|
1、读入数据的程序:
DATA LIST LIST
/id (F3) Interview_date (ADATE10) Age (F3) Gender (A1)
Income_category (F1) Religion (F1) opinion1 to opinion4 (4F1).
解释:要读入数据的变量,例如:id为变量名,后面的(F3)表示数字型的程度为3(可根据需要设计),依次类推A1表示字符的长度为1
BEGIN DATA(开始读入数据)
150 11/1/2002 55 m 3 4 5 1 3 1
272 10/24/02 25 f 3 9 2 3 4 3
299 10-24-02 900 f 8 4 2 9 3 4
227 10/29/2002 62 m 9 4 2 3 5 3
216 10/26/2002 39 F 7 3 9 3 2 1
228 10/30/2002 24 f 4 2 3 5 1 5
333 10/29/2002 30 m 2 3 5 1 2 3
385 10/24/2002 23 m 4 4 3 3 9 2
170 10/21/2002 29 f 4 2 2 2 2 5
391 10/21/2002 58 m 1 3 5 1 5 3
END DATA.(结束读入)
2、对变量添加名称(简单的说就是数据库中变量的意思是什么?)
VARIABLE LABELS
Interview_date "Interview date"
Income_category "Income category"
opinion1 "Would buy this product"
opinion2 "Would recommend this product to others"
opinion3 "Price is reasonable"
opinion4 "Better than a poke in the eye with a sharp stick".
解析:
VARIABLE LABELS
变量名( Interview_date ) 要赋予的变量名称(Interview date)
3、为变量中的数值添加lable
VALUE LABELS
Gender "m" "Male" "f" "Female"(对字符型赋值)
/Income_category 1 "Under 25K" 2 "25K to 49K" 3 "50K to 74K" 4 "75K+"
7 "Refused to answer" 8 "Don't know" 9 "No answer"
/Religion 1 "Catholic" 2 "Protestant" 3 "Jewish" 4 "Other" 9 "No answer"
/opinion1 TO opinion4 1 "Strongly Disagree" 2 "Disagree" 3 "Ambivalent"
4 "Agree" 5 "Strongly Agree" 9 "No answer".
解析:
VALUE LABELS
Gender "m" "Male" "f" "Female"(对字符型赋值Gernder表示要赋值的变量名,比如运行以后表示“F”代表female,“M”表示Male)
数据型的:Income_category 1 "Under 25K" 2 "25K to 49K" 3 "50K to 74K" 4 "75K+"(对数值型赋值,Income_category 表示要赋值的变量名,比如运行以后表示1代表Under 25K,2表示25K to 49K,依次类推)
4、缺失值处理:
MISSING VALUES
Income_category (7, 8, 9)
Religion opinion1 TO opinion4 (9).
解析:MISSING VALUES Income_category (7, 8, 9)(表示将Income_category变量中的7\8\9认为是缺失值,可根据自己的目的处理,自行修改)
5、修改变量的类型:
VARIABLE LEVEL
Income_category, opinion1 to opinion4 (ORDINAL)
Religion (NOMINAL).
6、检查
DISPLAY DICTIONARY.(运行即可,看见数据库的所有信息)
解析:VARIABLE LEVEL Income_category, opinion1 to opinion4 (ORDINAL)(表示把Income_category,opinion1 to opinion4的变量改为定序变量)可根据分析的进行修改
[ 本帖转载自中国统计网 ] |
|