Class Notes (1,100,000)
CA (620,000)
Dal (4,000)
SOSA (200)
Lecture 21

SOSA 2002X Lecture Notes - Lecture 21: The Sociological Imagination, W. M. Keck Observatory, The Control Group


Department
Soc & Social Anthropology
Course Code
SOSA 2002X
Professor
Dr.yokoyoshida
Lecture
21

This preview shows pages 1-2. to view the full 6 pages of the document.
SOSA 2002 NOV.18TH 11/18/2014
Wrapping up “Arguing with Numbers” and
Ethics and Research Proposal
According to Pierre Bourdieu, what is a sociologist’s job?
Not to explain what is right or wrong, but rather come up with explanation for why things come
up the way they do. The guest lecture supported this.
Social Research is a big part of observing these patterns in society.
Today: Research design alternative to survey based research
Why do some employers prefer to interview Matthew but not Samir?
Oreopolus, Phillip and Diane Dechief. 2011.
Although not written by sociologists, it tackles the idea of ethnic prejudice.
There is a strong notion amongst Canadians that we are past racism.
Individuals are not perceiving of themselves as racist so we cannot directly ask that question.
May not be intentionally racist.
This research is trying to examine the structural components; if we aren’t racist why are there
still racist things happening?
Investigation on the discrimination/prejudice in the job market.
What were the variables used?
a. Ethnic names (X1)
b. Foreign Degrees (X2)
c. Foreign Work Experience (X3)
… There are more exogenous variables as well.

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

How do these attributes affect callback rates? (Y- endogenous variables)
Data Collection
-Social Experiment-
They used an audit study. Very different than survey based social research.
In survey research there are different sets of questions designed to explore how they affect each
other.
“Audit Study” – A type of social experiment to measure social practices through a field study
In experiment design (Ch.2 text), there are people from the sample split into different groups.
The attributes being tested are applied to different groups.
 Control group, Treatment X1, Treatment X2, Treatment X3.
The rate of outcome is compared between the control group and the other groups.
 Outcomes Y
Different than a survey because everyone is asked the same question
Experimental Design in the “call back” study:
The control group was that with an English name, Canadian education and Canadian experience.
Treatment 1: foreign name, Canadian education, Canadian experience
Treatment 2: foreign name, foreign education, Canadian experience
Treatment 3: foreign name, foreign education, mixed experience
Treatment 4: foreign name, foreign education, foreign experience
The changes in the callback rates between control group one and treatment one allows them to
identify the impact of the name because education and work experience were the same, and so on
(differences between number 2 & 3 allows us to examine the importance of education location)
Because there are no really individual behind these resumes, we can see how they are evaluating
them without much legal or ethical concerns, unlike in surveys.
You're Reading a Preview

Unlock to view full version