William Bird, Director of the Media Monitoring Project

Being asked to be the data analyst for GMMP 2005 is a bit like being asked to fly the latest Airbus, it is both intimidating and a great honour, but a lot more exciting.  The GMMP is a spectacular achievement.  On one level it raises international concerns about gender issues and the media, on another, it highlights civil society’s interest and growing activism around gender issues in the media issues.  Delightfully for the Media Monitoring Project (MMP), the GMMP is indicative of a growing trend in the use of media monitoring as a tool to bring about and track changes in the media.

• Monitoring the media across the world for one day provides a global view of where, when and how gender is represented.  The GMMP, as the world’s largest media monitoring project, has the capacity to present some of the most fascinating data, but also some of the most stretching challenges.
• The only element more impressive than the idea of the GMMP is the exercise itself.  Roughly 100 countries, around 57% of the world participated in the GMMP 2005, an increase of 30 countries since the last GMMP.  Aside from the sheer scale of the exercise, the GMMP highlights the power of civil society.  Participation in the GMMP is voluntary and usually involves several organisations in each country.  From a purely logistical point of view the coordination and production of the material had to be intricately planned and carried out.  That the project is now entering the data capture and production of results phase is evidence of how well the GMMP has been organised and coordinated, and is a credit to WACC, Margaret Gallagher, and the planning committee.

What makes GMMP 2005 even more special?

• For the first time, the data for the GMMP is being analysed by a non-governmental organisation in the southern hemisphere.  The MMP is a relatively small human rights organisation, based in South Africa, that has been monitoring the media for more than 12 years.  During this time, the MMP has conducted over 75 media monitoring projects, including numerous gender monitoring projects, and
• the largest Southern African media monitoring project ever undertaken, the Gender Media Baseline Study (GMBS).  For the GMBS in 2003, the MMP designed and produced the methodology, as well as the results for 13 countries.  The GMBS produced over 25 000 news items after a month-long monitoring period.  The GMMP is a global study, even more extensive than the GMBS, and presents a range of different monitoring challenges.

The idea itself sounds simple enough: bring together the monitoring data gathered from 100 countries into one database and then produce results for the world.  To enable the MMP to undertake the exercise, the MMP’s database specialist created a database specifically for the GMMP.  The next step in the process of data analysis is the issue of how to collate all of the countries’ data.  If we were to simply add the data together, because of the disparities in the number and type of media per country, it would mean that the GMMP findings would be almost entirely determined by countries like the USA and China, which have thousands of media.  In undertaking the analysis for the GMMP, our assumption is that the results need to represent the manner in which gender is represented in the media around the world on the 16th of February (referred to in some parts of the world as G-Day).  In order to ensure that this is fulfilled, it was necessary to develop a weighting system that ensured that the countries’ monitoring, even those that have as few as three media, make a valuable difference to the overall outcome of the GMMP, along with larger countries like India.

To make matters even more interesting, because of the differences between the countries, as well as the differences within countries between radio, television and print media, each media type was allocated a different weighting.  In the production of the overall results, each of the three weightings per country will be brought together.  To ensure its validity and accuracy, the MMP has developed the weighting system together with the head of the Actuarial Science and Statistics Department at the University of the Witwatersrand.  The weighting system is based on a combination of factors including the reach of the media in each country, the population, circulation and media band of each country.  When the weightings are applied, they will ensure that all of the countries that participated in the GMMP will make a meaningful contribution to the results.

One of the developments in the GMMP 2005 is that in addition to being able to produce global results, the GMMP 2005, will offer comparisons with previous GMMPs, and provide regional comparisons.  Where possible, national data that enables individual countries to measure themselves against their region and the rest of the world will also be produced.  These specific comparisons add further complexities to the production of the results.  Each of the coding sheets contains approximately 20 pieces of information that need to be captured.  In order to produce national and regional results, each piece of information captured needs to be linked to the relevant country code.  In addition, each piece of information also needs to be weighted according to the type of media to determine each country’s weighting.  Only after all of this has been taken into account can the data be compiled to produce the global results.  To produce one of the most basic results, such as the gender breakdown of the total number of sources, requires a six-stage process.

• The data need will be captured into a specially designed database, which will enable the data to be searched and interrogated for various information.  A query or question is then used to extract the relevant information, such as the sex of the sources, from each type of media, and then linked to the country code.  The sources per media type then need to be added up and weighted according to the weighting of each media type for each country.  The weighted data can then be brought together, totalled, and normalised (put into a percentage).  This particular query will provide a result such as that produced during GMMP 2000, which showed that women comprise 18% of all known news sources, while men comprise 82%.  To produce the regional and national comparisons another stage is added.  
• In order to produce some of the more complicated results, for example, a breakdown of the function of the person in the story requires a further four stages, in addition to those already outlined.  The list of database questions or queries has not yet been completed, current calculations suggest that an estimated 5000 database queries will need to be run in order to produce the final results.

In GMMP 2000, over 50 000 people in the news, and over 16 000 news items were coded.  In 2005, with even more countries, it is fair to assume that more than 70 000 people and over 22 000 news items will be coded.  The MMP is committed to ensuring that the high standards achieved in the past will be met and with the innovative national and global comparisons, surpassed.  100 countries have already shown great commitment in monitoring the media on G-Day.  The next exciting phase of GMMP 2005 has only just begun, and all indications are that it will make a critical contribution to the struggle for gender equality.  

• The Media Monitoring Project is the data analyst and a member of the GMMP 2005 planning committee.

---

Add this page to your favorite Social Networking websites