0
$\begingroup$

I need to carry out a comparison test for the measures of absorbance between 2 types of tank: glass tank and plastic tank. The technician in the laboratory took randomly 5 plastic tanks and 5 glass tanks for the measurement. The absorbance in each tank will be measured 6 times. So i will have in total 60 measures of 10 tanks to do the test. Here is my sample data set:


ID tank test measure
1 plastic 1 xxx
2 glass 1 xxx
3 plastic 1 xxx
.........
9 plastic 6 xxx
10 glass 6 xxx

And here is my proc mixed that i used:

proc mixed data=data1;
class tank;
model measure=tank;
random id(tank);
repeated id(tank);
run;

The goal of the experiment is to justify the difference of the absorbance measure between 2 types of tanks. Can you tell me what do you think about my mixed model please?

$\endgroup$

closed as off-topic by Peter Flom Apr 27 '17 at 12:07

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – Peter Flom
If this question can be reworded to fit the rules in the help center, please edit the question.

1
$\begingroup$

In your situation, assuming the correlations between the 6 measurements from the same tank are the same, (it is reasonable assumption) you do not need both repeat and random. Just keep one of them, like this one.

proc mixed data=data1;

class id tank;

model measure=tank;

random intercept/subject = id;

run;

$\endgroup$
  • $\begingroup$ Thank you. Can i ask you why we should put "intercept" in the random statement please? $\endgroup$ – buiquanghai Apr 27 '17 at 11:44
  • $\begingroup$ Because you want a random intercept model. That is, one that lets the intercept vary by subject. $\endgroup$ – Peter Flom Apr 27 '17 at 12:05
  • $\begingroup$ Because "assuming the correlations between the 6 measurements from the same tank are the same", so one method is using "random intercept/subject=id". Another method is replace "random intercept/subject=id;" by "repeated/subject=id type = cs;". The results should be exact the same. $\endgroup$ – user158565 May 4 '17 at 19:44

Not the answer you're looking for? Browse other questions tagged or ask your own question.