Saturday, May 12, 2012

Confusing Correlation with Causation

When conducting any type of analysis, it is important to distinguish between activities that are correlated, with activities that are causally related. For example, it remains difficult to say for certain whether widespread use of broadband access "causes" economic activity, or whether places where there is a lot of economic activity drive use of broadband.

That can matter quite a lot whenever businesses or policymakers have to allocate capital for the purposes of stimulating economic growth, for example.

ABI Research has found, for example, that mobile users who download a retailer-branded app said the app caused them to visit the store more (45.8 percent), buy more of the store/brand’s products and services (40.4 percent), tell a friend about their store shopping experience (35.8 percent), and encourage friends to visit the store (30.8 percent), according to ABI Research.

The issue is whether those respondents already were patrons of stores who provided an app. It might be that the greatest benefit of a particular retail app to a particular shopper is the fact that the shopper already was spending significant money at a retail outlet.

In that case, the app does not cause shopping behavior, but only increases engagement with a retailer that already had high significance for the user. But a reasonable conclusion would be that a mobile app can increase spending and engagement, under some conditions, when a relationship already exists.

No comments:

AI "Performance Plateau" is to be Expected

There is much talk now about generative artificial intelligence model improvement rates slowing. But such slowdowns are common for most--if...