Now that you have some experience with the assumptions needed for causal inference as well as familiarity with a few causal estimands, we need to set up a structured way to apply what you’ve learned when approaching causal inference problems.

We can think of causal inference as a two-step process.

  1. Identification: During this stage we determine which causal estimand we will estimate based both on what we want to know and on what we are able to compute. We must also determine whether we will be able to meet the three assumptions in order to infer that the relationship is causal in nature.
  2. Estimation: Now that we’ve reasoned what can compute and that this measure will reflect a causal relationship between variables, we must carry out the statistical model to compute the treatment effect. We used some simple computations throughout this lesson, but we will soon use more complicated methods like multiple linear regression to obtain our estimand.


The flowchart in the learning environment summarizes the two steps in the causal inference process: Identification and Estimation. Think about these steps as we would have applied them in our therapy animal example.

  • Estimand: Are we interested in computing the ATE for all patients, or do we only care about the effect of therapy animals on the patients who were treated (ATT)? Are we able to compute the ATE in this case? When might we NOT be able to compute the ATE?
  • Assumptions:
  1. Conditional Exchangeability: Are the treatment and control groups exchangeable? Or do we need to account for other measures of confounding beyond just anxiety diagnosis?
  2. SUTVA: Does one patient receiving the therapy influence other patients to choose to receive the therapy? Is every patient receiving the same services (same frequency, same amount of time, same kind of animal)?
  3. Overlap: Does every patient have some probability of being in either therapy group? Or do some types of patients have no probability of being treated, such as those with allergies or those under age 18?
  • Which statistical model will we use to estimate the treatment effect?
  • Are there any additional assumptions that need to be met to use this technique?

Take this course for free

Mini Info Outline Icon
By signing up for Codecademy, you agree to Codecademy's Terms of Service & Privacy Policy.

Or sign up using:

Already have an account?