交叉表分析主要用来检验两个变量之间是否存在关系，或者说是否独立，其零假设为两个变量之间没有关系。我们在实际的工作中，经常用交叉表来分析比例是否相等。比如我们来分析一下，不同的性别对不同的报纸的选择有什么不同，就是要用交叉表分析了，下面是具体的方法。

方法/步骤

在spss中打开数据，然后依次打开：analyze--descriptive--crosstabs，打开交叉表对话框

将性别放到行列表，将对读物的选择变量放到列，这样就构成了一个交叉表

接下来我们要设置输出的结果，点击statistics，打开一个新的对话框

勾选chi-square（卡方检验），勾选phi and cramer's V（衡量交互分析中两个变量关系强度的指标），点击continue，回到交叉表对话框

​点击cells，设置cell中要展示的数据​

​在这里勾选observed（​各单元格的观测次数），勾选row（行单元格的百分比），点击continue，回到交叉表对话框

​点击ok按钮，输出检验结果​

​先看到的第一个表格就是交叉表，性别为行、选择的读物为列

​卡方检验结果：我们主要是看pearson卡方检验，sig值小于0.05，因此我们认为不同的性别的人对周末读物的选择有显著的差别​

​​最后一个表格，输出的是phi值和V值，两个都是代表两个变量之间的关系的紧密度的，数值小于0.1说明关系不紧密，即性别与周末读物的选择没有明显的关系，这个结论和上面的卡方检验有出入，所以我们需要进一步进行两两比较。

Cross table analysis is mainly used to test the existence of a relationship between two variables, or is independent, the null hypothesis for it doesn't matter between the two variables. We are in actual work,http://www.yeyanmai2.com/post/496.html often with cross table to analyze whether equal proportion. For example, we analyze, choose different gender on different newspapers have what different, is to use cross table analysis, the following are the specific method.

Methods / procedures

Open the data in SPSS, then in turn: analyze--descriptive--crosstabs, cross table dialog box opens

The sex on the list, select variables on the readings on the column, so as to form a cross table

Next we are going to set the output results, click statistics, open a new dialog

Check the chi-square (chi square test),http://www.mkykgw.com/post/484.html check the phi and cramer's V (a measure of interaction analysis of two variables relationship strength index), click continue, back cross table dialog box

If you click cells, you want to display the data set cell

If you check the observed here (we each cell observation times), check the row (percentage cell), click continue,http://www.jiyinyugen1.com/post/331.html back cross table dialog box

If you click the OK button, we output test results

The first table we first see is cross table, sex, selection of books for the column

New results of chi square test: we mainly see Pearson chi square test, SIG value of less than 0.05, so we think different gender have significant differences on the weekend just reading

We present the last table,http://www.sanbenshu.net/post/118.html the output is the phi value and V value, two are representative of the relationship between the two variables tightness,http://www.mnmbgw.com/post/635.html a value of less than 0.1 show the relationship is not close, namely, gender and weekend reading choice is not obvious, and chi square this conclusion and the above test, so we need a further two two.

spss如何做交叉表分析