Make Data Simple.

Latest Posts

Earthquake Events by the Numbers

On the wake of the 7.8-magnitude quake in Nepal, we ran into some numbers on earthquakes numbers by Mw scale distributed over the last few years.

2013 was clearly the most earthquake-hit year in recent times with 58 events while 2008 being the least with 18 events.

earthquakes

The April 25th Himalayan earthquake which has led to more than 3,000 deaths reported so far including after shocks and the destruction of homes, businesses and iconic sites one of which is the UNESCO World Heritage Site of Kathmandu Durbar Square remains the most destructive natural event of recent times.

Tanzania Statistics Act, 2013. Our Gross Misinterpretation.

This all started with a tweet that had this newspaper clipping:

IMG_3135

This quickly made its rounds around the world and understandably since Tanzania will be hosting this year’s Open Government Partnership Africa Forum.

Looking at the comments that came out of this (and having been interviewed by the media many times and my stories always changed for the sexier version), I decided to read this Statistics Bill, 2013 that was passed by the Tanzanian Parliament.

Here is what I think.

This bill is fantastic!

Every single government in the world has a right to constitute a body that is responsible for its data collection dissemination and publication. This bill goes to institutionalize that body (NBS) and to define its role. The Kenya National Bureau of statistics, like all other government agencies also went through this process in the Statistics bill, 2006.

I honestly do not find fault in this bill, in fact, like any other institutional document, I find this to be in order.

  • This act is aimed at the establishing the National Bureau of Statistics and the statistics governing body to provide for the coordination of the national statistical system.
  • The oath of secrecy goes to protect some government information and secrets that generally, for any government should be protected. This is not new and this exists in all bills that i have ready. Every government has this.
  • This bill actually encourages working together of the NBS with other government institutions and private agencies by joint collections and providing that they develop and maintain a comprehensive national databank by using sectoral data banks developed by the various agencies.
  • The journalist in the above article seemed to have an issue with the definition of the official statistics which according to this bill are statistics produced by:
    • the beureau
    • government institutions
    • agencies
      • The very surprisingly interesting thing is the definition of an ‘agency’ – these include research institutions, non – governmental organizations, development partners or any other user or producer of statistics.
      • Although this is subject to the approval of the bureau, this is possibly the best provision that I have seen anywhere.
    • So, contrary to what the journalist published, this bill actually makes provisions for those out of government.
  • This bill also ensures data protection in the disclosure of unidentified information by insisting that all data must be anonymized before publication and all individual forms and returns destroyed within 5 years.
  • In this bill also, are the plans by the government to fund its own data collection through appropriations by parliament. Although they are open to donations, looking at the ordering of their funds and resources to the bureau, this comes in last (if ordering is anything to go by.)
  • All institutions have penalty regulations for those that do offenses or those with gross misconduct. This is no different for government institutions and as the official carrier of national statistics, you do not want your staff to go unpunished for misrepresentation or miscommunication of statistics that might have heavy economic, social or security implications.

All in all, after reading the bill, the one thing that is clear is, do not trust every opinion you read. Create your own.

 

Too Much Data… Endless Possibilities.

Recently, I applied for a loyalty card in one of retail outlets. I was, as part of the procedure, required to fill out a form. The most striking aspect of the process was the actual amount and the level of detail in the information that I had to part with. This included, among other details:

  • Personal details – my date of birth, gender
  • Household and family details – my spouse, size of my household, number of children
  • Location details – residence
  • Contact details – email, phone number
  • Behavioral details – occupation, interests
  • Financial details – who I bank with

A quick look at this information could only lead to one inference: retail outlets in Kenya are, knowingly or not, effective collectors of staggeringly useful data.

Remember, these retail brands have outlets all over the country: In every major town you would expect to find one of, some of and in some cases, all of Tuskys, Naivas, Nakumatt or Uchumi outlets just to name some of the major brand names. This means they collect this data every single day while effortlessly covering the whole country.

Taking all these information about shoppers, adding to this the fact that the loyalty card provides a system to track shopping habits like frequency of shopping, items purchased and amount spent over time all mapped to an individual; and then overlaying this against already existing demographic data would create a rich and massive data set from which no limits exist when it comes to what could be achieved from it.

Association rules and patterns between variables could easily be used to inform targeted marketing strategies, shopping basket data analysis, product clustering, catalog design, stocking, store layout plans and channel expansion among other retail-specific business decisions.

The goal here would be to turn all this information into actionable insight. First and foremost, the right business questions must be asked. From these questions, appropriate analytic techniques and tools are developed to add value to this customer data by turning it into insight.

Whether this is happening within the Kenyan retail space and at a level high enough to influence outcome is debatable but one thing is for sure: Out there, brands like Budget are firmly in the current datafication-of-life wave and have been known to use customer data as the corner stone of their strategy.

The business needs are there, the data is there, the intellectual capacity is there – so why are we not utilizing these again?

The Rape Conundrum

Over the last few days, reports of a politician accused of alleged sexual assault has dominated the airwaves and resurrected the whole rape and sexual offenses debate about who to blame or rather how to rid our society of this heinous and insensate act.

The chart below shows the total number of reported offenses against morality – a category of crimes that includes rape, defilement, incest, sodomy, bestiality, indecent assault, abduction and bigamy according to the Kenya Police Service – in the 5 year period between 2009 and 2013 broken down into specific crimes. Please keep away from the defilement section if you do not wish to be appalled!

offenses breakdown
Evidently, there is a steady rise in the number annually, ignoring the slight fall in 2012.

This, though, is the number of reported crimes and hence would not include the crimes where victims keep quiet out of fear, embarrassment or just ignorance.

A slight peek into this debate always brings out an uncomfortable truth. It is almost conventional wisdom now that to reduce these incidents, ladies should change their mode of dressing, take self-defense classes, carry pepper spray in their handbags, avoid some designated areas, et al: all this in a free society!

This, to me, conveys a certain narrative. That it is your fault you were raped. You should have known how to fight. You should have been stronger. You should have dressed differently from what makes you comfortable. You should have avoided that street.

Why blame the victim, albeit indirectly?

This is not me assuming that rape only occurs against women and it is only perpetuated by men. No. This is a collection of thoughts about this predilection to lay blame on the victim while minimal exertion is directed at the perpetrator.

So How Do We Change This?

Well, a tweet I came across on “rape prevention advice” embodied what I have felt for long..

To some, it is just an insensitive attempt at making a mockery of the vice.

I think not. To me, it is a well thought-out strategy that rightly shifts the power of prevention from the victim, innocent and barely on the wrong, to the perpetrator, who is in control and can decide whether an actual rape takes place or not.

With the current narrative, a victim feels responsible and might make them hide the truth out of fear of blame and more victimization while the perpetrators move from victim to victim. But if it were to be packaged, instead, as an intolerable act of violence where the perpetrators are called out for who they are and what they did and the victims supported in their recovery then we might see a difference.

Let us move away from telling the victim/probable victim that it is/was their fault and they shouldn’t exercise their right to movement, dressing and association in a free society and shift to reminding the perpetrators that it is their fault and their decision to bring harm to another human being. That what they did cannot be condoned.

Time to stop saying, “You were wrong to be raped” to the victim but turn to the perpetrator and say, “You were wrong to rape.”

World Water Day

“No water, no life. No blue, no green.” – Sylvia Earle

On Sunday, March 22nd, 2015, the world marked the World Water Day, a United Nations initiative to celebrate clean water and highlight the plight of the not-so privileged when it comes to access to clean water and sanitation. The theme for 2015 is “water and sustainable development”.

Globally, 748 million people do not have access to an improved source of drinking water and 2.5 billion do not use an improved sanitation facility. Improved sources of water are less prone to contamination making them safer. They include piped water, rain harvested water, borehole water and water from protected wells.

Water and sanitation

In Kenya, 52.5% of the population have access to improved water – though the number significantly differs between rural (44%) and urban areas (71.7%) and between the various counties. 84% of Nairobi’s residents have access to improved water sources while only 20% of those in Narok are as fortunate.

water

Improved sanitation include connections to main sewer, septic tank, cesspit, Ventilated-Improved Pit latrine and covered pit latrine.

The national figure for access to improved sanitation is 61.1%. Nairobi is again the best placed county with Wajir as the lowest placed county, 13 times worse off than the capital and 9 times worse off than the national average.

sanitation

According to the World Water Development Report, demand for water around the world will increase by 55% over the next 15 years. This strain is expected to impact hugely on health and food sustainability with Sub-Saharan Africa most at risk.

Calls for improved security, job creation, end to corruption et al are no doubt benevolent but if we can’t formulate policies to ensure sustenance of basic life as well, then we might be putting the cart before the horse here.