data can be transmitted ί 2 2 library, and then by the screening group Take the side ^ In the quantitative program of the teacher group, the project analysis department can be based on the ancient 200907876 code test theory and the fitness test theory to generate various eye-catching parameters, ίί reliability, effectiveness, test case discrimination, test difficulty Descriptive information, etc. J ^ 'f project analysis of the project theory can analyze the level of the pre-test group (Mean), standard deviation (10), project-test related ((7) (10) (four)) tem-total coir -), skewness, kurtosis and factor load (factQr), according to the analysis of the difficulty, discrimination, and information of the analytic test group. In an embodiment of the present invention, the mode selection! and 1022 can also determine the critical value of each parameter parameter in the quantization program (eHteri module_make trait test topic. 仏t, strobe=column] hypothesis The characteristics of the creative function are generated by the module Cong 10 ^ solid test questions, the first time through the screening module to find 300 test questions (ie, the pre-test group) with a wide range of content, for this 300 test questions, this special Answer - shell material and reaction time, Xue The module performs project analysis and production, for example, the parameter parameter of the reliability 'According to the parameter of the topic, the selection module 1022 is more selectively selected from the previous _ _ _ _ _ _ _ _ _ _ _ _ _ For example, 100 test questions with a reliability of more than 0.7), and the loo test questions are the official test questions that can be selected by the selected question module 1041, which is called the Tebe test question. These trait test questions are available for the evaluation device. To estimate the precise creative function trait intensity of the subject. The evaluation device 1Q4 can call brain software, hardware, or a combination thereof. In a preferred embodiment of the present invention, the evaluation device of the present invention 1〇4 The system is selected from the test questions for the characteristics to be evaluated. The trait test society evaluates the characteristics of the tester according to the data of the test. In comparison 3, the evaluation equipment 4 104 selects the topic that has been screened and pre-tested. However, in another embodiment, the evaluation device 104 can also perform the test of the test questions arranged by the module 1021 or the test of the specific quality filter. The topic as a trait 11 200907876 test title 'it also does not affect the implementation of the invention. In a preferred embodiment of the invention, the evaluation device 1041 and the estimation module 1042 are provided. Among them, the topic selection module "the parameter selection module evaluation parameters" from the plurality of test questions caps 'quality system according to one of the evaluation materials, to estimate the characteristics of the tester. "The test question is in one embodiment of the present invention, such as the human traits required for the appointment of a manpower, and the choice is to be evaluated: the spring child officer may first associate with the required personality trait according to the number system. This evaluation is not limited to the norm parameter, the question bank parameter, the directional parameter can include but the program she, the secret number 参数 = parameter, other parameters known, and so on. The skill is in addition to the preset parameter 7 package included in the topic selection module 1〇41, which is listed as the screening module rent in the problem-solving device 102, and the parameter number and the parameter parameter. In other words, the quantization parameter includes not only the parameters in the evaluation device 104, but also the evaluation parameters ΐί:ίΐ. ΐϊEach evaluation parameter “i〇!:i quality: due to the quality of the quality test questions required. More sophisticated selection evaluation specific in a preferred embodiment, the selection module 1〇 41 砰 参数 参数 , , , 砰 砰 砰 砰 砰 砰 砰 砰 砰 砰 砰 砰 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据 据The maximum message method is a plurality of trait test questions D2, and the topic module 1041 is estimated to have an attitude estimate. The selected pair of 1〇41 can also be used by other skilled artists. 5^然=, The group meter module 1042 evaluates the feature test of the specific trait intensity for evaluation. 12 200907876 In a preferred embodiment, a special _ can be selected before the plurality of trait test questions, and the topic of the module is selected. According to the corresponding answering data, in the case of 'selecting ^ test body, whether the selected module and the mountaineering block will issue a warning message. The two test questions with the attitude scale of the measurement are corresponding to the selected model. Group 1041 can be compared to the subject name μ, as the test start The same, that is, the applicant does not concentrate on the reaction of the problem, if the reaction group _ can then provoke the interest from the formal question bank, if the candidate model is still not focused on the answer, that is, the line ^ extreme test questions are carried out, The problem module 1041 can provide a verification statement, the selection validity and the credibility of the present invention. The estimation module 1042 of the present invention for ensuring the data of the textual reviewer, the data is estimated to be tested. The characteristics of the person, ^ should 2 test questions - the author enters the corresponding trait test questions and the words, whenever the test =. However, other Navigators know that the material is applicable ^ test in a better implementation In the example, the estimated module 1〇42=the termination condition data, for example, ς 炊 verse: the final S-question module selects another trait = the estimated embodiment towel module 1G42 can also only be responsible for the calculation f And the estimated standard deviation is transmitted to a judgment module (not shown: the core, and (not shown) determines whether the condition of the test end is reached, the judgment module (not shown) terminates the test, ^达测目拉,, and (sheld) make the topic selection module _ continue to select the next - awaiting _ to 13 200907876 The first picture 'the traits evaluation device hardware device of the present invention' such as the database and/or the hard disk machine test questions. In addition, the six 骷 骷 (1) ten for the remaining ones to respond to the traits. ί, the information of the background of the subject. For example, the manager's data is not only limited to the management system, but is not limited to the transmission time, the status of the test, etc. Address, test question. The device 110 can also store a plurality of test questions generated by the library device 1021 in a preferred embodiment, and the quality of the stored problem is quantified and quantified. = Test questions and their title parameters of the screening module. ,. Through the pre-set (10) 3 - the physical storage device. Sex and / or __ interconnection is, % ^ real / = miscellaneous, through the wired or wireless connection. /,, 102, evaluation device 1〇4 may include-management other connection-type radio connections known to those skilled in the art, or device 106 connected to the problem-making device 丨02 and evaluation, 0. The way of managing the line is directly communicated, or only through the storage cable or between the user interface 108 and the user interface 108. Management splits by means of local or remote connections. ',, ''The location of the official, 14 200907876 In the preferred embodiment, the tube legs for analysis and answering information can be reliability, validity, and the quality of this data; 1G to carry out specific special understanding of this set of evaluation system 1 〇 current the latest second tube J to understand the scene ^ ^ analysis, immediately with the relevant functional model parameters · data, the vibration of the reading itself = the first map The management of the present invention (10) can also be used to search for the date, the last name t, the ϊί, the professional category, the health qualifications, the factory area, and the various items; It is easy to use the storage device. However, it should be noted that the 'the first figure shows the tube 106' in a dotted line. In another embodiment, the trait of the present invention does not include the tube JE. There are only independent analysis modules 2 and/or search modules 1062. Specifically, the traits evaluation system 1 〇 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , according to. The search module 1〇62 can also be directly connected to the storage device η for searching various materials in the storage device 110. In a preferred embodiment, the traits evaluation system 10 of the present invention may also include - enabling the manager to wheel into the first material and the second material. In addition, 15 200907876 I is used by the administrator to input the answer data and manager information for the search module 1〇62 search, such as the test period, name, ID number, surname, age, special category, work qualification, The factory area, the percentage of each trait score, the traits of each trait, the score of each trait, the assessment of the traits, and the assessment style. On the other hand, the user interface 108 is also available for the subject to input and / or its background information and / or for the pre-test personnel to enter the pre-test data & not = user 'user interface 108 can be adjusted appropriately = know = yes, the present invention is preferably applied to the client / feeding service The architecture of the device. In the actual example, the title generation module 1021 is presented in the server directory of your account, and immediately selects a trait test 108 to input the data tester and then passes through the user interface device 110). It can be stored in the data material, which is stored as an answering data. It is estimated that the subject can be judged according to the trait test of the subject, and whether the grading difference is _ age μ gu gu Estimate the title module 1 (Μ1 select another, one: upper cow: if the result does not reach the termination criteria, then the special test termination condition will be issued until the test result reaches the estimated standard deviation, and the ten modules will be 1042 may also only be responsible for calculating the conditions of the module (not shown; passing the feeding device end (not shown) - if the test has been reached, ϋ f does not) decide whether to end the test, check, ς The condition of the condition is judged by the module (not shown), which is the final 16 200907876. The Z module (not shown) makes the selection module _ continue to be selected, and the ship end (not shown) can have one Analyze the module, answer the information, and provide a normal model data immediately. However, in , Ίίί this data may also be reliability, validity, and comparison table module
趟^者/在一較佳實施例中’飼服器端（未圖示）可具有-f哥輪，组，供管理者搜尋儲存裝置110中之各種資料，苴包含 ，不限f㈣日期、姓名、身餘字號、姓別、i二專S 的評量分數、各鄕質分_作答域，以及評量風ϋ t較佳實知例中，飼服器端（未圖示）可具有一 管理者搜尋管理。 ”俯辦縣置110中，供 用雖述t施例以客戶端/飼服器架構描述本發明之應 μ m'者皆知，本發明之特質測評系統w亦可^ 於早一糸統中，受測者仍可透過使用者丌了只她 行特質評估。因此，本發明並不受此架構之限^糸統本區進 發=二質測評方法之流程圖。本 合’以產生複數侧材排列組 數’從這複數個測驗題目中選出特質測^題目目=之評估參 質測驗題目的作答資料，估計㈣據對應特 17 200907876 用ίίΠ?用Γ包含?題裝置102、-評估裝置⑽、一使 止儲存裝置110之特質測評系、统10，4 2;無線的方式作通訊、或熟此技藝者所ϊ的置他 ίϊϊΐ 110相連接；造題裝置102與評估裝錢4亦可t ======僅透過儲存裝置丨關接相i 太1 I根據糸統管理者或受測者所在位置，以 =^端連狀方式，與造概置1G2、評估裝置服置^ 作溝通互動。雖然以下將以基於職能甄選 本方法之各步驟’細，熟此技藝者當知，Ϊ發Ϊ 亦適用於其他特質測評之應用。 ％月 法^第示，本方法開始於步驟2〇〇。於步驟202，本方 tii題目產生模組贈對第一素材及第二素材進^ ίϊΐΐ一較Ϊ實施例中’第一素材及第二素材係可針3 刀/ ΐΓ"為許多試題_ ’依該企業實際工作中發 ^曰’或與現實生活相似醇件綱題為設計依據，則 型ί換成情境題型（即第—素材）及反應選項（即第二 述在公司，又可稱為題幹，係可以描 J遇至J衝大Jfi兄的方式呈現，而四個反應選項 又门^ ’v又可稱為項’係可為各有辦的解決方案，各有 答者根據自己的喜好進行排序。藉以= 笛_!妾=、於步驟204，題目產生模組1021可將情境題型（即 及反應選項（即第二素材）進行排列組合，進而產 =數^^題目。在—較佳實施例中，本發明之 〇22，可於步驟施對題目產生模組體 題目進行質简選，謂岭合—徽鮮之猶題所目有膽 18 200907876 當-牛，即使透過上述題目產生模組騰將第一素材及 12ΪΪ歸類後再作排列組合，仍有可能產“如「、當同 男畔请產假時’你會怎麼處理其待辦事項？」等不合 Ϊ 選模組職可建立各種標準及/或規 貝J將不合邏輯的试題刪除，僅剩下符合邏輯的測驗題目。 模观’本㈣之賊触係可利用筛選 、、甘 貝化過之測驗題目進行預先測試，以產生作夂資料 r 目Γί料。具體而言，預試模組簡係將篩選模‘1〇22 終 （即步驟綱）所篩選出之測驗題目（即符合邏 所產Γ夕H提供給預試人員進行施測，以收集預試人員 斤產生之作答>料與例如反應時間等其他相關資料。 在只鉍例中，預試模組1023所收集到的各種資料係可直 ^餘給ί選模組，進行量化的項目分析。在另-實施 例中’。預試模組1023所收集到的各種資料係可傳遞至例如一 飼服器資料庫，再由篩職組體擷取所需資料，進化 的項目分析。 接严’於步驟210，篩選模組1〇22可分別根據例如古典測 驗理論與適性測驗理論進行項目分析，以產生各種題目參數， 其可包含信j、效度、試題鑑別度、試題難度、描述性資訊等 參數。具體f之’依據古典測驗理論的項目分析可分析預試題 組的平均數(Mean)、標準差（幼）、項目_測驗的相關（c〇rrected item-total correiation)、偏態係數(skewness)、峰度係數 (Kurtosis)與因素負荷量（facrt〇r i〇a(^ng)等參數，依據當代 ，驗理論的項目分析可分析試題題組的難度、鑑別度、訊息量 等參數。在本發明之一實施例中，篩選模組1〇22亦可在量化 程序中（即步驟210)判斷各題目參數之臨界值(criteri〇n)， 以決定供選題模組1041使用之特質測驗題目。舉例而言，假 設創意職能特質經由題目產生模組丨021產生丨〇〇〇個測試題目 19 200907876 一次先透過篩職組皿2找出内容邏輯無誤 測驗韻目目（即預試題組）（步驟2〇6)，針對此3〇0個 f i 1(i23(^^2G8)收細騎應之作答資 進行項^心吝1據此t作答資料及反應時間’筛選模組ι〇22 選模组in^#-r、例如信度之題目參數’根據此題目參數，篩 组）、中，選出节性地從先前的300個測驗題目（即預試題 個iLr目tit)的(例如信度超過〇.7)的⑽ 组1041選摆的τ 4而纪00個測驗題目就是可供選題模 驗題目則可供^為特質測驗題目。此等特質測 職能特質強度。、 使用’以估計出受測者精顧創意 接著，於步驟212,在一輕祛眘故加士 ^ 先自複數個特質測驗題目題實=中選出選可 具體而言，選題模組104〗所、s山單T k出一個測驗碭目。 欲測量之特質的離戶尺产的1^出之兩個测驗題目，係對應到 點。 尺度的兩個極端賴題，作為測驗起始 資料。+在-較^圭中驟透過使用者介面1 〇8輸入作答 答;料 實_中，於步驟216由儲存裝置110儲存作 爾後’於步驟218，選題槿細1n/M _ ^ 端試題之反應，判斷受測者是否 °比^測者在此兩極 示受測者未專心作答，則於牛=乍。。若反應相同，即表 步驟212，由選題^ 1〇4= 告訊息，並再回到 試題進行施測。若於步驟218m庫中挑選另外兩題極端 專心作答之次數超過一既定次判斷受測者無 終止施測，本方法即於步驟2 碭杈組1041可選擇性 題模組可提供-驗證機制二話說’本發明之選 有效度及可信度。 確保受測者之作答資料之 20 200907876 接著來到步驟222 ^ff示受測者有專心作答，則本方法 複數個測驗題目中2 :選Z組1041根據-評估參數，自該 質相關聯。^―特質測驗題目，此評估參數係與-4； 管，例如人力資源部之主 數，此評估參數係盘所需而之°寺所貝’選擇欲進行評估之參 :題心數聯向=言算= 參數或熟频藝麵㈣表輸出 可評估參數亦 不僅包含評=10=:數選題，内的評估參數 模組所產生之參數。：内每的一參個%==;置，各 數，亦即，選擇任何-個評估參^4影ί:;ϊί 質所需之特質測驗ΐΐ 件以更精準的選擇評估特定特 222} ι〇4ΐ 選模組及f=中之筛 =測驗題目/正式題庫中(步驟204-步驟2V=In a preferred embodiment, the 'feeder end (not shown) may have a -f wheel, a group for the manager to search for various materials in the storage device 110, including, not limited to f (four) date, The name, the body size, the surname, the evaluation score of the i-second S, the quality scores, the evaluation field, and the evaluation method. In the preferred example, the feeding device end (not shown) may have An administrator search management. In the case of the use of the client/feeder architecture to describe the invention, it is known that the trait evaluation system of the present invention can also be used in the early days. The subject can still pass the user's evaluation of her traits. Therefore, the present invention is not subject to the limitations of this framework. The present invention is a flow chart of the second-class evaluation method. Arranging the number of groups 'Selecting the traits of the test questions from the plurality of test questions = the answer data of the evaluation test questions, and estimating (4) According to the corresponding special 17 200907876 ί Γ Γ Γ 102 102 102 102 102 102 102 、 、 、 、 、 The traits evaluation system of the storage device 110, the system 10, 4 2; the wireless way for communication, or the ϊ ϊϊΐ 110 connected by the skilled person; the title device 102 and the evaluation money 4 can also t ====== Only through the storage device to shut down the phase i too 1 I according to the location of the system manager or the subject, in the form of = ^ end, with the establishment of 1G2, evaluation device service ^ Communicate and interact. Although the following steps will be based on the function selection method, The artist knows that the hairpin is also suitable for other traits evaluation applications. %月法^第示, the method begins in step 2〇〇. In step 202, the party tii title generation module gives the first material and The second material is in the middle of the example. The first material and the second material can be 3 knives / ΐΓ" for many questions _ 'according to the actual work of the enterprise The outline is the design basis, then the type ί is replaced by the situational type (ie the first material) and the reaction option (that is, the second is in the company, which can also be called the stem, and the system can describe the J. The way to present, and the four reaction options are closed ^ 'v can also be called item ' can be a solution for each, each respondent sorts according to their own preferences. By = flute _! 妾 =, in the steps 204. The title generation module 1021 can arrange and combine the contextual problem types (ie, the response options (ie, the second material), and then generate the number of questions. In the preferred embodiment, the file 22 of the present invention can be In the step of applying the title of the module to make a simple selection of the title, said Linghe - Hui Xianzhi The eyes are bold 18 200907876 When - cow, even if the first material and 12 ΪΪ are sorted and combined after the above-mentioned problems are generated, it is still possible to produce "such as "When you are on the same date, please take maternity leave" How to deal with its to-do list?" If you don't have the right choice, you can create various standards and/or rules to delete the illogical test questions, leaving only the logical test questions. The system can be pre-tested by screening and gamma-tested test questions to generate 夂 r 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 The selected test questions (that is, the calibration test is provided to the pre-test personnel to perform the test to collect the pre-test personnel's response) and other relevant information such as reaction time. In the example only, the various data collected by the pre-test module 1023 can be directly used to select modules for quantitative project analysis. In another embodiment. The various data collected by the pre-test module 1023 can be passed to, for example, a feed server database, and the required data can be retrieved from the screening team for evolutionary project analysis. In step 210, the screening module 1 22 can perform project analysis according to, for example, classical test theory and fitness test theory, to generate various topic parameters, which can include letter j, validity, test title discrimination, test difficulty, Descriptive information and other parameters. Specific f's project analysis based on classical test theory can analyze the average number of pre-test groups (Mean), standard deviation (young), project_test correlation (c〇rrected item-total correiation), skewness coefficient (skewness) , Kurtosis coefficient (Kurtosis) and factor load (facrt〇ri〇a (^ng) and other parameters, according to the contemporary, the theory of the project analysis can analyze the difficulty, discrimination, message volume and other parameters of the test group. In an embodiment of the invention, the screening module 1 22 may also determine a critical value (criteri〇n) of each parameter parameter in the quantization program (ie, step 210) to determine the trait test topic used by the selection module 1041. For example, suppose the creative function traits generate a test question via the title generation module 丨 021. 19 200907876 First, through the screening of the class 2 to find out the content logic error test (the pre-test group) ( Step 2〇6), for this 3〇0 fi 1(i23(^^2G8) fine rider should be responsible for the fundraising item ^心吝1 according to this t answer data and reaction time 'screening module ι〇22 Select the module in^#-r, for example, the reliability parameter parameter 'according to The title parameter, the screen group, and the selected τ 4 of the (10) group 1041 of the previous 300 test questions (ie, the pre-test title iLr target tit) (eg, the reliability exceeds 〇.7) 00 test questions are available for the selected test questions to be used for the trait test questions. These traits measure the trait strength. Use 'to estimate the testee's enthusiasm for creativity. Then, at step 212, in a light祛 故 故 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ The two test questions of the ruler's production are corresponding to the point. The two extremes of the scale are used as the test start data. + In the case of -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ In the actual operation, in step 216, the storage device 110 stores the result in step 218, and the reaction of the selected question 1n/M _ ^ end test questions determines whether the subject is measured at the two poles. If you don't concentrate on answering, then it is in the cow = 乍. If the reaction is the same, that is, step 212, by the topic ^ 1〇4= message And then return to the test questions to perform the test. If the number of the other two questions selected in the 218m library is more than one predetermined time to judge the test subject without termination test, the method is selected in step 2 砀杈 group 1041 The question module can provide - verification mechanism to say 'the effectiveness and credibility of the invention. Ensure the respondent's answer to the information 20 200907876 Then go to step 222 ^ff shows that the subject has a special answer, then this Method 2 of the plurality of test questions: The selected Z group 1041 is associated with the quality according to the evaluation parameter. ^ ― trait test topic, this evaluation parameter is -4; tube, for example, the main number of the human resources department, this evaluation parameter is required for the tying of the temple, and the choice of the temple is selected. = statistic = parameter or familiar frequency (4) table output evaluable parameters also include not only the evaluation of the parameter parameter module, but also the parameters generated by the evaluation parameter module. : Each one of the parameters is set to %==; set, each number, that is, select any one of the evaluation parameters: ϊί Qualitative test traits to evaluate the specific 222} ι 〇4ΐ Select module and f=中之筛=test question/formal question bank (step 204-step 2V=
舉例而言，選題模組随可以最大訊息S ^之試題°然而’選題模組麗亦可以其 …此技者所知的其他方法’於步驟222，選出得以供估 私組1G42評估特定特質強度的特質測驗題目。 ' 在本發明之較佳實施例中’於步驟DA，估計模組係 21 200907876 應ίί測驗題目之—作答資料，估計受測者之特質。具 7 ^測者輸入對應特質測驗題目的作答資料時，# MLE任^+ <+例而'r，估計模、组1042可採用最大概似估計法 及估物其纖纖所知的模式 計佳實施例中，於步驟226 ’估計模組可 特定特質之評估ΪΪ。步驟228，由估計模組1042產生 僅出另:口:】，於:驟226，估計模組辦2亦可 之評估結杲，反之，、^Si ^;質=回到步驟^選題= 分= 可 =數=較=r，物^^常 質的評量，==ig進行特定特 利用使用者介面108，經過分析斤效度等數據，便可 22 200907876 瞭解公司目前各部門人員特質的分佈、測驗本身最新的信效度 與相關職能模型的參數等數據。 °又 r 接著，本方法結束於步驟232。然而，本方法亦包含許多選 擇性之步驟。舉例而言，本方法亦可包含一搜尋步驟。於此步 驟，搜尋模組1062可供搜尋施測日期、姓名、身份証字號、 =別、年齡、專業類別、工作資歷、廠區、各項特質分數的百 分比、各項特質分數的評量分數、各項特質分數的作答信度， 以及評量風格等。搜尋模組1062所提供的搜尋功能，彳^管 理者得以更容易地使用儲存裝置110所儲存之各種資料。 ^者’本方法亦可包含另一輸人步驟之供#理者透過 者二面108輸入第一素材以及第二素材。再者，輸入步驟亦可 者透過使用者介面⑽輸人欲供搜尋模組1〇62搜尋之 :::=別、工作資歷、廠區、各項特匕數t百 以及評猶、鴻嫩_答信度， 旦/Ϊί ’本發日狀輸人步驟亦可供受_輸入其其背 扣，使 ===== 存包含另一儲存步驟’供儲存裝置⑽儲 坪估，喊其他魏可作為控⑽、統與進-步 4估的依據’14些貧訊包括但不限於 适乂 址、試題傳輸時間、受測狀態等資料。^者位 資料。又，本發明之儲在卡驟^+斤屬σ卩門、應徵職位等 月之儲存步驟可包含儲存褒置〗丨0儲存複數個 23 200907876 測驗題目。 在一較佳實施例中，儲存Is 模組1021所產生之測驗題目 儲存題目產生 1022質化與量化後的測驗趨目’ 儲存:經過篩選模叙 預試分析的測驗題目及其題目參數。氣預試模組⑽3進行 在—實施例中，儲存裝置11〇 〇 然而，儲存裝置110亦可為許夕;、择早只體的儲存裝置。 性方式相互連接，及/或與同=的儲存裝置，透過電 線或無線之連接。 4裝置102、評健置1G4作有 本發明已透過以上具體實 所述者，僅棉用^ 乍评細a兄明，惟以上 能i明本發明之較佳實施例而已，並不 之均等變化與修偷耸，比/ ，本^明申明靶圍所作 内。 ’、^ 白應仍屬本發明專利涵蓋範圍 【圖式簡單說明】 第-圖係本發明之特f _系統之方塊圖 第二圖係本翻之特質測評方法之流程圖 【主要元件符號說明】 10特質測評系統 忉2造題裝置 1 〇4評估裝置 106管理裝置 24 200907876 108使用者介面 110儲存裝置 1021題目產生模組 1022篩選模組 1023預試模組 1041選題模組 1042估計模組 1061分析模組 1062搜尋模組 、申請專利範圍： L 一種特質測評系統，該系統包含： 題目 少-’係具有一題目產生模、组，該題目產生模組將至 弟素材及至少-第二素材排列組合，以產生複數個測驗 —儲存㈣’供儲存截數侧驗題目；以及 有—選賴組以及—料模組， —特質測驗目，估參數，自該儲存f置中選出 模 _挪對應該特相關聯’而該估計 碭目之一作答資料，估計該特質。 2.如請求項丨所述之 料，提供是否達丨、該估计模組更根據該作答資 j顺終场件的-_雜。 3·如請求項1所述之系 '、、先其t該儲存裝置儲存該作答資料。For example, the topic selection module can be used with the maximum message S^. However, the 'choice topic module can also be used by other methods known to the skilled person'. In step 222, the selected private group 1G42 is evaluated for specific evaluation. The trait test topic of trait strength. In the preferred embodiment of the present invention, in step DA, the estimation module system 21 200907876 should evaluate the problem--answering information to estimate the characteristics of the subject. When the test data of the corresponding trait test question is input by the tester, #MLE任^+ <+example and 'r, the estimated mode and group 1042 can adopt the most approximate estimation method and estimate the mode of the fiber. In a preferred embodiment, the evaluation of the module's specific traits is estimated at step 226'. Step 228, the estimation module 1042 generates only the other: mouth:], at: 226, the estimation module can also evaluate the balance, and vice versa, ^Si ^; quality = back to step ^ topic = Points = can = number = comparison = r, the object ^ ^ regular quality assessment, = = ig for the specific use of the user interface 108, after analysis of the efficiency and other data, you can 22 200907876 to understand the company's current departmental personnel traits The distribution, the latest reliability and validity of the test itself and the parameters of the relevant functional models. ° and r Next, the method ends at step 232. However, the method also includes a number of optional steps. For example, the method can also include a search step. In this step, the search module 1062 can search for the date, name, ID number, number, age, professional category, job qualification, factory area, percentage of each trait score, score of each trait score, The responsiveness of each trait score, as well as the assessment style. The search function provided by the search module 1062 makes it easier for the administrator to use the various materials stored by the storage device 110. ^者' This method may also include another input step for the input of the first material and the second material by the two sides 108. In addition, the input step can also be searched through the user interface (10) for the search module 1〇62 to search for:::=, work qualifications, factory area, various special t-numbers and evaluations, Hongnong _ Answer degree, Dan / Ϊί 'The daily input step of the hair is also available for input _ input its back buckle, so that ===== save contains another storage step for storage device (10) storage evaluation, shouting other Wei It can be used as the basis for the control (10), system and advance-step 4 evaluation. '14 Some of the poor news includes but is not limited to the appropriate address, test transmission time, and measured status. ^ Person position information. In addition, the storage step of the present invention stored in the card is + 斤 、 、 、 、 、 、 、 、 、 、 、 、 、 、 可 可 储存 储存 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 In a preferred embodiment, the test questions generated by the Is module 1021 are stored. The stored questions are generated. 1022 Qualified and quantized test trends ‘Storage: The test questions and their title parameters are analyzed by the screening test. The gas pre-test module (10) 3 is carried out. In the embodiment, the storage device 11 〇 〇 However, the storage device 110 may also be a stagnation device; Sexually connected to each other, and/or connected to the same storage device via wire or wireless. 4 device 102, evaluation device 1G4 for the present invention has been described above through the above specific, only cotton use ^ 乍 a a 兄 兄 , , , , , , , , , , , , , , , , , , , , , , , Changes and repairs sneak, than /, this ^ Ming stated that the target is made inside. ', ^ 白 should still be covered by the patent of the invention [Simple description of the drawing] The first figure is the block diagram of the special f _ system of the present invention. The second picture is the flow chart of the trait evaluation method of the present [main component symbol description 】 10 traits evaluation system 忉 2 problem-making device 1 〇 4 evaluation device 106 management device 24 200907876 108 user interface 110 storage device 1021 title generation module 1022 screening module 1023 pre-test module 1041 topic selection module 1042 estimation module 1061 analysis module 1062 search module, patent application scope: L A trait evaluation system, the system includes: a small problem--" has a problem-generating module, a group, the problem-generating module will be to the brother material and at least - second The materials are arranged in combination to generate a plurality of tests—storage (4) for storing the cut-off side test questions; and having the selected group and the material module, the trait test object, estimating the parameter, and selecting the mode from the storage f setting _ The opposition should be specifically related to 'and one of the estimated items to answer the information, estimate the trait. 2. If the information stated in the request item is provided, the estimated module is further based on the answer sheet. 3. The system described in claim 1 is preceded by the storage device storing the answer data.