In a message dated 9/1/98 11:49:42 AM, firstname.lastname@example.org wrote:
>They are trying to use the timing of variables to infer causality;
>things that are measured earlier likely cause, and are not caused by,
>things that are measured later. So to disambiguate cause, they measure
>soc/psych both before and after internet use. Of course there's always
>the possiblity of unmeasured variables causing everything, such as
>unmeasured "during" soc/psych. But it seems to me they have done the
>reasonable thing with the data the have.
Even by that standard, then, they've overreached in deducing causation. Neither the family communication-usage nor depression-usage correlations changed significantly from pre to post, although loneliness probably did.
Also, the *change* in psychological state is contemporaneous with internet use, and so with respect to that as a causative factor, they don't have temporal control. If *losing* friends makes people turn to the internet (as opposed to not having them in the first place) then you get the reported results, even with the reverse of the inferred causation.
>Huh? They used log of internet hours, without truncation.
Oops, my bad.