Lecture 023

Reading

Political Science is biased by choice of methodology under constraints.

• Solution: debate to resolve bias

Value Freedom:

• epistemic value: good

• political (moral) value: bad

• constitutive value: necessary

• contextual value: environment

String Thesis of Value Freedom: Science should have no value.

Moderate Thesis of Value Freedom: Science is objective when only epistemic values are constitutive of science practice; moral and political consideration must always remain contextual.

How value can be in science

• in choosing acceptable p value

• in choosing loss function

Definitions of Objectivity

1. freedom from bias
2. intersubjectivity
3. reliability (free from error)

Questions

1. define value a. beliefs has intersubjectivity b. beliefs is a subset of values c. values are not tied to behavior, since one can have one value and behave in a totally opposite direction. d. They are a black box machine since values are infinite.

2. reading: research violent games

3. conclusion: no difference

4. objection:

   • game not representative
   • violent doesn't matter as much as FPS.
   • better to compare gamers from non-gamers
   • reaction doesn't tie to violence in society
   • less reaction can be more violent

   a. objection that is epistemic values - game not representative - better to compare gamers from non-gamers b. objection that is non-epistemic value - violent doesn't matter as much as FPS. - reaction doesn't tie to violence in society - less reaction can be more violent c. objection that is constitutive value - game not representative - better to compare gamers from non-gamers d. objection that is contextual value - violent doesn't matter as much as FPS. - reaction doesn't tie to violence in society - less reaction can be more violent

5. Adding value in gun violence research Yes. Gun issue is not objective since it does not have intersubjectivity. Since the truth may lie in the spectrum of those different believes, the insertion of such bias into science research will lock down the possibility of otherwise.

"Damage done" is a constitutive value.

1. about bias "Objectivity as freedom from bias" identify a research as biased if non-epistemic value can alter the result of a research. In this sense, surveys are not more objective than interviews because in either of the research, researchers can design questions that contains researcher's own bias. "Objectivity as intersubjectivity" allows non-epistemic values as long as they are open to challenges by diverse opinions. In this sense, surveys are not more objective than interviews because both research can accepted by science community. "Objectivity as reliability" concerns with the replicability of research methodology. Survey studies are easier to replicate than interviews because survey tends to use quantitative data which is easier to control whereas a typical interviews tends to involve natural language, gesture, emotional expressions.

Class

constitutive value: generated from goal of an activity (epistemic) contextual value: environment, independent from goal (non-epistemic)