企業(yè)人力資源管理師考試的等值研究
本文選題:測驗(yàn)等值 + 企業(yè)人力資源管理師。 參考:《魯東大學(xué)》2017年碩士論文
【摘要】:以經(jīng)典測驗(yàn)理論為依據(jù),對2015年5月和2015年11月兩次企業(yè)人力資源管理師考試(三級(jí))的理論考試部分進(jìn)行等值研究。運(yùn)用隨機(jī)等組設(shè)計(jì)與錨測驗(yàn)相結(jié)合的方式,同時(shí)使用等值標(biāo)準(zhǔn)誤法和交叉驗(yàn)證法對三種基于經(jīng)典測驗(yàn)理論的線性等值方法(即Tucker線性等值法、Levine觀察分?jǐn)?shù)法、Levine真分?jǐn)?shù)法)進(jìn)行檢驗(yàn)與比較,篩選出誤差最小且具有最強(qiáng)跨樣本穩(wěn)定性的方法。運(yùn)用最佳等值方法設(shè)計(jì)出等值分?jǐn)?shù)轉(zhuǎn)換表,進(jìn)而為未來相關(guān)領(lǐng)域的研究提供思路和參考。研究中所選擇的錨測驗(yàn)與兩個(gè)原測驗(yàn)的相關(guān)系數(shù)分別為0.835和0.759,這說明錨測驗(yàn)與原測驗(yàn)相關(guān)度很高,錨測驗(yàn)比較科學(xué),可以對被試施測。在總體權(quán)重為1時(shí),Tucker線性等值法的截距和斜率分別為-3.264和1.051;Levine觀察分?jǐn)?shù)法的截距和斜率分別為-8.719和1.188。在總體權(quán)重為0.501時(shí),Tucker線性等值法的截距和斜率分別為-3.774和1.063;Levine觀察分線法的截距和斜率分別為-8.52和1.183。Levine真分?jǐn)?shù)法的截距和斜率分別為-10.1和1.219。根據(jù)等值標(biāo)準(zhǔn)誤法計(jì)算得出,總體權(quán)重為1時(shí)的Tucker法的等值標(biāo)準(zhǔn)誤最小,為0.741。緊隨其后誤差依次變大的的分別是總體權(quán)重為0.501時(shí)的Tucker線性等值法,標(biāo)準(zhǔn)誤為0.856、總體權(quán)重為0.501時(shí)的Levine觀察分?jǐn)?shù)法,標(biāo)準(zhǔn)誤為2.086、總體權(quán)重為1時(shí)的Levine觀察分?jǐn)?shù)法,標(biāo)準(zhǔn)誤為2.137,而Levine真分?jǐn)?shù)法的等值標(biāo)準(zhǔn)誤最大,為2.431。根據(jù)交叉驗(yàn)證法計(jì)算得出,在總體權(quán)重為1時(shí),Tucker線性等值法的T值最小,為1.343。緊隨其后T值依次變大的分別為總體權(quán)重為0.5時(shí)的Tucker線性等值法,T值為2.523、總體權(quán)重為1時(shí)的Levine觀察分?jǐn)?shù)法,T值為2.973、總體權(quán)重為0.5時(shí)的Levine觀察分?jǐn)?shù)法,T值為4.199,而Levine真分?jǐn)?shù)法的T值最大,為5.799。對企業(yè)人力資源管理師理論考試進(jìn)行等值研究表明:(1)作為評(píng)價(jià)等值方法優(yōu)劣的一種方式,交叉驗(yàn)證法是實(shí)用且有效的。(2)無論使用等值標(biāo)準(zhǔn)誤法還是交叉驗(yàn)證法來比較等值方法的優(yōu)劣,均表明,Tucker線性等值法最優(yōu),Levine觀察分?jǐn)?shù)法其次,Levine真分?jǐn)?shù)法誤差最大?傮w權(quán)重為1時(shí)的Tucker法被證明是最優(yōu)且最適合該類型考試的方法。(3)在企業(yè)人力資源管理師理論考試中,等值結(jié)果受到總體權(quán)重的影響較小。
[Abstract]:Based on the classical test theory, this paper makes an equivalent study on the theoretical examination part of the Enterprise Human Resource Management examination (Grade 3) in May 2015 and November 2015. Using the combination of random equal group design and anchor test, At the same time, three linear equivalence methods based on classical test theory (Tucker linear equivalence method and Levine true fraction method) are tested and compared by using equivalent standard error method and cross validation method. The method with minimum error and strongest stability across samples is selected. An equivalent score conversion table is designed by using the optimal equivalence method, which provides ideas and references for future research in related fields. The correlation coefficients between the anchor test and the two original tests were 0.835 and 0.759 respectively, which indicated that the anchor test had a high correlation with the original test, and the anchor test was more scientific and could be tested on the subjects. The intercept and slope of Tucker linear equivalent method are -3.264 and 1.051g Levine respectively when the total weight is 1 and the intercept and slope of Levine observation fraction method are -8.719 and 1.1888 respectively. The intercept and slope of the Tucker linear equivalent method are -3.774 and 1.063 1.183.Levine, respectively. The intercept and slope of the Tucker linear equivalent method are -8.52 and 1.219, respectively. According to the calculation of equivalent standard error method, the equivalent standard error of Tucker method with total weight of 1 is the smallest, which is 0.741. The Tucker linear equivalence method with a total weight of 0.501, the standard error of 0.856, the Levine observation score method with a total weight of 0.501, the standard error of 2.086 and the Levine observation score method with a total weight of 1 were followed. The standard error was 2.137, while the equivalent error of Levine true fraction method was 2.431. According to the cross validation method, the minimum T value of the Tucker linear equivalent method is 1.343 when the total weight is 1. The value of T of Tucker linear equivalence method is 2.523 when the total weight is 0.5, the value of T of Levine observation fraction method is 2.973 when the total weight is 1, and the value of Levine observation fraction method is 4.199 when the total weight is 0.5, and the value of Levine is 4.199. The true fraction method has the largest T value, 5.799. A study on the equivalent value of the theoretical examination of human resource managers in enterprises shows that "1" is a way to evaluate the merits and demerits of the equivalence method. The cross validation method is practical and effective.) whether the equivalent standard error method or the cross validation method is used to compare the advantages and disadvantages of the equivalent method, it is shown that the Tucker linear equivalent method is the best Levine observation fraction method and the Levine true fraction method has the largest error. The Tucker method with a total weight of 1 has been proved to be the best and most suitable method for this type of examination.) in the theoretical examination of human resource managers in enterprises, the equivalent results are less affected by the total weight.
【學(xué)位授予單位】:魯東大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2017
【分類號(hào)】:C975;B849
【相似文獻(xiàn)】
相關(guān)期刊論文 前9條
1 余培英;;巧解“贏”字[J];咬文嚼字;2010年05期
2 ;封面人物尹曉燕[J];中國人力資源開發(fā);2011年10期
3 ;學(xué)員風(fēng)采[J];中國人力資源開發(fā);2008年04期
4 ;學(xué)員風(fēng)采[J];中國人力資源開發(fā);2006年08期
5 ;黑龍江省CHRP學(xué)員入選奧運(yùn)火炬手[J];中國人力資源開發(fā);2008年08期
6 ;李剛[J];人力資源管理;2013年11期
7 ;學(xué)員風(fēng)采[J];中國人力資源開發(fā);2006年12期
8 ;理事風(fēng)采[J];人力資源管理;2012年11期
9 ;羅箭華副教授[J];柳州師專學(xué)報(bào);2009年04期
相關(guān)重要報(bào)紙文章 前10條
1 記者 董煒;國際高級(jí)人力資源管理師考試在京舉行[N];中國人事報(bào);2009年
2 智夢尋;誰在冷落人力資源管理師認(rèn)證[N];國際商報(bào);2010年
3 本報(bào)記者 何瑞琳;人力資源管理師最熱門[N];揚(yáng)州日報(bào);2010年
4 本報(bào)記者 蘇茂華;人力資源管理師行情持續(xù)看漲的職業(yè)[N];濟(jì)南日報(bào);2005年
5 記者 張?jiān)伱?人力資源管理師職業(yè)標(biāo)準(zhǔn)調(diào)整[N];北京人才市場報(bào);2005年
6 宋德松 吳瓊;人力資源管理師:做企業(yè)的戰(zhàn)略推動(dòng)者[N];天津日報(bào);2005年
7 潘波;人力資源管理師今年考兩次[N];中國人事報(bào);2007年
8 劉靜;“全球人力資源管理師”登陸中國[N];北京人才市場報(bào);2007年
9 李敏芳;宜城成校建成人力資源管理師培訓(xùn)基地[N];宜興日報(bào);2010年
10 ;天津開發(fā)區(qū)引進(jìn)人力資源管理師培訓(xùn)[N];中國高新技術(shù)產(chǎn)業(yè)導(dǎo)報(bào);2003年
相關(guān)碩士學(xué)位論文 前1條
1 張s,
本文編號(hào):1888455
本文鏈接:http://sikaile.net/shekelunwen/xinlixingwei/1888455.html