广告方式与地区与销售额关系的spss分析.docx

广告方式与地区与销售额关系的SPSS分析经济08403班闫晶晶某企业在制定某商品的广告策略时，收集了该商品在不一致地区使用不一致广告形式促销后的销售额数据，分析广告形式与地区是否影响商品销售额。自变量为广告方式(Xl)与地区(X2),因变量为销售额(X3)。涉及地区18个，每个地区抽取样本8个,共有案例144个。下表中，Xl一列中，1表示报纸，2表示广播，3表示宣传品，4表示体验。具体数据如下：广告形式地区销售额1.001.0075.002.001.0069.004.001.0063.003.001.0052.001.002.0057.002.002.0051.004.002.0067.003.002.0061.001.003.0076.002.003.00100.004.003.0085.003.003.0061.001.004.0077.002.004.0090.004.004.0080.003.004.0076.001.005.0075.002.005.0077.004.005.0087.003.005.0057.001.006.0072.002.006.0060.004.006.0062.003.006.0052.001.007.0076.002.007.0033.004.007.0070.003.007.0033.001.008.0081.002.008.0079.004.OO8.0075.003.008.0069.001.009.0063.002.009.0073.004.009.0040.003.009.0060.001.0010.0094.002.0010.00100.004.0010.0064.003.0010.0061.001.0011.0054.002.0011.0061.004.0011.0040.001.0012.0070.002.0012.0068.004.0012.0067.003.0012.0066.001.0013.0087.002.0013.0068.004.0013.0051.003.0011.0041.003.0013.0065.001.0014.0065.002.0014.0063.004.0014.0061.003.0014.0058.001.0015.0065.002.0015.0083.004.0015.0075.003.0015.0050.001.0016.0079.002.0016.0076.004.0016.0064.003.0016.0044.001.0017.0062.002.0017.0073.004.0017.0050.003.0017.0045.001.0018.0075.002.0018.0074.004.0018.0062.003.0018.0058.001.001.0068.002.001.0054.004.OO1.0058.003.001.0041.001.002.0075.002.002.0078.004.002.0082.003.002.0044.001.003.0083.002.003.0079.004.003.0078.003.003.0086.001.004.0066.002.004.0083.004.004.0087.003.004.0075.001.005.0066.002.005.0074.004.005.0070.003.005.0075.001.006.0076.002.006.0069.004.006.0077.003.006.0063.001.007.0070.002.007.0068.004.007.0068.003.007.0052.001.008.0086.002.008.0075.004.008.0061.003.008.0061.001.009.0062.002.009.0065.004.009.0055.003.009.0043.001.0010.0088.002.0010.0070.004.0010.0076.003.0010.0069.001.0011.0056.002.0011.0053.004.0011.0070.003.0011.0043.001.0012.0086.002.0012.0073.004.0012.0077.003.0012.0051.001.0013.0084.002.0013.0079.004.0013.0042.003.0013.0060.001.0014.0077.002.0014.0066.004.0014.0071.003.0014.0052.001.0015.0078.002.0015.0065.004.0015.0065.003.0015.0055.001.0016.0080.002.0016.0081.004.0016.0078.003.0016.0052.001.0017.0062.002.0017.0057.004.0017.0037.003.0017.0045.001.0018.0070.002.0018.0065.004.0018.0083.003.0018.0060.00图1.1数据文件一、单因素方差分析在这里，以商品销售额为观测变量，广告形式与地区为操纵变量，通过单因素方差分析方法分别对广告形式、地区对销售额的影响进行分析。两个单因素方差分析的原假设分别设为：不一致广告形式没有对销售额产生显著影响（即不一致广告形式对销售额的效应同时为0）；不一致地区的销售额没有显著差异（即不一致地区对销售额的效应同时为0）SPSS自动计算出组间方差、组内方差、F统计量与对应的概率P值，完成单因素方差分析的有关计算，并将计算结果输出到SPSS输出窗口中。分析结果如下图所示。ANOVO销售额SumofsquaresdfMeansquareFSigBetweengroups5866.08331955.36113.483.000WithingroupsTotal20303.22226169.306140143145.023图1.3广告形式对销售额的单因素方差分析结果ANOVA销售额SumofSquaresdfMeanSquareFSig.BetweenGroups9265.30617545.0184.062.000WithinGroups16904.000126134.159Total26169.306143图1.4地区对销售额的单因素方差分析结果结果分析：图1.4是广告形式对销售额的单因素方差分析结果。能够看到：观测变量销售额的离差平方与为26169.306；假如仅考虑广告形式单个因素的影响，则销售额总变差中，不一致广告形式可解释的变差为5866.083,抽样误差引起的变差为20303.222,它们的方差分别为1955.361与145.023；所得到的F统计量的观测值为13.483,对应的概率P值近似为0.由于概率值P小于显著性水平0.05,则应该拒绝原假设，认为不一致广告形式对销售额产生了显著影响。能够看到，假如仅考虑地区单个因素的影响，则销售额总变差(26169.306)中不一致地区可解释的变差为9265.306,抽样误差引起的变差为16904.,它们的方差分别为545.018与134.159,所得到的F统计量为4.062,对应的概率P值近似为0.由于概率P值小于显著性水平0.05,因此应该拒绝原假设，认为不一致地区对销售额产生了显著影响。同时对比图1.4与图1.5容易发现：假如从单因素的角度考虑，广告形式对销售额的影响较地区来讲更明显。一、多重比较检验在上述单因素方差分析中，发现不一致广告形式对产品销售额有显著影响，不一致地区的产品销售额存在显著差异，为了进一步研究哪种广告形式的作用比较明显，哪种不明显，对变量进行多重比较检验。输出结果如下：HomogeneousSubsets销售覆NSubsetforalpha=.0512Student-Newman-Keulsa宣醐3656.5556体验3666,6111播3670.8889三3673,2222Sig.1.000.055TukeyHSDa3656.5556体脸3666.6111播3670,8889三3673.2222Sig.1.000.096Scheffea宣传品3656.5556帽3666,6111播3670.8889三3673,2222Sig.1.000.148Meansforgroupsinhomogeneoussubsetsaredisplayed,a.UsesHarmonicMeanSampleSize=36.000.PostHocTestsMultipleCoinpaiisosDePendentVariabIe销售额(I)广告形式(J)广告形式MeanDifference(I-J)Std.ErrorSig.95%ConfidenceIntervalLowerBoundUpperBoundTukeyHSD报班广播2.333332.83846.844-5.04719.7138宣传品16.66667*