雙因素及多因素SPSS方差分析PPT學(xué)習(xí)教案

1、會計學(xué)1 雙因素及多因素雙因素及多因素SPSS方差分析方差分析 第1頁/共53頁 第2頁/共53頁 第3頁/共53頁 第4頁/共53頁 第5頁/共53頁 不同種系、劑量的子宮重量例題數(shù)據(jù)不同種系、劑量的子宮重量例題數(shù)據(jù) 方差分析的數(shù)據(jù)安排方差分析的數(shù)據(jù)安排 劑劑 量量 種系種系0.2（1）0.4（2）0.8（3） A（1）106116145 B（2）4268115 C（3）70111133 D（4）426387 第6頁/共53頁 Tests of Between-Subjects Effects Dependent Variable: WUTERI 12531.667a52506.33327.

2、677.000 100467.01100467.01109.452.000 6457.66732152.55623.771.001 6074.00023037.00033.537.001 543.333690.556 113542.012 13075.00011 Source Corrected Model Intercept MOUSE ETROGEN Error Total Corrected Total Type III Sum of Squaresdf Mean SquareFSig. R Squared = .958 (Adjusted R Squared = .924)a. 主效應(yīng)

3、方差分析檢驗結(jié)果主效應(yīng)方差分析檢驗結(jié)果 第7頁/共53頁 Tests of Between-Subjects Effects Dependent Variable: 紅細(xì)胞增加數(shù) 2.963a3.98898.750.000 19.507119.507 1950.750.000 1.68811.688168.750.000 .9071.90790.750.000 .3681.36836.750.000 8.000E-028 1.000E-02 22.55012 3.04311 Source Corrected Model Intercept DRUGA DRUGB DRUGA * DRUGB E

4、rror Total Corrected Total Type III Sum of Squaresdf Mean SquareFSig. R Squared = .974 (Adjusted R Squared = .964)a. 方差分析表 第8頁/共53頁 A藥效應(yīng)均值圖藥效應(yīng)均值圖 B藥效應(yīng)均值圖藥效應(yīng)均值圖 A、B藥對紅細(xì)胞增加數(shù)交互效應(yīng)邊際圖藥對紅細(xì)胞增加數(shù)交互效應(yīng)邊際圖 第9頁/共53頁 Between-Subjects Factors 121212121212121212121212121212121212N 123456 REP 123456 COL 123456 VARIE

5、TY 因素變量表因素變量表 Tests of Between-Subjects Effects Dependent Variable: YIELD 27.717a151.8481.339.211 22588.751122588.75116364.072.000 4.4605.892.646.666 1.6955.339.246.940 21.56354.3133.124.015 77.302561.380 22693.77072 105.01971 Source Corrected Model Intercept REP COL VARIETY Error Total Corrected To

6、tal Type III Sum of Squaresdf Mean SquareFSig. R Squared = .264 (Adjusted R Squared = .067)a. 方差分析表 第10頁/共53頁 P Pa ai ir rw wi is se e C Co om mp pa ar ri is so on ns s Dependent Variable: YIELD .192.480.691-.7691.153 -.167.480.730-1.128.794 -.083.480.863-1.044.878 .333.480.490-.6281.294 .550.480.25

7、6-.4111.511 -.192.480.691-1.153.769 -.358.480.458-1.319.603 -.275.480.569-1.236.686 .142.480.769-.8191.103 .358.480.458-.6031.319 .167.480.730-.7941.128 .358.480.458-.6031.319 .083.480.863-.8781.044 .500.480.302-.4611.461 .717.480.141-.2441.678 .083.480.863-.8781.044 .275.480.569-.6861.236 -.083.480

8、.863-1.044.878 .417.480.389-.5441.378 .633.480.192-.3281.594 -.333.480.490-1.294.628 -.142.480.769-1.103.819 -.500.480.302-1.461.461 -.417.480.389-1.378.544 .217.480.653-.7441.178 -.550.480.256-1.511.411 -.358.480.458-1.319.603 -.717.480.141-1.678.244 -.633.480.192-1.594.328 -.217.480.653-1.178.744

9、(J) REP 2 3 4 5 6 1 3 4 5 6 1 2 4 5 6 1 2 3 5 6 1 2 3 4 6 1 2 3 4 5 (I) REP 1 2 3 4 5 6 Mean Difference (I-J) Std. ErrorSig.a Lower Bound Upper Bound 95% Confidence Interval for Difference a Based on estimated marginal means Adjustment for multiple comparisons: Least Significant Difference (equivale

10、nt to no adjustments). a. 第11頁/共53頁 B Be et tw we ee en n- -S Su ub bj je ec ct ts s F Fa ac ct to or rs s 大于等于10年12 不足10年16 1 2 暴露于鎘 塵年段 Value LabelN 因素變量表 Tests of Between-Subjects Effects Dependent Variable: VITALCP 11.085a25.54310.073.001 41.936141.93676.216.000 10.881110.88119.775.000 .5421.542

11、.985.330 13.75525.550 483.62528 24.84127 Source Corrected Model Intercept AGE TIME Error Total Corrected Total Type III Sum of Squaresdf Mean SquareFSig. R Squared = .446 (Adjusted R Squared = .402)a. 方差分析表 第12頁/共53頁 P Pa ar ra am me et te er r E Es st ti im ma at te es s Dependent Variable: 肺活量 7.9

12、77.8868.998.0006.1519.803 -8.70E-02.020-4.447.000-.127-4.670E-02 .300.303.993.330-.323.924 0a. Parameter Intercept AGE TIME=1 TIME=2 BStd. ErrortSig.Lower BoundUpper Bound 95% Confidence Interval This parameter is set to zero because it is redundant.a. 參數(shù)估測值的輸出結(jié)果表 按時間分組的肺活量均值表 暴暴 露露 于于 鎘鎘 塵塵 年年 段段 D

13、ependent Variable: 肺活量 4.219a.2233.7614.678 3.919a.1913.5264.312 暴露于鎘塵年段 大于等于10年 不足10年 MeanStd. ErrorLower BoundUpper Bound 95% Confidence Interval Evaluated at covariates appeared in the model: 年齡 = 46.64. a. 第13頁/共53頁 Between-Subjects Factors t130 t230 t330 t430 d140 d240 d340 l160 l260 1 2 3 4 ta

14、rget 1 2 3 device 1 2 light Value LabelN 因素變量表因素變量表 Tests of Between-Subjects Effects Dependent Variable: SCORE 783.467a2334.06446.451.000 3162.13313162.133 4312.000.000 235.200378.400106.909.000 86.467243.23358.955.000 76.800176.800104.727.000 104.200617.36723.682.000 93.867331.28942.667.000 186.93

15、3823.36731.864.000 70.40096.733 4016.000120 853.867119 Source Corrected Model Intercept TARGET DEVICE LIGHT TARGET * DEVICE TARGET * LIGHT TARGET * DEVICE * LIGHT Error Total Corrected Total Type III Sum of Squaresdf Mean SquareFSig. R Squared = .918 (Adjusted R Squared = .898)a. 方差分析結(jié)果方差分析結(jié)果 第14頁/共

16、53頁 device * light * target Dependent Variable: SCORE 2.000.3831.2402.760 8.000.3837.2408.760 8.800.3838.0409.560 7.200.3836.4407.960 1.600.383.8402.360 4.400.3833.6405.160 5.600.3834.8406.360 6.800.3836.0407.560 .800.3834.E-021.560 8.400.3837.6409.160 9.600.3838.84010.360 9.200.3838.4409.960 5.200.

17、3834.4405.960 6.400.3835.6407.160 4.800.3834.0405.560 2.800.3832.0403.560 4.000.3833.2404.760 5.200.3834.4405.960 6.000.3835.2406.760 2.000.3831.2402.760 3.200.3832.4403.960 1.200.383.4401.960 4.400.3833.6405.160 5.600.3834.8406.360 target t1 t2 t3 t4 t1 t2 t3 t4 t1 t2 t3 t4 t1 t2 t3 t4 t1 t2 t3 t4

18、t1 t2 t3 t4 light l1 l2 l1 l2 l1 l2 device d1 d2 d3 Mean Std. Error Lower Bound Upper Bound 95% Confidence Interval 各單元格觀測量均值（部分）各單元格觀測量均值（部分） 第15頁/共53頁 第一種照度下心理得分 第二種照度下心理得分 邊際均值圖邊際均值圖 第16頁/共53頁 第17頁/共53頁 第18頁/共53頁 第19頁/共53頁 變量名變量名標(biāo)標(biāo) 簽簽中文標(biāo)簽中文標(biāo)簽值值值標(biāo)簽值標(biāo)簽值標(biāo)簽值標(biāo)簽(中文中文) losLength of stay 住院時間長住院時間長 短短 co

19、stTreatment costs治療花費(fèi)治療花費(fèi) clotsolv Clot-dissolving drugs 凝塊消溶藥凝塊消溶藥 1 Streptokina se 鏈激酶鏈激酶 2reteplase瑞替普酶瑞替普酶 3alteplase阿替普酶阿替普酶 procSurgical treatment手術(shù)治療手術(shù)治療 1PTCA 經(jīng)皮冠狀動脈成經(jīng)皮冠狀動脈成 型術(shù)型術(shù) 2CABG搭橋術(shù)搭橋術(shù) 第20頁/共53頁 B Be et tw we ee en n- -S Su ub bj je ec ct ts s F Fa ac ct to or rs s 經(jīng)皮冠狀動脈成型術(shù)907 搭橋術(shù)574

20、鏈激酶116 瑞替普酶696 阿替普酶669 1 2 外科 手術(shù) 1 2 3 凝塊 消溶 藥 Value LabelN D De es sc cr ri ip pt ti iv ve e S St ta at ti is st ti ic cs s 4.941.10568 4.811.072441 4.681.048398 4.771.066907 7.251.26348 6.621.137255 6.481.135271 6.601.163574 5.901.633116 5.471.399696 5.411.396669 5.481.4221481 28.38383.2738868 29.

21、66743.18096441 29.80733.60094398 29.63263.39406907 44.72255.4278048 44.62515.22506255 44.74325.63081271 44.68905.42789574 35.14479.14344116 35.14768.27021696 35.85758.62337669 35.46818.503141481 凝塊消溶藥 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total

22、 鏈激酶 瑞替普酶 阿替普酶 Total 外科手術(shù) 經(jīng)皮冠狀動脈成型術(shù) 搭橋術(shù) Total 經(jīng)皮冠狀動脈成型術(shù) 搭橋術(shù) Total 住院時間(天) 治療花費(fèi) MeanStd. DeviationN 各單元頻數(shù)各單元頻數(shù) 描述統(tǒng)計量描述統(tǒng)計量 第21頁/共53頁 Boxs Test of Equality of Covariance MatricesBoxs Test of Equality of Covariance Matrices a a 270.509 17.908 15 358296.5 .000 Boxs M F df1 df2 Sig. Tests the null hypothe

23、sis that the observed covariance matrices of the dependent variables are equal across groups. Design: Intercept+clotsolv+proc+clotsolv * proc a. Levenes Test of Equality of Error VariancesLevenes Test of Equality of Error Variances a a 1.50751475.185 10.00151475.000 住院時間(天) 治療花費(fèi) Fdf1df2Sig. Tests th

24、e null hypothesis that the error variance of the dependent variable is equal across groups. Design: Intercept+proc+clotsolv+proc * clotsolv a. 協(xié)方差陣的齊性檢驗協(xié)方差陣的齊性檢驗 第22頁/共53頁 不同手術(shù)花費(fèi)直方圖不同手術(shù)花費(fèi)直方圖住院時間直方圖住院時間直方圖 注：第注：第3 3版第版第1 1次印刷書中丟掉了這個部分，圖標(biāo)號順序有誤次印刷書中丟掉了這個部分，圖標(biāo)號順序有誤 第23頁/共53頁 Between-Subjects FactorsBetw

25、een-Subjects Factors 鏈激酶68 瑞替普酶441 阿替普酶398 鏈激酶48 瑞替普酶255 阿替普酶271 1 2 3 凝塊 消溶 藥 1 2 3 凝塊 消溶 藥 外科手術(shù) 經(jīng)皮冠狀動 脈成型術(shù) 搭橋術(shù) Value LabelN Descriptive StatisticsDescriptive Statistics 4.941.10568 4.811.072441 4.681.048398 4.771.066907 1.4504.0480268 1.4700.04392441 1.4715.04883398 1.4692.04671907 7.251.26348 6.6

26、21.137255 6.481.135271 6.601.163574 1.6478.0479648 1.6470.04538255 1.6478.04910271 1.6474.04731574 凝塊消溶藥 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 鏈激酶 瑞替普酶 阿替普酶 Total 住院時間 (天) 治療花費(fèi)對 數(shù)(底10) 住院時間 (天) 治療花費(fèi)對 數(shù)(底10) 外科手 術(shù)經(jīng)皮冠 狀動脈 成型術(shù) 搭橋術(shù) MeanStd. DeviationN 組間因素各單元頻數(shù)組間因素各單元頻數(shù) 描述統(tǒng)計量描述統(tǒng)計量

27、第24頁/共53頁 Boxs Test of Equality of Covariance MatricesBoxs Test of Equality of Covariance Matrices a a 12.208 2.021 6 246014.6 .059 7.804 1.288 6 126223.5 .259 Boxs M F df1 df2 Sig. Boxs M F df1 df2 Sig. 經(jīng)皮冠狀動脈成型術(shù) 搭橋術(shù) Tests the null hypothesis that the observed covariance matrices of the dependent v

28、ariables are equal across groups. Design: Intercept+clotsolv a. L Le ev ve en ne e s s T Te es st t o of f E Eq qu ua al li it ty y o of f E Er rr ro or r V Va ar ri ia an nc ce es s a a .5452904.580 1.9502904.143 .5242571.592 .8202571.441 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 外科手術(shù) 經(jīng)皮冠狀 動脈成型 術(shù) 搭橋術(shù)

29、 Fdf1df2Sig. Tests the null hypothesis that the error variance of the dependent variable is equal across groups. Design: Intercept+clotsolv a. 協(xié)方差陣相等的協(xié)方差陣相等的Box檢驗檢驗 誤差方差相等的誤差方差相等的Leven檢驗檢驗 第25頁/共53頁 Tests of Between-Subjects EffectsTests of Between-Subjects Effects 5.725a22.8632.529.080.006 .026b2.0

30、136.112.002.013 10695.320110695.3209448.8.000.913 989.8451989.845458854.000.998 5.72522.8632.529.080.006 .0262.0136.112.002.013 1023.2549041.132 1.950904.002 21624.000907 1959.676907 1028.979906 1.976906 24.504c212.2529.316.000.032 .000d2.000.017.984.000 14546.869114546.86911061.000.951 858.8181858.

31、818382466.000.999 24.504212.2529.316.000.032 .0002.000.017.984.000 750.9315711.315 1.282571.002 25800.000574 1559.156574 775.436573 1.282573 Dependent Variable 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10)

32、住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) 住院時間(天) 治療花費(fèi)對數(shù)(底10) Source Corrected Model Intercept clotsolv Error Total Corrected Total Corrected Model Intercept clotsolv Error Total Corrected Total 外科手術(shù) 經(jīng)皮冠狀動脈成型術(shù) 搭橋術(shù) Type III Sum of SquaresdfMean SquareFSig. Partia

33、l Eta Squared R Squared = .006 (Adjusted R Squared = .003) a. R Squared = .013 (Adjusted R Squared = .011) b. R Squared = .032 (Adjusted R Squared = .028) c. R Squared = .000 (Adjusted R Squared = -.003) d. 多元方差分析檢驗結(jié)果多元方差分析檢驗結(jié)果 第26頁/共53頁 外科手術(shù)=經(jīng)皮冠狀動脈成型術(shù)外科手術(shù)=經(jīng)皮冠狀動脈成型術(shù) -.129.020 00 -.129.020 .139.006 .

34、351.001 -.401.008 .143.031 -.258.021 00 -.258.021 .140.006 .065.001 -.532.009 .016.033 Contrast Estimate Hypothesized Value Difference (Estimate - Hypothesized) Std. Error Sig. Lower Bound Upper Bound 95% Confidence Interval for Difference Contrast Estimate Hypothesized Value Difference (Estimate -

35、Hypothesized) Std. Error Sig. Lower Bound Upper Bound 95% Confidence Interval for Difference 凝塊消溶藥 Simple Contrast a Level 2 vs. Level 1 Level 3 vs. Level 1 住院時間(天) 治療花費(fèi) 對數(shù)(底10) Dependent Variable Reference category = 1 a. 不同凝塊消溶藥對住院時間和治療花費(fèi)的影響（不同凝塊消溶藥對住院時間和治療花費(fèi)的影響（PTCAPTCA） 第27頁/共53頁 外科手術(shù)=搭橋術(shù)外科手術(shù)=搭橋

36、術(shù) -.634-.001 00 -.634-.001 .180.007 .000.922 -.989-.015 -.280.014 -.774.000 00 -.774.000 .180.007 .000.999 -1.127-.015 -.421.015 Contrast Estimate Hypothesized Value Difference (Estimate - Hypothesized) Std. Error Sig. Lower Bound Upper Bound 95% Confidence Interval for Difference Contrast Estimate

37、Hypothesized Value Difference (Estimate - Hypothesized) Std. Error Sig. Lower Bound Upper Bound 95% Confidence Interval for Difference 凝塊消溶藥 Simple Contrast a Level 2 vs. Level 1 Level 3 vs. Level 1 住院時間(天) 治療花費(fèi) 對數(shù)(底10) Dependent Variable Reference category = 1 a. 不同凝塊消溶藥對住院時間和治療花費(fèi)的影響（不同凝塊消溶藥對住院時間和治

38、療花費(fèi)的影響（CABG） 第28頁/共53頁 外科手術(shù)=經(jīng)皮冠狀動脈成型術(shù)外科手術(shù)=經(jīng)皮冠狀動脈成型術(shù) Tukey HSDa,b 681.4504 4411.4700 3981.4715 1.000.958 凝塊消溶藥 鏈激酶 瑞替普酶 阿替普酶 Sig. N12 Subset Means for groups in homogeneous subsets are displayed. Based on Type III Sum of Squares The error term is Mean Square(Error) = .002. Uses Harmonic Mean Sample S

39、ize = 153.957. a. Alpha = .05. b. 外科手術(shù)=搭橋術(shù)外科手術(shù)=搭橋術(shù) Tukey HSDa,b 2551.6470 2711.6478 481.6478 .993 凝塊消溶藥 瑞替普酶 阿替普酶 鏈激酶 Sig. N1 Subset Means for groups in homogeneous subsets are displayed. Based on Type III Sum of Squares The error term is Mean Square(Error) = .002. Uses Harmonic Mean Sample Size = 1

40、05.467. a. Alpha = .05. b. 治療花費(fèi)（對數(shù)）的一致性子集治療花費(fèi)（對數(shù)）的一致性子集 第29頁/共53頁 第30頁/共53頁 原始數(shù)據(jù)原始數(shù)據(jù) 數(shù)據(jù)文件結(jié)構(gòu)數(shù)據(jù)文件結(jié)構(gòu) 第31頁/共53頁 第32頁/共53頁 第33頁/共53頁 第34頁/共53頁 不同藥物對受試者的影響不同藥物對受試者的影響 第35頁/共53頁 M Mu ul lt ti iv va ar ri ia at te e T Te es st ts s b b .97728.412a3.0002.000.034 .02328.412a3.0002.000.034 42.61828.412a3.0002.

41、000.034 42.61828.412a3.0002.000.034 Pillais Trace Wilks Lambda Hotellings Trace Roys Largest Root Effect MED ValueF Hypothesis df Error dfSig. Exact statistica. Design: Intercept Within Subjects Design: MED b. 藥物間多元檢驗藥物間多元檢驗 第36頁/共53頁 Tests of Within-Subjects EffectsTests of Within-Subjects Effects

42、Measure: MEASURE_1 698.2003232.73324.759.000 698.2001.815384.76324.759.001 698.2003.000232.73324.759.000 698.2001.000698.20024.759.008 112.800129.400 112.8007.25815.540 112.80012.0009.400 112.8004.00028.200 Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound Sphericity Assumed Greenhouse-G

43、eisser Huynh-Feldt Lower-bound Source MED Error(MED) Type III Sum of SquaresdfMean SquareFSig. 被試者內(nèi)效應(yīng)方差分析結(jié)果被試者內(nèi)效應(yīng)方差分析結(jié)果 第37頁/共53頁 返回返回 第38頁/共53頁 返回返回 第39頁/共53頁 返回返回 第40頁/共53頁 返回返回 第41頁/共53頁 ANOVAANOVA方差分析結(jié)果方差分析結(jié)果 ANOVAANOVA 783.4672334.064 3162.13313162.133 76.800176.800 235.200378.400 86.467243.233

44、 12.60026.300 93.867331.289 278.5331223.211 70.40096.733 4016.000120 853.867119 Source Corrected Model Intercept LIGHT TARGET DEVICE LIGHT * DEVICE LIGHT * TARGET LIGHT * TARGET * DEVICE Error Total Corrected Total Type III Sum of SquaresdfMean Square Dependent Variable: SCORE 返回返回 第42頁/共53頁 V Va ar

45、 ri ia an nc ce e E Es st ti im ma at te es s 1.040 -.846a .539 4.496 .733 Component Var(LIGHT) Var(LIGHT * DEVICE) Var(LIGHT * TARGET) Var(LIGHT * TARGET * DEVICE) Var(Error) Estimate Dependent Variable: SCORE Method: ANOVA (Type III Sum of Squares) For the ANOVA and MINQUE methods, negative varian

46、ce component estimates may occur. Some possible reasons for their occurrence are: (a) the specified model is not the correct model, or (b) the true value of the variance equals zero. a. Variance EstimatesVariance Estimates .759 .700 4.012 .733 Component Var(light) Var(light * target) Var(light * target * device) Var(Error) Estimate Dependent Variable: score Method: ANOVA (Type III Sum of Squares) 方差成分表（去掉方差成分表（去掉LightLight* *devicedevice ） 返回返回 方差成分表方差成分表 第43頁/共53頁 A AN NO OV VA A 783.4672334.064 3162.13313162.133 235.200378.400 86.467243.233 76.8