I'm very new in the field I'm working now and have a simple question:

If I'm given such following data:

conds <- rep(c("old","young","kid"),each=9)
measures <- rep(rep(c("weight","height","heartbeats"),each=3),3)
values <- c(sample(50:100,9),sample(150:190,9),sample(60:120,9),
            sample(50:110,9),sample(150:200,9),sample(60:120,9),
            sample(5:50,9),sample(50:150,9),sample(60:100,9))
d <- data.frame(measures,conds,values)

How can I do test the differences between old vs young, old vs kid and kid vs young considering all other measures together?

These are two ways I used to find the differences between each pairs of old-young-kid based on the observed measures but I think non of these are correct:

aov.out = aov(values ~ conds , data=d)
summary(aov.out)
               Df Sum Sq Mean Sq F value Pr(>F)  
    conds        2  14515    7257   3.643 0.0415 *
    Residuals   24  47811    1992                 
    ---
    Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
     TukeyHSD(aov.out)
      Tukey multiple comparisons of means
        95% family-wise confidence level

    Fit: aov(formula = values ~ conds, data = d)

    $conds
                   diff        lwr       upr     p adj
    old-kid   47.000000  -5.543622  99.54362 0.0856646
    young-kid 51.111111  -1.432511 103.65473 0.0576516
    young-old  4.111111 -48.432511  56.65473 0.9791911

I decided to reorder my data and find the differences in the following way:

conds <- rep(c("old","young","kid"),each=3)
w <- values[c(1:3,10:12,19:21)]
h <- values[c(4:6,13:15,22:24)]
hb <- values[c(7:9,16:18,25:27)]
d <- data.frame(conds,w,h,hb)
av.out <- aov(w*h*hb~conds,d)
     summary(av.out)
                Df    Sum Sq   Mean Sq F value  Pr(>F)   
    conds        2 3.075e+12 1.537e+12    13.4 0.00612 **
    Residuals    6 6.885e+11 1.147e+11                   
    ---
    Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
     TukeyHSD(av.out)
      Tukey multiple comparisons of means
        95% family-wise confidence level

    Fit: aov(formula = w * h * hb ~ conds, data = d)

    $conds
                 diff     lwr     upr     p adj
    old-kid   1172527  323913 2021141 0.0128705
    young-kid 1297829  449215 2146443 0.0079956
    young-old  125302 -723312  973916 0.8949227

Would you let me know which is correct? if none of them are correct then how could I find the differences?

Thanks,

Maah