这个问题已解决.
由gcb1(食物分类取值如何规范化.xlsx生成的）生成gcb2的的code:

data gcb2;
    set gcb1;

    array food(9) food1-food9;
    do i = 1 to 9;
        if kindexc(food(i),"菜","瓜","竹","柿","葱","蒜","卜","椒","菇","姜")  then food(i)="蔬菜";
        else if kindexc(food(i),"肉","骨","脚","腿","翅","鱼","尖")  then food(i)="禽畜肉";
        else if kindexc(food(i),"面","米","粉","包") then food(i)="粮谷类";
        else if kindexc(food(i),"蛋")  then food(i)="鸡蛋";
        else if kindexc(food(i),"奶")  then food(i)="牛奶";
        else if kindexc(food(i),"油")  then food(i)="植物油";
        else if indexc(food(i),"盐")  then food(i)="盐";
        else if indexc(food(i),"抽") then food(i)="生抽";
    end;
     run;


之前出错：将indexc()混为kindexc(),
indexc()只能输出字节所在的位置，
而kindexc()是能输出单个（如果是词也只识别该次第一个字）汉字所在的位置（或者可理解为该函数是识别汉字的）.