Correlation Is Not Causation
A colleague called yesterday to discuss one community’s spay/neuter program. It had been noted in this community that a large TNR program had significantly decreased the amount of S/N they were able to do a bit over a year ago. Over the past year, the intake of cats had significantly increased – and the community was pointing to the decrease in TNR for this increase. How awesome that they are thinking about these data points and how they might relate to one another!
One of the first things one learns in a class on research methods is “correlation is not causation.” Meaning, just because 2 things happen in relation to one another, it does not mean that one necessarily causes the other. Certainly, if one thing causes the other, there will be correlation, but just because there may be correlation between sleeping with your shoes on and a headache, it might not be that the shoes cause the headache (it may instead be the behaviors earlier in the evening…).
Going back to the community my colleague called me regarding, unfortunately there is much that we need to uncover before we can move past correlation to causation. And, as much as it kills me, it is likely that we will not be able to get to causation at this point, as many variables are likely correlated with the increase in intake of cats. First, we need to break out type of cat intake – free roaming vs. owner relinquished. If the intake of owned animals increased, we cannot easily point to a decrease in TNR for causation. We also need to break type out another way – neonate, juvenile, adult. We would expect that the population to first increase to be neonate and juveniles when we cease a S/N program. If there is no increase in kittens, it becomes more difficult to argue that the decrease in TNR is causing the increase in intake.
With a new shelter in this community and lots of new advertising and programming, it is just as likely that the increased public presence increased intake – which simply “gets in the way” of us being able to measure the impact of S/N, especially retrospectively. What does this mean to you?
As an industry, we tend to take the kitchen sink approach – meaning, we throw everything at a problem in hopes we can save lives, and do not often stop to think about measuring impact. Who can blame us – as each pause may mean another life lost? The problem is, if we do save lives, how do we know what program or process had a real effect? With the tremendous urgency we have to affect change, and the limited resources we have to do so, measuring the change can be a challenge – and it is a vital piece of the puzzle that we must address so that we can truly learn to apply program and process to save lives.
Tags: data, intake, research, spay/neuter, TNR
You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
Leave a Reply
We appreciate your participation. Your comment will be published once it's approved. Please read our comment policy.