The Bakshy et al. study found that individual choice was the larger filter reducing cross-cutting exposure, with algorithmic effects being smaller than user self-selection.

[UNDERMINE → premise #1] Bakshy et al. (2015) found that both algorithmic ranking and individual choice reduce cross-cutting exposure, but user choice was the larger filter. The study's own conclusion emphasizes that individual choice dominates algorithmic filtering in determining exposure patterns. Therefore, The Bakshy et al. study found that individual choice was the larger filter reducing cross-cutting exposure, with algorithmic effects being smaller than user self-selection. (Warrant: When a study finds that homophily effects exceed algorithmic effects, citing it as evidence that algorithms contribute 'beyond homophily' understates the relative magnitude of user self-selection.)

• Bakshy et al. (2015) found that both algorithmic ranking and individual choice reduce cross-cutting exposure, but user choice was the larger filter.
• The study's own conclusion emphasizes that individual choice dominates algorithmic filtering in determining exposure patterns.

These are challenges this argument’s reasoning pattern must still withstand. Answering them on Isonomia strengthens the argument.

• Are there other events or causes that could account for the presence of the sign?Open
• What is the strength of the correlation between the sign and the event/state it is taken to indicate?Open
• Has the sign actually been observed reliably in this case?Open

Cited by 1 · 1 contest