Class Notes (1,200,000)
CA (650,000)
UTSG (50,000)
SOC (3,000)


Course Code
Alexandra Marin

This preview shows half of the first page. to view the full 3 pages of the document.
Experiments are done in social psychology subfield of soc and psych different but related
Why first thing, really important bc bold standard for showing causality,
Research are explanatory , exp we talk about sub, treatment what applied to subject
Group 1 or more groups treatment applied and control group assigned randomly
Elegible sec 8, sub housing in us, took list applied divided random 3 group 1 sec 8 voucher,
another moving to opp less than 10 poverty, sec 8 evrywhere, control nothing
Treatment and control
Assin random
Apply treatment what happens how outcomes differ treatment exp group and treatment
Yet correlation time non spur
Show cor, behavior improved, not in ones no voucher
Time order, pretest, postets, before and after treatment.
Spurious vain research, some 3rd thing. Cause boy beh improve and move to better nigh, in
this case home was random
Random rule out spuriousness
Groups differ even randomly, could be same, fluke, stat calc prop of fluke, 2 guards random
and possib existed and ass random better stat to non random
Have to have control group
Have to have pretreatment measure, different begin or how they change not all have both.
Logic of exp, counter factual say something causes siemthing drop purse let go, no let go not
drop, imagine other sit not happen never observe it, never see person do what they havent
Cannot observe counter fct cannot see causal
In exp can because two groups same rel perspective 1 group treatment, not apply have
counter factual not 1 on 1
Cannot take two different people different to start with
You're Reading a Preview

Unlock to view full version