Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


3010 - Class 16

No description

Kelly Blidook

on 23 November 2017

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of 3010 - Class 16

Basis: Physics, Chemistry, Biology

Transition to: Psychology, Social/behavioural sciences/ Economics
photo credit Nasa / Goddard Space Flight Center / Reto Stöckli
3.) Internal vs. External Validity

First: "construct validity" – variables must actually measure theoretical concepts.

There can be no external validity w/o internal validity (reality inside => reality outside).

Does experiment simulate real world conditions?

External validity ultimately comes with replication.
Potential problems/trade-offs
2.) Mundane vs. Experimental realism

Criticism: Lack of “reality” in experiment

Mundane (actual) realism – may not elicit anything worthy of study.

Experimental (created) realism – may lack external validity

Tradeoff example: Iyengar (2000)
Potential problems/trade-offs
1.) Impact vs. Control

Want participant to be engaged, but want to avoid “baggage” (bias).

Economists do so by using money, but political behaviour is not always this simple.
Potential problems/trade-offs

Difficult to convince and generalize – lack of “reality”

Shared understanding of method doesn’t exist as in Psych and Econ
Why are experiments rare in P.S.?

Risk-taking (use of limited information)
Resource allocation
Individual vs. Collective behaviour

Experiments and common uses
Experiments and common uses
Experiments provide certainty of:

1.) specific treatment

2.) measured amount of treatment (i.e. information

3.) lack of “second-hand” treatment (treatment by
non-measured means).

4.) time element
Effective means of controlling variables
2 Groups:

1. “Control” Group (normal, or non-treated group)

2. “Experimental” Group (receives treatment)

Pretest - (non) Treatment - Posttest

Results: Does Experimental Group result differ from Control Group, etc.?
Basic elements
Effective means of controlling variables

Effective means of assessing causality
The Experimental Method
Posc 3010 – Empirical Methods

1.) Internal validity: causation, not just correlation

2.) Effective control of variables – isolate causal IV, also conditional variables.

3.) Precision in measurement
Experiment requires measured manipulation of IV

In field, lack of control over variables

If “field” provides internally valid component, can be more convincing externally.

"Natural" experiments
Field vs laboratory experiment
Not necessarily “unethical” in itself – depends on risks to participant.

Problems arise with subjects becoming aware.

1.) Potential for researcher bias in design

2.) Experimenter demands or perceived demands – i.e. Hawthorne effect, Pygmalion effect

3.) Validity - do results tell us about the real world?

4.) Range of research topics limited
Concepts of value to Political Science:

Double-blind Experiment

Pygmalion effect

Randomization vs Matching
May not know relevant variables
Depends on size of sample
Considerations affecting outcomes
Full transcript