spss缺失值分析概要1.docxVIP

第一组 学号：201521430049 严安2016春季 高级心理统计 第一次作业A．异常值分析（HBAT.sav）语句1：EXAMINE VARIABLES=x6 /PLOT STEMLEAF HISTOGRAM NPPLOT /STATISTICS DESCRIPTIVES /CINTERVAL 95 /MISSING LISTWISE /NOTOTAL.解释：生成变量x6的描述统计表（表1）、正态检验表（表2）、直方图（图1）、茎叶图、Q-Q图等。结果发现，x6集中在中心（Skewness=-0.245），但较为平坦（Kurtosis=-1.132）；K-S检验显示x6分布不是正态分布。表1DescriptivesStatisticStd. ErrorX6 - Product QualityMean7.810.139695% Confidence Interval for MeanLower Bound7.533Upper Bound8.0875% Trimmed Mean7.833Median8.000Variance1.950Std. Deviation1.3963Minimum5.0Maximum10.0Range5.0Interquartile Range2.6Skewness-.245.241Kurtosis-1.132.478表2Tests of NormalityKolmogorov-SmirnovaShapiro-WilkStatisticdfSig.StatisticdfSig.X6 - Product Quality.109100.005.950100.001a. Lilliefors Significance Correction图1语句2：EXAMINE VARIABLES=x6 BY x1 /PLOT=BOXPLOT/STATISTICS=NONE/NOTOTAL /MISSING=REPORT.EXAMINE VARIABLES=x7 BY x1 /PLOT=BOXPLOT/STATISTICS=NONE/NOTOTAL /MISSING=REPORT.解释：根据x1把x6和x7分别分成三组进行统计分析，并作出箱式图。以x6为例见表3、图2。表3Case Processing SummaryX1 - Customer TypeCasesValidMissingTotalNPercentNPercentNPercentX6 - Product QualityLess than 1 year32100.0%00.0%32100.0%1 to 5 years35100.0%00.0%35100.0%Over 5 years33100.0%00.0%33100.0%图2语句3：ONEWAYx6 x7 BY x1 /MISSING ANALYSIS /POSTHOC = SCHEFFE ALPHA(.05).解释：根据x1分别对x6和x7进行分组方差分析，并进行事后检验。（表略）B．缺失值分析（HBAT_MISSING.sav）语句1：Count nmiss = v1 to v14 (missing) .freq nmiss .解释：统计每个被试出现缺失值数的频数、百分比。数据表中生成被试缺失值频数列。结果发现，缺失值最多的被试共缺失7个值，这样的被试有6个（表4）表4nmissFrequencyPercentValid PercentCumulative PercentValid.002637.137.137.11.001521.421.458.62.001927.127.185.73.0045.75.791.47.0068.68.6100.0Total70100.0100.0语句2：MVA v1 v2 v3 v4 v5 v6 v7 v8 v9 v10 v11 v12 v13 v14 /MAXCAT = 25 /ID = id /CATEGORICAL = v10 v11 v12 v13 v14解释：统计每个连续变量的有效值的频数、均值、标准差以及缺失值的频数、百分比。结果表明变量v1的缺失值最大，缺失比例达30%（表5）。表5Univariate StatisticsNMeanStd. DeviationMissingNo. of ExtremesaCountPercentLowHighv1494.008000v2571.944600v3538.0621300v4635.1681.1714710.000v5612.856.7760912.900v6642.611.717468.600v7616.823