The dust raised by 2016 KCSE results having settled, I wish to analyse the data in order to separate the lies from the truth .Yes ,separate the sheep from the goats ,figuratively speaking.This is after it was reported in the news and said on the by-ways and highways that the results showed the expected normal distribution curve pattern.
Now a Normal distribution is a bell shaped curve. The data tended to cluster around the Average score. The further data points are from the mean the less its possible for the data to occur at all.
Many things such as height and intelligence follow a normal distribution curve. For example if we took the height measurement of 300 men randomly and plot their height on x axis and frequency on y axis you will get a normal curve.
In this little project I set out to find out if KCSE result for 2016 describe a normal curve.Then as a matter of course I offer my two pence about the findings.As matter of information ,expect zero spin or politics.
To investigate the problem,I laid my hands on the KCSE results data from online sources for the Year 2016 and 2014. KNEC does not provide free download of aggregate KCSE data to members of the public who incidentaly fund it through their taxes.After some cleaning ,I uploaded the data into my dataframe .The computer progarmming language I used in the analysis is Python.Please click here to view the data frame The Year 2016 KCSE analysis
My findings and recommendations
First the KCSE 2016 results curve is abnormal and looks strange. Its skewed with more than 50% of the students failing to score the average marks. Approximately 68% of the students should have scored between C-minus and C-plus( i.e 388,421).Instead only 138,025 students attained that score.This means about 250,396 students who have the ability to score between C-minus and C-Plus scores failed to do so.This begs the question,why is that the case?
In my personal opinion its because most school do not have the facilities to meet the learning needs of this students.Also its possible the a majority of our Teachers are not equal to the task of teaching in 21 century.Small wonder most high flying schools are reported to have resorted to cheating if recent reports in media is anything to go by.
Lastly this dismal results could be attributed to the perceived culture of cheating among the high and mighty in our political arena.This perception began with the bungled 2007/2008 General election .All and sundry agree that we will never know who won the election,but it was reported that both sides in the contest engaged in one or other nefarious activities.From that time,our teachers and students took that as clue.You get what you want in life not by honesty and hard work;rather by being “clever”.You rig the system in your favor.That why national school with first class KCPE pupils ,are reported to have engaged in cheating on a massive scale in recent history.That beats logic.Why should you cheat when you have students who could and can score A plain through diligence and discipline?.
The grade that most students scored in 2016 is D-( about 150,000 students) compared with D-plus in 2014 with about (76 ,000 students). E -grade had higher number of students in 2016(33,000) than in 2014(about 6,000).The A-s and B-s grades compares relatively well with normal curve distribution pattern.
Surprisingly the 2014 KCSE result are skewed but at least there is close resemblance to a normal distribution.This is a year reported to have experienced a lot of exam leakage and cheating.Does it mean Kenya students requires cheating in order to post a normal distribution curve results?.This an area that require further investigation to get the insights about this absurd phenomena.Hon Matiangi should be applauded for beginning the process of draining the swamp in the education sector.The high water was hiding problems in the,sector.The strange look of the curves,should tell you a lot need to done,before the sector returns to minimum of norms expected from it.
©copyright 2016 newspeakthone.wordpress.com