再访依赖相关性比较检验:蒙特卡洛研究。

IF 2.2 4区教育学 Q1 Social Sciences

Journal of Experimental Education Pub Date : 1997-01-01 DOI:10.1080/00220973.1997.9943458

K. May, J. Hittner

{"title":"再访依赖相关性比较检验:蒙特卡洛研究。","authors":"K. May, J. Hittner","doi":"10.1080/00220973.1997.9943458","DOIUrl":null,"url":null,"abstract":"Abstract A Monte Carlo evaluation of 4 test statistics for comparing dependent zero-order correlations was conducted. In particular, the power and Type I error rates of Hotelling's t; Williams' t; Olkin's z; and Meng, Rosenthal, and Rubin's Z were evaluated for sample sizes of 20, 50, 100, and 300 under 3 different population distributions (normal, uniform, and exponential). For the power analyses, 3 different magnitudes of discrepancy or effect sizes between ρy, x1 , and ρy, x2 were examined (values of .1, .3, and .6). Likewise, for the Type I error rate analyses, 3 different magnitudes of the predictor-criterion correlations were evaluated (ρy, x1 = ρy, x2 = .1, .4, and .7). All of the analyses were conducted at 3 different levels of predictor intercorrelation (ρx1, x2 = .1, .3, and .6). The results indicated that the choice as to which test statistic is optimal, in terms of power and Type I error rate, depends not only on sample size and population distribution but also on (a) the predictor intercorrel...","PeriodicalId":47911,"journal":{"name":"Journal of Experimental Education","volume":"65 1","pages":"257-269"},"PeriodicalIF":2.2000,"publicationDate":"1997-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/00220973.1997.9943458","citationCount":"35","resultStr":"{\"title\":\"Tests for Comparing Dependent Correlations Revisited: A Monte Carlo Study.\",\"authors\":\"K. May, J. Hittner\",\"doi\":\"10.1080/00220973.1997.9943458\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract A Monte Carlo evaluation of 4 test statistics for comparing dependent zero-order correlations was conducted. In particular, the power and Type I error rates of Hotelling's t; Williams' t; Olkin's z; and Meng, Rosenthal, and Rubin's Z were evaluated for sample sizes of 20, 50, 100, and 300 under 3 different population distributions (normal, uniform, and exponential). For the power analyses, 3 different magnitudes of discrepancy or effect sizes between ρy, x1 , and ρy, x2 were examined (values of .1, .3, and .6). Likewise, for the Type I error rate analyses, 3 different magnitudes of the predictor-criterion correlations were evaluated (ρy, x1 = ρy, x2 = .1, .4, and .7). All of the analyses were conducted at 3 different levels of predictor intercorrelation (ρx1, x2 = .1, .3, and .6). The results indicated that the choice as to which test statistic is optimal, in terms of power and Type I error rate, depends not only on sample size and population distribution but also on (a) the predictor intercorrel...\",\"PeriodicalId\":47911,\"journal\":{\"name\":\"Journal of Experimental Education\",\"volume\":\"65 1\",\"pages\":\"257-269\"},\"PeriodicalIF\":2.2000,\"publicationDate\":\"1997-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.1080/00220973.1997.9943458\",\"citationCount\":\"35\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Experimental Education\",\"FirstCategoryId\":\"95\",\"ListUrlMain\":\"https://doi.org/10.1080/00220973.1997.9943458\",\"RegionNum\":4,\"RegionCategory\":\"教育学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"Social Sciences\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Experimental Education","FirstCategoryId":"95","ListUrlMain":"https://doi.org/10.1080/00220973.1997.9943458","RegionNum":4,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Social Sciences","Score":null,"Total":0}

引用次数: 35

摘要

摘要对4个检验统计量进行蒙特卡罗评价，以比较相关零阶相关性。特别是霍特林t的幂和I型错误率;威廉姆斯的t;Olkin z;在3种不同的总体分布(正态分布、均匀分布和指数分布)下，对样本量为20、50、100和300的孟、Rosenthal和Rubin’s Z进行了评估。对于功效分析，检验了ρy, x1和ρy, x2之间的3个不同的差异大小或效应大小(值为.1，.3和.6)。同样，对于I型错误率分析，评估了3种不同程度的预测-标准相关性(ρy, x1 = ρy, x2 = .1， .4和.7)。所有分析均在三个不同的预测因子相关水平(ρx1, x2 = .1， .3和.6)下进行。结果表明，选择哪个检验统计量是最优的，在功率和I型错误率方面，不仅取决于样本量和总体分布，而且取决于(a)预测因子的相互关系。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Tests for Comparing Dependent Correlations Revisited: A Monte Carlo Study.

Abstract A Monte Carlo evaluation of 4 test statistics for comparing dependent zero-order correlations was conducted. In particular, the power and Type I error rates of Hotelling's t; Williams' t; Olkin's z; and Meng, Rosenthal, and Rubin's Z were evaluated for sample sizes of 20, 50, 100, and 300 under 3 different population distributions (normal, uniform, and exponential). For the power analyses, 3 different magnitudes of discrepancy or effect sizes between ρy, x1 , and ρy, x2 were examined (values of .1, .3, and .6). Likewise, for the Type I error rate analyses, 3 different magnitudes of the predictor-criterion correlations were evaluated (ρy, x1 = ρy, x2 = .1, .4, and .7). All of the analyses were conducted at 3 different levels of predictor intercorrelation (ρx1, x2 = .1, .3, and .6). The results indicated that the choice as to which test statistic is optimal, in terms of power and Type I error rate, depends not only on sample size and population distribution but also on (a) the predictor intercorrel...

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Experimental Education Multiple-

CiteScore

6.70

自引率

0.00%

发文量

期刊介绍： The Journal of Experimental Education publishes theoretical, laboratory, and classroom research studies that use the range of quantitative and qualitative methodologies. Recent articles have explored the correlation between test preparation and performance, enhancing students" self-efficacy, the effects of peer collaboration among students, and arguments about statistical significance and effect size reporting. In recent issues, JXE has published examinations of statistical methodologies and editorial practices used in several educational research journals.