Hi ,
For most people, data science is synonymous with machine learning, and many see the role of the data scientist as simply being to build predictive models. Yet, predictive analytics can only get you so far. Predicting what will happen next is great, but what good is knowing the future if you
don’t know how to change it?
That’s where causal analytics can help. However, causal inference is rarely taught as part of traditional prediction-centric data science training. Where it is taught, though, is in the social sciences.
In the latest episode of
Value Driven Data Science, I'm joined by Joanne Rodrigues, author of Product Analytics to discuss how techniques drawn from the social sciences, in particular, causal inference, can be combined with data science techniques to give data scientists the ability to understand and change consumer behaviour at scale.
Highlights include:
- (00:49) Combining social sciences with data science
- (02:01) Joanne's journey from social sciences to data science
- (04:15) Understanding causal inference
- (07:40) Real-world applications of causal inference
- (12:22) Challenges in causal inference
- (19:41) Correlation vs. causation in data
science
- (26:12) Operationalising randomness in experiments
- (27:16) Observational experiments vs. medical trials
- (27:47) Designing experiments with existing data
- (28:50) Challenges in natural experiments
- (29:55) Ethical considerations in experimentation
- (31:50) Qualitative frameworks
in causal inference
- (35:58) Integrating causal inference with machine learning
- (38:59) Common techniques in causal inference
- (41:02) Marketing causal inference to management
- (43:48) Ethical implications of predictive modelling
- (48:08) Final advice for data scientists
You can listen to this episode at the link below, or find it on Apple Podcasts, Amazon Music, Spotify or Google Podcasts.
