Get 1 week of unlimited access
Class Notes (1,034,030)
US (405,976)
MiamiOH (2,634)
SOC (24)
SOC 153 (24)
Lecture 3

SOC 153 Lecture Notes - Lecture 3: Scientific Control

2 pages100 viewsSpring 2016

Department
Sociology
Course Code
SOC 153
Professor
Steven Nelson
Lecture
3

This preview shows half of the first page. to view the full 2 pages of the document.
Scientific method dramatically improves the reliability and usefulness of our factual
knowledge- more accurate deductions and predictions
Science helps us avoid biases
Science helps us avoid mistakes b/c it uses a set of rules and procedures, also must
submit findings
Science is apolitical- if not for science, then “facts” are determined by those with power
10 features of Science:
1. skepticism- don’t take anyones word for it
2. empiricism- the world is what we see it is- what we detect or measure
3. objectivity- accepting reality without judgement, despite what we expect or wish
4. methodology- using systematized procedures to avoid human biases
5. theory- attempts to understand complex things are made explicit and are supported
over time by successive findings or discarded
6. definition- strict precision of concepts and terms to avoid confusion
7. evidence- all claims must be judged solely by date presented
8. logic- claims are argued within the rules of formal logic and mathematics
9. falsification- subjecting all explanations to critical tests designed to disprove them
10. publication- making your findings and procedures open for the public critique and
collaboration
Validity- are you really measuring what it's saying you’re measuring? food stamps don’t
measure poverty
Reliability- is it predictable or unpredictable?
Accuracy- ballpark answer (about $100,000) think target all shots to middle not close
Precision- very precise, but inaccurate (exact answer family earns $95,740) think targets
all shots are very close together
Causation:
if X happens/exists then Y will happen
thing that is causing something to happen(independant variable)
thing that is happening(dependant variable)
3 Conditions necessary to demonstrate causation: X------>Y
1. correlation: in order for X to cause Y then X has to be correlated to Y- occur together or
absent together
2. temporal priority/ correct directionality- Ex: studying happens first then get good test
grades.
3. non-spuriousness- show that X and Y are not accidentally together because of some Z;
trying to show why two things are correlated
a. Ex: eating ice cream and drowning incidents. Together they seem like they have
no correlation. X=ice cream Y=drowning Z=summer(causes ice cream and
drowning)
How we can demonstrate non-spuriousness of X&Y’s relationship
arguing tons of different Z’s
statistically controlling for specific group of variables(control variables)
doing a controlled experiment (which eliminates all spurious variables)
Probability and Prediction:
You're Reading a Preview

Unlock to view full version


Loved by over 2.2 million students

Over 90% improved by at least one letter grade.