实验主题 | SPSS统计分析 |
实验题目 | 方差分析 |
一 实验目的
掌握单因素方差分析的原理与步骤、多因素方差分析的原理与步骤、协方差分析的原理与步骤。
二 实验内容
题目一:某农场为了比较4种不同品种的小麦产量的差异,选择土壤条件基本相同的土地,分成16块,将每一个品种在4块试验田上试种,测得小表亩产量(kg)的数据如表6.17所示(数据文件为data6-4.sav), 试问不同品种的小麦的平均产量在显著性水平0.05和0.01下有无显著性差异。(数据来源:《SPSS实用统计分析》 郝黎仁,中国水利水电出版社)
表6.17 小麦产量的实测数据
品种 | A1 | A2 | A3 | A4 |
产量 | 277.5 | 244.2 | 249.2 | 273 |
276.4 | 249.5 | 244.2 | 240.9 | |
271 | 236.8 | 252.8 | 257.4 | |
272.4 | 239 | 251.4 | 266.5 | |
实验结果截图:
Multiple Comparisons | ||||||
产量 LSD | ||||||
(I) 品种 | (J) 品种 | Mean Difference (I-J) | Std. Error | Sig. | 95% Confidence Interval | |
Lower Bound | Upper Bound | |||||
A1 | A2 | 31.70000* | 5.57044 | .000 | 19.5631 | 43.8369 |
A3 | 24.67500* | 5.57044 | .001 | 12.5381 | 36.8119 | |
A4 | 14.87500* | 5.57044 | .020 | 2.7381 | 27.0119 | |
A2 | A1 | -31.70000* | 5.57044 | .000 | -43.8369 | -19.5631 |
A3 | -7.02500 | 5.57044 | .231 | -19.1619 | 5.1119 | |
A4 | -16.82500* | 5.57044 | .011 | -28.9619 | -4.6881 | |
A3 | A1 | -24.67500* | 5.57044 | .001 | -36.8119 | -12.5381 |
A2 | 7.02500 | 5.57044 | .231 | -5.1119 | 19.1619 | |
A4 | -9.80000 | 5.57044 | .104 | -21.9369 | 2.3369 | |
A4 | A1 | -14.87500* | 5.57044 | .020 | -27.0119 | -2.7381 |
A2 | 16.82500* | 5.57044 | .011 | 4.6881 | 28.9619 | |
A3 | 9.80000 | 5.57044 | .104 | -2.3369 | 21.9369 | |
*. The mean difference is significant at the 0.05 level. | ||||||
实验结果分析:
根据不同小麦的平均产量在显著性水平0.05和0.01下的奇性检验结果、方差检验结果、多重比较结果、均值折线图可以看出,不管是方差还是均值,差异较大,而它的均值折线图分布比较陡峭。所以,不同小麦的平均产量有显著差异。
题目二:2. 某公司希望检测四种类型的轮胎A,B,C,D的寿命(由行驶的里程数决定),见表6.18(单位:千英里)(数据文件为data6-5.sav),其中每种轮胎应用在随机选择的6辆汽车上。在显著性水平0.05下判断不同类型轮胎的寿命间是否存在显著性差异?(数据来源:《统计学(第三版)》,M.R.斯皮格尔,科学出版社)
表6.18 四种轮胎的寿命数据
A | 33 | 38 | 36 | 40 | 31 | 35 |
B | 32 | 40 | 42 | 38 | 30 | 34 |
C | 31 | 37 | 35 | 33 | 34 | 30 |
D | 29 | 34 | 32 | 30 | 33 | 31 |
实验结果截图:
ANOVA | |||||
里程 | |||||
Sum of Squares | df | Mean Square | F | Sig. | |
Between Groups | 77.500 | 3 | 25.833 | 2.388 | .099 |
Within Groups | 216.333 | 20 | 10.817 | ||
Total | 293.833 | 23 | |||
Multiple Comparisons | |||||||
Dependent Variable:里程 | |||||||
(I) 轮胎 | (J) 轮胎 | Mean Difference (I-J) | Std. Error | Sig. | 95% Confidence Interval | ||
Lower Bound | Upper Bound | ||||||
LSD | A | B | -.500 | 1.899 | .795 | -4.46 | 3.46 |
C | 2.167 | 1.899 | .267 | -1.79 | 6.13 | ||
D | 4.000* | 1.899 | .048 | .04 | 7.96 | ||
B | A | .500 | 1.899 | .795 | -3.46 | 4.46 | |
C | 2.667 | 1.899 | .176 | -1.29 汽车lsd | 6.63 | ||
D | 4.500* | 1.899 | .028 | .54 | 8.46 | ||
C | A | -2.167 | 1.899 | .267 | -6.13 | 1.79 | |
B | -2.667 | 1.899 | .176 | -6.63 | 1.29 | ||
D | 1.833 | 1.899 | .346 | -2.13 | 5.79 | ||
D | A | -4.000* | 1.899 | .048 | -7.96 | -.04 | |
B | -4.500* | 1.899 | .028 | -8.46 | -.54 | ||
C | -1.833 | 1.899 | .346 | -5.79 | 2.13 | ||
Tamhane | A | B | -.500 | 2.349 | 1.000 | -8.40 | 7.40 |
C | 2.167 | 1.701 | .797 | -3.46 | 7.79 | ||
D | 4.000 | 1.538 | .176 | -1.34 | 9.34 | ||
B | A | .500 | 2.349 | 1.000 | -7.40 | 8.40 | |
C | 2.667 | 2.201 | .838 | -5.04 | 10.37 | ||
D | 4.500 | 2.078 | .352 | -3.22 | 12.22 | ||
C | A | -2.167 | 1.701 | .797 | -7.79 | 3.46 | |
B | -2.667 | 2.201 | .838 | -10.37 | 5.04 | ||
D | 1.833 | 1.302 | .722 | -2.51 | 6.18 | ||
D | A | -4.000 | 1.538 | .176 | -9.34 | 1.34 | |
B | -4.500 | 2.078 | .352 | -12.22 | 3.22 | ||
C | -1.833 | 1.302 | .722 | -6.18 | 2.51 | ||
*. The mean difference is significant at the 0.05 level. | |||||||
实验结果分析:
由上述结果图可以看出,虽然奇性结果相伴概率等于0.05满足了方差检验的前提条件,在ANOVA图中,相伴概率也大于显著性水平,表示,四种轮胎中方差显著区别不大,但是,在多重比较结果图里,有4个组之间的相伴概率都小于显著性水平,而且,在各组均值的折线图里,我们也可以看出四组存在了显著性差异。所以四种不同类型轮胎的寿命间存在显著性差异。
题目三:将4种不同的水稻品种A1,A2,A3,A4安排在面积相同的4种不同土质的地块B1,B2,B3,B4中试种,测得各地块的产量(kg)如表6.19(数据文件为data6-6.sav),试分别在显著性水平为0.05和0.01下检验不同水稻品种、不同土质及二者交互作用对水稻产量的影响。(数据来源:《SPSS实用统计分析》郝黎仁,中国水利水电出版社)
表6.19 四种水稻的产量数据
B1 | B2 | B3 | B4 | |
A1 | 135 | 120 | 147 | 132 |
A2 | 154 | 129 | 125 | 125 |
A3 | 125 | 129 | 120 | 133 |
A4 | 115 | 124 | 119 | 123 |
实验结果截图:
Warnings |
Post hoc tests are not performed for 水稻 because error term has zero degrees of freedom. |
Post hoc tests are not performed for 土地 because error term has zero degrees of freedom. |
Between-Subjects Factors | |||
Value Label | N | ||
水稻 | 1 | A1 | 4 |
2 | A2 | 4 | |
3 | A3 | 4 | |
4 | A4 | 4 | |
土地 | 1 | B1 | 4 |
2 | B2 | 4 | |
3 | B3 | 4 | |
4 | B4 | 4 | |
Tests of Between-Subjects Effects | |||||
Dependent Variable:产量 | |||||
Source | Type III Sum of Squares | df | Mean Square | F | Sig. |
Corrected Model | 1571.938a | 15 | 104.796 | . | . |
Intercept | 263939.062 | 1 | 263939.062 | . | . |
rice | 474.687 | 3 | 158.229 | . | . |
soil | 94.688 | 3 | 31.563 | . | . |
rice * soil | 1002.562 | 9 | 111.396 | . | . |
Error | .000 | 0 | . | ||
Total | 265511.000 | 16 | |||
Corrected Total | 1571.938 | 15 | |||
a. R Squared = 1.000 (Adjusted R Squared = .) | |||||
1. 水稻 | ||||
Dependent Variable:产量 | ||||
水稻 | Mean | Std. Error | 99% Confidence Interval | |
Lower Bound | Upper Bound | |||
A1 | 133.500 | . | . | . |
A2 | 133.250 | . | . | . |
A3 | 126.750 | . | . | . |
A4 | 120.250 | . | . | . |
2. 土地 | ||||
Dependent Variable:产量 | ||||
土地 | Mean | Std. Error | 99% Confidence Interval | |
Lower Bound | Upper Bound | |||
B1 | 132.250 | . | . | . |
B2 | 125.500 | . | . | . |
B3 | 127.750 | . | . | . |
B4 | 128.250 | . | . | . |
3. 水稻 * 土地 | |||||
Dependent Variable:产量 | |||||
水稻 | 土地 | Mean | Std. Error | 99% Confidence Interval | |
Lower Bound | Upper Bound | ||||
A1 | B1 | 135.000 | . | . | . |
B2 | 120.000 | . | . | . | |
B3 | 147.000 | . | . | . | |
B4 | 132.000 | . | . | . | |
A2 | B1 | 154.000 | . | . | . |
B2 | 129.000 | . | . | . | |
B3 | 125.000 | . | . | . | |
B4 | 125.000 | . | . | . | |
A3 | B1 | 125.000 | . | . | . |
B2 | 129.000 | . | . | . | |
B3 | 120.000 | . | . | . | |
B4 | 133.000 | . | . | . | |
A4 | B1 | 115.000 | . | . | . |
B2 | 124.000 | . | . | . | |
B3 | 119.000 | . | . | . | |
B4 | 123.000 | . | . | . | |
实验结果分析:
由上述各种结果图可以看出不同水稻品种、不同土质对水稻产量的影响差异性显著,而且,两控制变量对观测变量的交互作用图中,两因素相交,说明了有交互作用的影响。同理可得在显著性水平为0.01下也有上述结论。
题目四:某超市将同一种商品做3种不同的包装(A)并摆放在3个不同的货架区(B)进行销售试验,随机抽取3天的销售量作为样本,具体资料见表6.20。要求检验:在显著性水平0.05下商品包装、摆放位置及其搭配对销售情况是否有显著性影响。(数据来源:《应用统计学》 耿修林,科学出版社;数据文件:data6-7.sav)
发布评论