SPSS相关性与偏相性关分析教材.doc
上传人:sy****28 上传时间:2024-09-13 格式:DOC 页数:25 大小:21KB 金币:14 举报 版权申诉
预览加载中,请您耐心等待几秒...

SPSS相关性与偏相性关分析教材.doc

SPSS相关性与偏相性关分析教材.doc

预览

免费试读已结束,剩余 15 页请下载文档后查看

14 金币

下载此文档

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

SPSS数据统计分析与实践主讲:周涛副教授北京师范大学资源学院2007-11-13教学网站:http://www.ires.cn/Courses/SPSS本页已使用福昕阅读器进行编辑。福昕软件(C)2005-2007,版权所有,仅供试用。第十一章相关分析本章内容:一、相关分析基本原理二、SPSS相关分析实例三、等级相关及SPSS实例四、偏相关分析及SPSS实例五、距离分析及SPSS实例一、相关分析基本原理CorrelationCorrelationCorrelation心血管健康CorrelationCorrelationCorrelationCorrelationCorrelationCorrelationCorrelationSignificanceofrSignificanceofr二、皮尔逊相关分析的注意事项(AttentionsofPearsonCorrelation)Attention1Toevaluateacorrelationcoefficient,firstdetermineitssignificance.Iftheprobabilitythatthecoefficientresultedfromchanceisnotacceptablylow,theanalysisshouldendthere;neitherthecoefficients’signnoritsmagnitudemayreflectanythingotherthansamplingerror.Ifthecoefficientisstatisticallysignificant,thesignshouldgiveanindicationofthedirectionoftherelationship,andthemagnitudeindicatesitsstrength.Remember,however,thatallstatisticsbecomesignificantwithahighenoughN.Keepinmind!!!Attention2Evenifit’sstatisticallysignificant,whetheracorrelationofagivenmagnitudeissubstantivelyorpracticallysignificantdependsgreatlyonthephenomenonbeingstudied.Generally,correlationstendtobehigherinthephysicalsciences,whererelationshipsbetweenvariablesoftenobeyuniformlaws,andlowerinthesocialsciences,whererelationshipmaybehardertopredict.Acorrelationof0.4betweenapairofsociologicalvariablesmaybemoremeaningfulthanacorrelationof0.7betweentwovariablesinphysics.Attention3Bearinmindalsothatthecorrelationcoefficientmeasuresonlystraight-linerelationships.Notallrelationshipsbetweenvariablestraceastraightline.TheFigureshowsacurvilinearrelationshipsuchthatvaluesofyincreasealongwithvaluesofxuptoapoint,thendecreasewithhighervaluesofx.Thecorrelationcoefficientforthisplotis0.Attention4Correlationdoesnotimplycausation.ThefactthatVariableAandVariableBarecorrelateddoesnecessarilymeanthatAcausedBorthatBcausedA(thougheithermaybetrue).Attention5Alsonotethatthescalesusedtomeasuretwovariableshavenoeffectontheircorrelation.Ifyouhadconvertedhoursofexerciseperweektominutesperweek,an