spss 命令
2012-07-01 22:18阅读：
基本操作
读取数据
读取文件
data list file=’’
fixed/free
/{varname} {format}.
读取数列
data list
/{varname} {format}.
begin data
{data list}
end data.
读取矩阵
matrix data variable
{varname}
/file=’’
/n={n}
/contents=corr.
数据编辑
变量名
变量名不超过8个字符, 名首必须是字母
变量标签
variable labels {varname}
’’.
不能超过40个字符
值标签
value labels {varname} {value}
’’.
不能超过20个字符
形式
format {varname} (f{w.d}).
缺失值
missing values {varname}
().
排序
sort cases by {varname}
(a/b).
数据合并
add files
/file=*
/file=’’
/rename {old varname}={new
varname}
/drop {varname}.
数据转换
Compute
compute {target
varname}={expression}.
算术运算：+,-,*,/,**(幂)
算术函数：sqrt, rnd(四舍五入),
trunc(取整)
统计函数：mean, sum, sd,
max, min
缺失值函数：SYSMIS，MISSING，NMISS，NVALID，VALUE（不算缺失值）
时间函数：
CTIME.DAYS,$JDATE
If
if () {target
varname}={expression}.
Recode
recode {varname} (old
value /else/lowest
through {value}/{value} through
{value}/{value} through high=new
value) into {newvarname}.
Missing values：user-defined 也会被重编码，因此应小心user-defined.
user-defined 不包括在范围内
Split file
Sort cases
by
{varname}
.
split file
layered
by
{varname}
.
split file
off.
Flip
flip variables =
{varname}/all.
Rank
rank variable = {varname}.
数据分析
描述性统计
Frequencies
Frequencies {varname}
/histogram
NORMAL /barchart
(freq/percent)/piechart.
Descriptive
Descriptive {varname}
/statistics=sum mean min
max (集中趋势)
RANGE stddev SEMEAN
VARIANC
(离中趋势)
skewness（偏度）
SESKEW
KURTOSIS（峰度）
SEKURT（形状测量）
/{varname} (z{newvarname}).
Explore
examine {varname}
/plot BOXPLOT
STEAMLEAF
HISTOGRAM(不带有正态曲线)
NPPLOT（证明正态）
/ STATISTICS
DESCRIPTIVES EXTREME
/MESTIMATORS（修正均值）.
Crosstabs
Crosstabs
/tables={row
varname} BY {column
varname}
/Statistic=chisq（默认）
corr
KAPPA（评定者间一致性系数）PHI
CC（修正）
/CELLS= .
相关分析
Graph
scatterplot={varname} with
{varname}.
Person
correlations
/variables={varlist}
/
MISSING=PAIRWISE/LISTWISE（数据足够时更稳定）.
Spearman/kendall(顺序变量)
nonpar corr
/variables={varlist}
/print=spearman(默认)/kendall（有重复）/both.
差异性检验
![spss <wbr>命令]( "spss <wbr>命令")
单样本ｔ检验
t-test testval={value}
/variables={varname}.
独立样本ｔ检验
t-test groups={varname}(value1
value2)
/variables={varname}.
相关样本ｔ检验
t-test pairs={varname} with
{varname}.
曼－惠特尼U检验
npar tests
/ m-w={varname} with { value1
value2}.
维克尔松检验
npar test
/ wilcoxon={varname} with
{varname}.
方差分析
前提假设：独立、等距、正态、同质
数据要求：多元正态，线性(散点图)
Oneway anova（单一自变量平-单一因变量）
oneway
{varlist} by {varname}
/statistics descriptives
homogeneity
/contrast {valuelist}
/posthoc=LSD TURKEY（敏感）BONFERRONI
Scheffe(保守) snk
DUNNETT.
Unianova（多个自变量-单一因变量）
unianova
{dependent varname}
BY {factor
varlist}
/posthoc {varlist}=lsd snk
turkey
/plot=profile({varname/varname*varname})
/desigh={factor
varlist}.
manova
{dependent varname} BY
{factor
varlist}
/DESIGN={FACNAME1} within
{FACNAME2}(FACNUM1)
{FACNAME1} within
{FACNAME2}(FACNUM2)…
Multivariate（多个自变量-多个因变量）
glm
{dependent varlist} BY
{factor
varlist}
/desigh={factor varlist}
/PRINT =
HOMOGENEITY（同质性前提）
/plot=profile({varname/varname*varname})
/posthoc {varlist}=lsd snk
turkey
manova
{dependent varlist}
BY {factor
var
