Ashley Anderson and Andrea Hill's Comments on "Oil, Islam, and Women" (doi:10.7910/DVN/OTF0JB)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Ashley Anderson and Andrea Hill's Comments on "Oil, Islam, and Women"

doi:10.7910/DVN/OTF0JB

Harvard Dataverse

2010-04-29

1

Anderson, Ashley and Hill, Andrea, 2010, "Ashley Anderson and Andrea Hill's Comments on "Oil, Islam, and Women"", https://doi.org/10.7910/DVN/OTF0JB, Harvard Dataverse, V1

Ashley Anderson and Andrea Hill's Comments on "Oil, Islam, and Women"

doi:10.7910/DVN/OTF0JB

Anderson, Ashley and Hill, Andrea (Harvard University)

2010-04-29

Harvard Dataverse

Andrea Hill

2010-04-29

2010-4-29

https://doi.org/10.7910/DVN/OTF0JB

Social Sciences

Michael Ross' paper "Oil, Islam and Women" (The American Political Science Review, Feb 2008: 102,1, pg 107) has significantly impacted both public debate and social science research on the relationship between Islamic culture and gender equality. It challenges the commonly-held notion that women in the Middle East are repressed because of cultural and religious factors, and instead finds that "women in the Middle East are underrepresented in the workplace and government because of oil". In this article, we use the data from Ross to reassess the impact of oil rents on measures of female influence in society. Our procedure is to replicate the main results of Ross and then examine the impact of changing model specifications and sample frame. We find that one does not have to depart much from Ross' original specifications before oil rents cease to matter in statistical terms. When the data is imputed to correct for list-wise deletion of missing observations, oil rents no longer correlates significantly with female labor force participation. Furthermore, when more recent data are used, we also fail to find a strong correlation between oil rents and political inclusion of women. The same tools used above to highlight the flaws in Ross' original model also revealed a new and unique picture of the variables affected female labor and governmental participation. Specifically, rather than oil being correlated with patriarchy, we found working age, income, and cultural variables, such as being a Middle Eastern country or being a predominantly Islamic country, were powerful predictors of female empowerment. From these findings, we develop an alternative theoretical framework for understanding the relationship between oil, income, culture and female labor and governmental participation. We show how female labor participation and female governmental participation actually move in opposite directions as the wealth of a country increases, but that other variables such as the percentage of the population that is of a working age and the location of the country play important roles as well.

1960-2006

International

<a href="http://creativecommons.org/publicdomain/zero/1.0">CC0 1.0</a>

Ross, Michael, "Oil, Islam and Women". The American Political Science Review, Feb 2008: 102,1, pg 107.

10.1017/S0003055408080040

Ross, Michael, "Oil, Islam and Women". The American Political Science Review, Feb 2008: 102,1, pg 107.

Label:

imputeme.R

Text:

Imputation Commands

text/plain; charset=US-ASCII

Label:

new data commands.R

Text:

Commands to run 2003-2006 data

text/plain; charset=US-ASCII

Label:

newwomens.csv

Text:

Imputed data for running revised versions of the original models

text/plain; charset=US-ASCII

Label:

owomen.csv

Text:

Data for the imputed values

text/plain; charset=US-ASCII

Label:

Replication Commands - All.R

Text:

R commands to replicate "Oil, Islam, and Women"

text/plain; charset=US-ASCII

Label:

womdata1.csv

Text:

Original dataset used by Ross (2008)

text/plain; charset=US-ASCII

Label:

Women in Gov data - 2010.csv

Text:

Data needed to run 2003-2006 models

text/plain; charset=US-ASCII