1、l有度量衡单位有度量衡单位l没有度量衡单位，但是连续性资料没有度量衡单位，但是连续性资料l量表条目的取值（条目等级量表条目的取值（条目等级=5）lt检验检验（两样本均数比较、样本均数与总体均数比较）（两样本均数比较、样本均数与总体均数比较）l配对配对t检验检验（同体自身前后，配对资料）（同体自身前后，配对资料）lu检验检验（两大样本资料）（两大样本资料）l方差分析方差分析（两组及两组以上）（两组及两组以上）l秩和检验秩和检验（偏态分布资料）（偏态分布资料）l可信区间可信区间x 表表1 甲乙两组病人年龄状况甲乙两组病人年龄状况 分组分组例数例数 n S 甲组甲组2444.717.1 乙组乙组27
2、45.716.11,122112(22(1nvnvSSF，较小）较大）.,05.028.2,128.11.161.17),(222623差异无统计学意义PFFvvGroup例数例数均数均数标准差标准差方差方差观察组观察组2354.997.5156.40对照组对照组2371.8911.98143.52安慰组安慰组2374.6417.43303.80合计合计6967.1715.48239.63变异来源变异来源SSVMSFP总总 16289.89682603.9615.510.0000组间组间 5207.932167.91 组内组内11081.9666 2,212121nnvSxxtxx2)1()1
3、(2122221121nnsnsnsxxGroup StatisticsGroup Statistics7247.000096.7384836.563715341.800057.8074425.85227分组试验组对照组指标1NMeanStd.DeviationStd.ErrorMean Levenes Te s t for Equality o f Variances t-test for Equalit y o f Means FSig.tdfSig.(2-tailed)M e a n DifferenceS t d.E r r o r Difference95%Confidence In
4、terval o f t h e Difference LowerUpperEqual variances assumed.341.57-1.94210.081-94.800048.82042-203.578613.9786Equal variances not assumed -2.1179.81.061-94.800044.77996-194.82725.227271,/0nvnsxtO On ne e-S Sa am mp pl le e S St ta at ti is st ti ic cs s10687.541.411.043平均睡眠时间NMeanStd.DeviationStd.
5、ErrorMeanO On ne e-S Sa am mp pl le e T Te es st t35.6841067.0001.541.461.63平均睡眠时间tdfSig.(2-tailed)MeanDifferenceLowerUpper95%ConfidenceInterval of theDifferenceTest Value=6O On ne e-S Sa am mp pl le e S St ta at ti is st ti ic cs s107786.740915.40382.46938107773.769737.277781.13591107782.364020.099
6、40.61246107759.761421.02131.64055107761.810614.62732.44571107775.263314.90322.45412107771.302138.732741.18024107763.335212.69912.38696PFRPBPGHVTSFREMHNMeanStd.DeviationStd.ErrorMeanO On ne e-S Sa am mp pl le e T Te es st t35.6661076.00016.740915.820017.66193.3191076.0013.76971.54095.998620.1881076.0
7、0012.364011.162213.5657-15.9841076.000-10.2386-11.4955-8.9818-18.3741076.000-8.1894-9.0640-7.314811.5901076.0005.26334.37226.15441.1031076.2701.3021-1.01383.6179-17.2241076.000-6.6648-7.4241-5.9055PFRPBPGHVTSFREMHtdfSig.(2-tailed)MeanDifference LowerUpper95%ConfidenceInterval of theDifferenceTest Va
8、lue=70 x分组分组 n S甲组甲组2936.59.7乙组乙组2734.15.122212121nsnsxxux 分组分组 N S 甲组甲组 1504.651012/L0.5481012/L乙组乙组 1004.181012/L0.6011012/L FSig.tdfSig.(2-tailed)95%C I LowerUpperPF9.818.0021.7261075.085-.228673.56322 1.781949.232.075-.169763.50432RP29.048.000-3.3791075.001-12.44183-3.30107 -3.268771.525.001-12.
9、60022-3.14268BP4.865.028-.1091075.913-2.614742.33986 -.107810.961.915-2.662782.38791GH2.357.125-.5061075.613-3.258411.92285 -.501839.025.616-3.281561.94600VT1.136.287-2.7541075.006-4.01371-.67393 -2.738846.191.006-4.02424-.66340SF5.311.021-.3141075.753-2.131031.54253 -.308807.166.758-2.169291.58080R
10、E11.498.001-2.6581075.008-11.14994-1.67984 -2.582782.991.010-11.29264-1.53714MH4.915.027-7.4181075.000-7.29806-4.24483 -7.283812.239.000-7.32694-4.215951,对子数vSdtd表表5.三棱莪术液抑瘤试验的效果三棱莪术液抑瘤试验的效果小白鼠小白鼠对子号对子号注注 射射药液组药液组对照组对照组 差值差值 d d2 13.03.6 0.6 0.36 22.34.5 2.2 4.84 32.44.2 1.8 3.24 41.14.4 3.310.89 54
11、.03.7-0.3 0.09 63.75.6 1.9 3.61 72.77.0 4.318.49 81.94.1 2.2 4.84 92.65.0 2.4 5.76101.34.5 3.210.24合计合计21.662.36组别组别例例数数疗前值疗前值疗后值疗后值前后差值前后差值前后差值前后差值 变化率变化率甲组甲组2585198116-4.05.2-5.65.5%乙组乙组25732875211.78.81.97.9%组间比较组间比较 t值（值（P值）值）.1.771.142.65(P0.05)3.90(P0.01)方差齐性方差齐性检验检验 F 值（值（P值）值）2.86(P0.05)表表6.
12、两组治疗前后差值（及差值变化率）两组治疗前后差值（及差值变化率）P Pa ai ir re ed d S Sa am mp pl le es s S St ta at ti is st ti ic cs s16.69712084.66662.3235713.65382084.57629.3173116.71432104.66015.3215811.67142104.48406.3094316.71432104.66015.321589.72862104.35260.3003616.78372084.62782.320887.90872084.11974.2856516.73432074.659
13、46.323866.34302074.02821.2799816.85222034.64990.326364.74882033.97778.2791916.91132034.56331.320283.73892033.86221.27107总分0总分1Pair1总分0总分2Pair2总分0总分3Pair3总分0总分4Pair4总分0总分5Pair5总分0总分6Pair6总分0总分7Pair7MeanNStd.DeviationStd.ErrorMeanPaired Samples CorrelationsPaired Samples Correlations208.798.000210.751
14、.000210.672.000208.585.000207.510.000203.399.000203.358.000总分0&总分1Pair 1总分0&总分2Pair 2总分0&总分3Pair 3总分0&总分4Pair 4总分0&总分5Pair 5总分0&总分6Pair 6总分0&总分7Pair 7NCorrelationSig.P Pa ai ir re ed d S Sa am mp pl le es s T Te es st t3.04332.93621.203592.64193.444614.948207.0005.04293.23158.223004.60325.482522.614
15、209.0006.98573.66214.252716.48757.483927.643209.0008.87504.00890.277978.32709.423031.928207.00010.39134.33618.301399.797110.985534.478206.00012.10344.76415.3343811.444112.762836.197202.00013.17244.80664.3373612.507213.837639.046202.000总分0-总分1Pair 1总分0-总分2Pair 2总分0-总分3Pair 3总分0-总分4Pair 4总分0-总分5Pair 5
16、总分0-总分6Pair 6总分0-总分7Pair 7MeanStd.DeviationStd.ErrorMeanLowerUpper95%ConfidenceInterval of theDifferencePaired DifferencestdfSig.(2-tailed)2,2)1()1(lglg21212lg22lg12121nnvnnsnsnGGtGGGroup StatisticsGroup Statistics10580.0000444.22217140.475388770.6250603.48060213.36261分组试验组对照组抗体滴度NMeanStd.DeviationS
17、td.ErrorMeanI In nd de ep pe en nd de en nt t S Sa ammp pl le es s T Te es st t4.640.047-.77316.451-190.6250246.62692-713.451332.20072-.74612.550.469-190.6250255.45438-744.520363.26962Equal variances assumedEqual variances notassumedFSig.Levenes Test forEquality of VariancestdfSig.(2-tailed)MeanDiff
18、erenceStd.ErrorDifferenceLowerUpper95%ConfidenceInterval of theDifferencet-test for Equality of MeansG Gr ro ou up p S St ta at ti is st ti ic cs s102.6431.35472.1121782.7102.46103.16300分组试验组对照组LGGNMeanStd.DeviationStd.ErrorMeanI In nd de ep pe en nd de en nt t S Sa am mp pl le es s T Te es st t2.52
19、5.132-.35016.731-.0671.19196-.47404.33982-.33912.943.740-.0671.19787-.49477.36055Equal variancesassumedEqual variancesnot assumedLGGFSig.Levenes Test forEquality ofVariancestdfSig.(2-tailed)MeanDifferenceStd.ErrorDifferenceLowerUpper95%ConfidenceInterval of theDifferencet-test for Equality of Meansl
20、假设检验用于推断两总体均假设检验用于推断两总体均数有无差别数有无差别l可信区间用于推断总体均数可信区间用于推断总体均数在哪个范围，可信区间可以在哪个范围，可信区间可以回答假设检验的问题回答假设检验的问题l假设检验能得到假设检验能得到P值，比较精值，比较精确地说明结论的概率保证确地说明结论的概率保证l可信区间提供在可信区间提供在水平上有无水平上有无统计学意义，但不易得到精统计学意义，但不易得到精确的概率。确的概率。l抽样误差抽样误差l样本含量样本含量l检验水平检验水平l检验效能检验效能l统计术语统计术语l检验统计量检验统计量l单侧检验还是双侧检验单侧检验还是双侧检验lP值的确切范围值的确切范围l给出总体均数的可信区间或两总体给出总体均数的可信区间或两总体均数之差的可信区间均数之差的可信区间l结合专业知识作出有实际意义的专结合专业知识作出有实际意义的专业结论业结论
