Session Ready
Exercise

Power analysis August

In the video, we ran a power analysis assuming we were going to run the experiment in January. Run a new power analysis assuming we'll run the experiment in August. To compute the conversion rate for our control and test you'll need to look at the dataset. Be sure to round all values to the hundredth of a percent (e.g., 0.13453 to 0.13). The data click_data_month is available pre-loaded for you to look up the conversion rate for August.

Instructions
100 XP
  • Fill in the proportion of the sample from the test condition (B), the alpha (alpha), and the power (power).
  • Fill in the conversion rate for our control condition using the conversion rate for August (p1). Remember to round.
  • Fill in the expected conversion rate for our test condition (p2), assuming a 10 percentage point increase (p1 + 0.1). Remember to round.
  • Look at total_sample_size to see how many data points you need in total across both conditions.