Difference between revisions of "Group03 Proposal"
Wklim.2017 (talk | contribs) |
Wklim.2017 (talk | contribs) |
||
(16 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
− | + | [[Image:Corruption1.jpg|400px|center]] | |
− | [[Image:Corruption1.jpg|400px]] | + | <div style="background:#FFFFFF; border:#FFFFFF; padding-left:20px; text-align:center;"> |
− | < | + | <font size = 6; color="#2B3856"><span style="font-family:Garamond;">Perceiving Evil: The Study of the Corruption Perception Index |
− | </div> | + | </span></font> |
+ | </div> | ||
<!--MAIN HEADER --> | <!--MAIN HEADER --> | ||
− | {|style="background-color:# | + | {|style="background-color:#D1F765;" width="100%" cellspacing="0" cellpadding="0" valign="top" border="0" | |
− | | style="font-family:Garamond; font-size:80%; solid #000000; background:# | + | | style="font-family:Garamond; font-size:80%; solid #000000; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
− | [[ | + | [[Group03 Proposal|<b><font size="3.5"><font color="#2c3e50 ">Proposal</font></font></b>]] |
− | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:# | + | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
− | [[ | + | [[Group03 Poster|<b><font size="3.5"><font color="#2c3e50 ">Poster</font></font></b>]] |
− | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:# | + | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
− | [[ | + | [[Group03 Application|<b><font size="3.5"><font color="#2c3e50 ">Application</font></font></b>]] |
− | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:# | + | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
− | [[ | + | [[Group03 Report| <b><font size="3"><font color="#2c3e50 ">Report</font></font></b>]] |
− | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:# | + | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
− | [[ | + | [[Group03 Conclusion| <b><font size="3"><font color="#2c3e50 ">Conclusion & Comments</font></font></b>]] |
− | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:# | + | | style="font-family:Garamond; font-size:100%; solid #1B338F; background:#D1F765; text-align:center;" width="20%" | |
; | ; | ||
+ | | | ||
+ | |} | ||
+ | |||
+ | =Key Motivations= | ||
+ | First launched in 1995, the <b>Corruption Perceptions Index (CPI)</b> has been widely credited with putting the issue of corruption on the forefront of the international policy agenda. Transparency International (TI), is an international non-governmental organization based in Berlin, Germany which acts to combat global corruption and prevent criminal activities arising from corruption. | ||
− | + | TI publishes the CPI, annually ranking countries "by their perceived levels of corruption, as determined by expert assessments and opinion surveys. The CPI generally defines corruption as "the misuse of public power for private benefit". | |
− | + | ||
+ | The CPI currently ranks 176 countries on a scale from 100 (very clean) to 0 (highly corrupt). Denmark is the least corrupt country in the world, ranking consistently high among international financial transparency, while the most corrupt country in the world is North Korea, remaining on 8 out of 100 since 2012. | ||
+ | |||
+ | In our project, we married the data set from Transparency International on their CPI records (from 2012 to 2016) versus the World Bank data set through the years, which contains economical, agricultural, social, environmental data of the same countries. We will seek to find out if there is indeed any correlations between the perceived corruption level of a country, and its internal conditions. | ||
+ | <br> | ||
+ | <br> | ||
+ | |||
+ | =Objectives (Questions we like to answer)= | ||
+ | |||
+ | It has been a challenge to validate whether CPI is an accurate index to represent corruption. | ||
+ | |||
+ | A study in 2002 found a “strong and significant correlation” between CPI and 2 proxies: black market activity and overabundance of regulation. But it is hard to find any clear indicators of black market activities and regulations. | ||
+ | |||
+ | There were some claims by other studies as well: | ||
+ | * Researchers found a correlation between higher CPI and higher long-term economic growth | ||
+ | * There is an increase of 1.7% in GDP for every unit increase in a country GPI’s score | ||
+ | * There is a “power-law” dependence linking higher CPI score with higher rates of foreign investment in a country | ||
+ | |||
+ | There is also criticism in the usage of CPI’s methodology, some flaws pointed are: | ||
+ | * Corruption is too complex to be captured by a single score. The nature of corruption in rural Kansas will, for instance, be different than in the city administration of New York, yet the Index measures them in the same way | ||
+ | * By measuring perceptions of corruption, as opposed to corruption itself, the Index may simply be reinforcing stereotypes and clichés | ||
+ | * The Index only measures public-sector corruption, leaving out private actors | ||
+ | |||
+ | The <b>objective of our study</b> is to find out if: | ||
+ | |||
+ | * There is a correlation between CPI and economic growth (through GDP) | ||
+ | * There is a correlation between CPI and education | ||
+ | * There is a correlation between CPI and gender equality | ||
+ | * Attempt to debunk any stereotypes and myths we may have for individual countries | ||
+ | <br> | ||
+ | |||
+ | =Data Sources= | ||
+ | |||
+ | The data came from two sources. | ||
+ | |||
+ | The first one came from Transparency International (2015 CPI as an example): | ||
+ | https://www.transparency.org/cpi2015#downloads | ||
+ | |||
+ | The data set contains the following important columns: | ||
+ | * CPI 2012 - 2016 Rank | ||
+ | * Country | ||
+ | * Country Code | ||
+ | * Region | ||
+ | * Corruption Perceptions Index from 2012 to 2016 | ||
+ | |||
+ | The second data set from the World Bank came from: | ||
+ | https://datacatalog.worldbank.org/dataset/world-development-indicators | ||
+ | |||
+ | This data set is a collection of development indicators, compiled from officially-recognized international sources. It presents the most current and accurate global development data available, and includes national, regional and global estimates | ||
+ | However, due to the huge amount of data, we only kept the data for countries which appeared in the CPI data set and only indices from 2006 to 2016. | ||
+ | The fields we are looking at are the World Development Indicators (WDI). | ||
+ | |||
+ | The filtered dataset for the World Bank data was <b>259,750</b> rows across <b>171</b> countries. | ||
+ | <br> | ||
+ | <br> | ||
+ | |||
+ | =Methodology= | ||
+ | |||
+ | The first factor to assess CPI is to understand the methodology of calculating the index. The CPI scores and ranks countries and territories around the world on the perceived level of corruption in the public sector. CPI is an aggregate index, which draws on relevant questions from several different data sources that capture business and expert views. | ||
+ | |||
+ | In 2012, there is an updated methodology in calculating CPI. The following steps are followed to calculate the CPI: | ||
+ | # Select Data Sources: carefully selected from numerous independent institutions | ||
+ | # Standardise Data Sources: It is standardised to a scale of 0-100 where a 0 equals the highest level of perceived corruption and 100 equals the lowest level of perceived corruption. This is done by subtracting the mean of the data set and dividing by the standard deviation and results in z-scores, which are then adjusted to have a mean of approximately 45 and a standard deviation of approximately 20 so that the data set fits the CPI’s 0-100 scale. | ||
+ | # Calculate the average: For a country or territory to be included in the CPI, a minimum of three sources must assess that country. A country’s CPI score is then calculated as the average of all standardised scores available for that country. Scores are rounded to whole numbers. | ||
+ | # Report a measure of uncertainty: The CPI is accompanied by a standard error and confidence interval associated with the score, which capture the variation in scores of the data sources available for that country/territory. | ||
+ | |||
+ | We can also further analyse the individual survey scores to see if there is an impact to the overall index score by country. | ||
+ | <br> | ||
+ | <br> | ||
+ | |||
+ | =Tools and Packages= | ||
+ | |||
+ | R Studio, Tableau (only for preliminary EDA) and associated R libraries will be used: | ||
+ | |||
+ | * shiny | ||
+ | * plyr | ||
+ | * shinydashboard | ||
+ | * tidyr | ||
+ | * ggplot2 | ||
+ | * plotly | ||
+ | <br> | ||
+ | |||
+ | =References to Related DataViz= | ||
+ | |||
+ | The references we are using for this project: | ||
+ | # E.V., T. I. (n.d.). Transparency International. Retrieved from https://www.transparency.org/research/cpi | ||
+ | # World Bank Open Data. (n.d.). Retrieved from https://data.worldbank.org/ | ||
+ | <br> | ||
+ | <br> | ||
Latest revision as of 21:43, 12 August 2018
Perceiving Evil: The Study of the Corruption Perception Index
|
|
|
|
|
|
Contents
Key Motivations
First launched in 1995, the Corruption Perceptions Index (CPI) has been widely credited with putting the issue of corruption on the forefront of the international policy agenda. Transparency International (TI), is an international non-governmental organization based in Berlin, Germany which acts to combat global corruption and prevent criminal activities arising from corruption.
TI publishes the CPI, annually ranking countries "by their perceived levels of corruption, as determined by expert assessments and opinion surveys. The CPI generally defines corruption as "the misuse of public power for private benefit".
The CPI currently ranks 176 countries on a scale from 100 (very clean) to 0 (highly corrupt). Denmark is the least corrupt country in the world, ranking consistently high among international financial transparency, while the most corrupt country in the world is North Korea, remaining on 8 out of 100 since 2012.
In our project, we married the data set from Transparency International on their CPI records (from 2012 to 2016) versus the World Bank data set through the years, which contains economical, agricultural, social, environmental data of the same countries. We will seek to find out if there is indeed any correlations between the perceived corruption level of a country, and its internal conditions.
Objectives (Questions we like to answer)
It has been a challenge to validate whether CPI is an accurate index to represent corruption.
A study in 2002 found a “strong and significant correlation” between CPI and 2 proxies: black market activity and overabundance of regulation. But it is hard to find any clear indicators of black market activities and regulations.
There were some claims by other studies as well:
- Researchers found a correlation between higher CPI and higher long-term economic growth
- There is an increase of 1.7% in GDP for every unit increase in a country GPI’s score
- There is a “power-law” dependence linking higher CPI score with higher rates of foreign investment in a country
There is also criticism in the usage of CPI’s methodology, some flaws pointed are:
- Corruption is too complex to be captured by a single score. The nature of corruption in rural Kansas will, for instance, be different than in the city administration of New York, yet the Index measures them in the same way
- By measuring perceptions of corruption, as opposed to corruption itself, the Index may simply be reinforcing stereotypes and clichés
- The Index only measures public-sector corruption, leaving out private actors
The objective of our study is to find out if:
- There is a correlation between CPI and economic growth (through GDP)
- There is a correlation between CPI and education
- There is a correlation between CPI and gender equality
- Attempt to debunk any stereotypes and myths we may have for individual countries
Data Sources
The data came from two sources.
The first one came from Transparency International (2015 CPI as an example): https://www.transparency.org/cpi2015#downloads
The data set contains the following important columns:
- CPI 2012 - 2016 Rank
- Country
- Country Code
- Region
- Corruption Perceptions Index from 2012 to 2016
The second data set from the World Bank came from: https://datacatalog.worldbank.org/dataset/world-development-indicators
This data set is a collection of development indicators, compiled from officially-recognized international sources. It presents the most current and accurate global development data available, and includes national, regional and global estimates However, due to the huge amount of data, we only kept the data for countries which appeared in the CPI data set and only indices from 2006 to 2016. The fields we are looking at are the World Development Indicators (WDI).
The filtered dataset for the World Bank data was 259,750 rows across 171 countries.
Methodology
The first factor to assess CPI is to understand the methodology of calculating the index. The CPI scores and ranks countries and territories around the world on the perceived level of corruption in the public sector. CPI is an aggregate index, which draws on relevant questions from several different data sources that capture business and expert views.
In 2012, there is an updated methodology in calculating CPI. The following steps are followed to calculate the CPI:
- Select Data Sources: carefully selected from numerous independent institutions
- Standardise Data Sources: It is standardised to a scale of 0-100 where a 0 equals the highest level of perceived corruption and 100 equals the lowest level of perceived corruption. This is done by subtracting the mean of the data set and dividing by the standard deviation and results in z-scores, which are then adjusted to have a mean of approximately 45 and a standard deviation of approximately 20 so that the data set fits the CPI’s 0-100 scale.
- Calculate the average: For a country or territory to be included in the CPI, a minimum of three sources must assess that country. A country’s CPI score is then calculated as the average of all standardised scores available for that country. Scores are rounded to whole numbers.
- Report a measure of uncertainty: The CPI is accompanied by a standard error and confidence interval associated with the score, which capture the variation in scores of the data sources available for that country/territory.
We can also further analyse the individual survey scores to see if there is an impact to the overall index score by country.
Tools and Packages
R Studio, Tableau (only for preliminary EDA) and associated R libraries will be used:
- shiny
- plyr
- shinydashboard
- tidyr
- ggplot2
- plotly
References to Related DataViz
The references we are using for this project:
- E.V., T. I. (n.d.). Transparency International. Retrieved from https://www.transparency.org/research/cpi
- World Bank Open Data. (n.d.). Retrieved from https://data.worldbank.org/
Back to Project Group Page