去势正态分布Q-Q图-2 Detrended Normal Q-Q Plot of空腹血清胰岛素样 男性 2.0 1.5 10期望正态分布直线 口 00 Observed value
Detrended Normal Q-Q Plot of 空腹血清胰岛素样生长因子-1 男性 Observed Value 0 200 400 600 800 1000 Dev from Normal 2.0 1.5 1.0 .5 0.0 -.5 去势正态分布Q-Q图-2 期望正态分布直线
均数箱图 1200 奇异值 昏1000 15 16 800 600 木 400 P25P75 200 P5均值 -200 女性 男性 性别
N = 66 83 性 别 女性 男性 空腹血清胰岛素样生长因子- 1 1200 1000 800 600 400 200 0 -200 115 79 142 78 16 15 均数箱图 奇异值 P50均值 P25-P75
分布自然对数-水平自然对数散点图 Spread vs Level Plot of IGF_ 1F By sEX 5.4 5.2 5.1 5.0 5.2 5.5 Plot of Ln of Spread vs ln of Level Slope =2.063 Power for transformation =-1.063 数据转换力度不接近0,表示要使两组方差齐性,需做数据转换
Spread vs. Level Plot of IGF_1F By SEX * Plot of LN of Spread vs LN of Level Slope = 2.063 Power for transformation = -1.063 Level 5.2 5.3 5.4 5.5 Spread 5.5 5.4 5.3 5.2 5.1 5.0 分布自然对数-水平自然对数散点图 数据转换力度不接近0，表示要使两组方差齐性，需做数据转换
数据转换ma Descriptive Continue a Factor levels together M Stem-and-leaf C Dependents together F Histogram Cancel Help v Normality plots with tests Spread vs Level with Levene Test C Non 选择自然 C Power estimation 基于均数的显 对数转换 C Transformed Power:Natural log 著性为0.135> C Untransformed 0.05,表示接 受两组均数方 差的假设 经数据转换后的方差正态齐性检验 Test of Homogeneity of variance Levene Statistic df1 df2 si 空腹血清胰岛 Based on mean 2.262 147 135 素样生长因子1 Based on median 2.422 147 122 Based on median and with adjusted df 2.422 1142127 122 Based on trimmed mean 2.351 147 127
数据转换 Test of Homogeneity of Variance 2.262 1 147 .135 2.422 1 147 .122 2.422 1 142.127 .122 2.351 1 147 .127 Based on Mean Based on Median Based on Median and with adjusted df Based on trimmed mean 空腹血 清胰岛 素样生 长因子-1 Levene Statistic df1 df2 Sig. 经数据转换后的方差正态齐性检验 选择自然 对数转换 基于均数的显 著性为0.135> 0.05，表示接 受两组均数方 差的假设
Spread vs Level Plot of IGF_1F By SEX 8 Leve Data transformed using P Slope = 893
Spread vs. Level Plot of IGF_1F By SEX * Data transformed using P = Slope = .893 Level 5.2 5.3 5.4 5.5 Spread 1.0 .9 .8 .7
先进行IGF-1F数据转换(自然对数) IE Sample 53. sav -SPSS Data Editor File Edit View Data Transform Analyze Graphs utilities Window Help 回。國蝇面国 149: igf_1f 281.14 64 33925 05555 83 205.04 00000000000000000000000 294.24 5.8 120.37 4.79 19195 5.26 4.76 4.98 01234567 20757 5 54 41986 6.04 621.73 6 689 841 6.74 33086580 29 27367 5.61 79075.19 215.55 5.37 20372532 aDNData Vier Variable View T oSS Processor is ready
先进行IGF-1F数据转换（自然对数）
数据转换(Ln)后 Descriptive 别 Statistic Std. Emol 的 LNIGF1 0 Mean 545447798E02 描述性统计指标 95% Confidence Lower Bound 5.2987 Interval for Mean Upper Bound 5.6102 5% Trimmed Mean 5.4532 54346 Std. deviatio Minimum 3.98 Maximum 689 Range 2.91 Interquart ile range Skewness 046 295 Kurtosis -.377 Mean 530935727E-02 95%Confidence Lower bound 5.1954 Interval for Mean Upper Bound 5.4233 5%Trimmed Mean 5.2899 5.2468 Variance .272 Std. deviation 5217 Minimum 4.39 6.71 2.32 Interquart ile Range .7591 Skewness 264 Kurtosis 329
Descriptives 5.4544 7.798E-02 5.2987 5.6102 5.4532 5.4346 .401 .6335 3.98 6.89 2.91 .9268 -.046 .295 -.377 .582 5.3093 5.727E-02 5.1954 5.4233 5.2899 5.2468 .272 .5217 4.39 6.71 2.32 .7591 .481 .264 -.329 .523 Mean Lower Bound Upper Bound 95% Confidence Interval for Mean 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis Mean Lower Bound Upper Bound 95% Confidence Interval for Mean 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis 性 别 0 1 LNIGF1 Statistic Std. Error 数据转换（Ln）后 的 描述性统计指标
数据转换(Ln)后的4种似然估算值 M-Estimators Hubers Tukey's. Hampels Andrews' 性别 M-Estimator| Biweight M-Estimator LNIGF1 0 5.4589 5.4531 545515.4531 5.2617 5.2538 2698 a. the weighting constant is 1. 339 b. The weighting constant is 4.685 C. The weighting constants are 1.700 3 400, and 8.500 d. the weighting constant is 1.340*pi 数据转换(Ln)后的百分位值及 Tukeys折点 Percentiles percentile 性别 5 25 50 75 90 Weighted LNIGF1 0 437534.5638499215434659189623986.5137 Average Definition 1) 456984.63524.8999 5.2468 5.6590 6.0298 6.2756 Tukey's Hinges LNIGF1 0 49922543465.9168 4.9146 5.2468 5.6465
M-Estimators 5.4589 5.4531 5.4551 5.4531 5.2617 5.2538 5.2698 5.2534 性 别 0 1 LNIGF1 Huber's M-Estimatora Tukey's Biweightb Hampel's M-Estimatorc Andrews' Waved a. The weighting constant is 1.339. b. The weighting constant is 4.685. c. The weighting constants are 1.700, 3.400, and 8.500 d. The weighting constant is 1.340*pi. Percentiles 4.3753 4.5638 4.9921 5.4346 5.9189 6.2398 6.5137 4.5698 4.6352 4.8999 5.2468 5.6590 6.0298 6.2756 4.9922 5.4346 5.9168 4.9146 5.2468 5.6465 性 别 0 1 0 1 LNIGF1 LNIGF1 Weighted Average(Definition 1) Tukey's Hinges 5 1 0 2 5 5 0 7 5 9 0 9 5 Percentiles 数据转换（Ln）后的百分位值及Tukey’s折点 数据转换（Ln）后的4种似然估算值
数据转换(Ln)后的奇异值 Extreme values 性别 Value LNIGF1 0 Highest 1 15 6.89 2 16 27 6.53 29 6.48 5 14 6.43 Lowest 1 26 3.98 48 4.31 60 4.36 4 35 4.40 5 42 4.42 Highest 1 78 6.71 2 142 6.59 79 6.34 115 6.31 5 117 6.15 Lowest 1 130 4.39 2 123 127 4.54 143 4.56 5 111 4.60
Extreme Values 1 5 6.89 1 6 6.74 2 7 6.53 2 9 6.48 1 4 6.43 2 6 3.98 4 8 4.31 6 0 4.36 3 5 4.40 4 2 4.42 7 8 6.71 142 6.59 7 9 6.34 115 6.31 117 6.15 130 4.39 123 4.47 127 4.54 143 4.56 111 4.60 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Highest Lowest Highest Lowest 性 别 0 1 LNIGF1 Case Number Value 数据转换（Ln）后的奇异值
数据转换(Ln)后的正态检验 Tests of Normality Kolmogorov-SmimoV 性别|Stti Sig LNIGF1 0 042 66 200* 078 83 200* This is a lower bound of the true significance a. Lilliefors Significance Correction 两组间 方差齐性 数据转换(Ln)后的方差正态齐性检验 P>0.05 Test of Homo geneity of variance Levene Stat ist ic df1 d f2 Sig LNIGF1 Based on mean 2.262 147 135 Based on median 2.422 147 122 Based on median and with adjusted df 2.422 1142.127 122 Based on trimmed mean 2.351 147 127
Tests of Normality .042 6 6 .200* .078 8 3 .200* 性 别 0 1 LNIGF1 Statistic d f Sig. Kolmogorov-Smirnova *. This is a lower bound of the true significance. a. Lilliefors Significance Correction Test of Homogeneity of Variance 2.262 1 147 .135 2.422 1 147 .122 2.422 1 142.127 .122 2.351 1 147 .127 Based on Mean Based on Median Based on Median and with adjusted df Based on trimmed mean LNIGF1 Levene Statistic df1 df2 Sig. 数据转换（Ln）后的正态检验 数据转换（Ln）后的方差正态齐性检验 两组间 方差齐性 P>0.05