- This topic has 3 replies, 2 voices, and was last updated 4 months, 3 weeks ago by
.
-
Posts
-
Hello. We are leading a teacher PD on Statistics, and while using the random sampler, one participant was wondering if you can see which samples are pulled when they random sample and then compare them to the others that were left, or compare them to the full dataset.
Hello Taryn,
I don’t fully understand the question. Is this using the Collector device in the Sampler? If so, you could add an attribute to the source dataset that uniquely identified each case. This would let you see which ones got sampled. But I’m not sure what is meant by “compare them to the full dataset.”
Bill
Hi Bill,
Yes, sort of. We used the collector in the sampler plugin and got a random sample of 10. We wanted to pull those 10 out of the total population and compare the mean of the sample with the mean of the total population without the 10. We can do it manually finding each observation from the sample and setting it aside in the main dataset, but that can be quite tedious. So we were wondering if there was a way to highlight all of the observations the sampler pulled in the larger dataset.
Ah. Yes, there is. Here is a shared document that demonstrates a technique that might work for you. Basically, we define an attribute in the source that attempts to look up cases in the sample collection and returns true if found and false if not found. It’s a little tricky. Let me know if it works for your situation?
Bill
- You must be logged in to reply to this topic.