Data Journalism Fellowship Blog.

There is no greater joy for a mentor than seeing his or her mentees blossom and develop confidence. I was pleasantly surprised when the five Internews data journalism fellows who only a few weeks ago were reluctant to analyze and find stories in data volunteered  do  a post mortem on the coverage of the preliminary report of the second Kenya AIDS Indicator Survey (KAIS 2012), a day after it was released.

They agreed that most of the stories in the media failed to understand what the numbers mean. But the most misleading was a story published in The Standard with the title: Push for male circumcision in Nyanza fails to reduce infections.

The main misrepresentations of data in the story were:

• The unfounded assumption that male circumcision would have a population-wide impact so quickly. Kenya launched a national program of voluntary medical male circumcision (VMMC), in November 2008.  It is not expected to make a significant difference in prevalence (where many factors are involved) until about 2025.
• The writer of the story forgot that Nyanza had a much higher prevalence to start off with (much more than other regions) – so no single intervention is going to bring the prevalence to the same level as other regions, for many reasons. It also did not consider that with more HIV positive people accessing antiretroviral drugs many of them are living longer and this could lead to a higher prevalence of HIV compared to just a few years ago when the lifespan of an infected person was much shorter so the total number of surviving infected people at any one time would have been lower.

Further the story did not report that the KAIS 2012 finding that HIV prevalence is more than five times higher among uncircumcised men: 17 per cent of uncircumcised men are infected with HIV, compared to three percent of circumcised men.

Some of the story ideas journalists can explore based on the report include:

Increased testing:  The proportion of HIV positive people who were aware of their HIV status tripled from 16 per cent in 2007 to 47 per cent in 2012.

Shift in age group: The age group 45 – 54 years had the highest HIV prevalence at 9.1 per cent, compared to age group 25 – 34 years in 2007. Is this generation that had the highest HIV prevalence seven years ago still the highest now – only that they have aged into an older age group? Is this an indicator that HIV will eventually be phased out if the trend continues?

Link between formal education and HIV: Men and women who completed secondary education had a higher rate of HIV infection than those who had not attended primary education.  Why? Have HIV prevention campaigns left out the educated? Is stigma higher among the educated than the less educated?