Education in Kenya: The Problem with KCPE and KCSE
Every time Kenyan students do their KCPE or KCSE exams, there is always a national conversation centered on the nature and importance of these exams. In this issue, I will share my views about what ails both exams and what can be done about it.
Before I begin, here’s a list of questions underlying the objectives of this article. I will try to answer each question as briefly and succinctly as possible.
1. What’s the purpose of schools?
2. What’s the nature of students?
3. What’s the relationship between education and intelligence?
4. Do exams measure intelligence?
5. What’s wrong with KCPE and KCSE?
6. What can we do about it?
What’s the purpose of schools?
The reason we get education policy awfully wrong is because many people do not know what is the purpose of schools. Why do we have schools and what are they supposed to accomplish?
The basic purpose of schools is to teach numeracy, writing, and reading. This is the reason why schools exist. These skills when put together, prepare children not only to meet their obligations in life but also to educate themselves further. Everything else outside reading, writing, and counting is beyond the mandate of a school.
Of course, reading, writing, and counting have their complexities and a school should address them all. Algebra, for example, is part of the complexities of counting and so is trigonometry. However, some of the things we think should be done in schools are beyond the mandate of a school. For example, a school teaches you to read but it does not help you decide what to read. By learning to read, a school prepares you with the necessary skills to educate yourself on things beyond the mandate of the school. You cannot blame a school for the inadequacies of students. After it teaches them to read, its on them to educate themselves.
Other things beyond the mandate of a school include teaching children how to build robots, learn to swim, learn a foreign language etc. If a school has the time and resources, they can do it, but they do not have to. The same is true of extracurricular activities. These are programs that improve the experience of students in schools, but they are optional.
A good education policy is that which helps schools meet their core mandates first, before venturing on other luxuries. Good policy may include actions aimed at reducing class sizes, increasing access to stationary and books for both teachers and students, increasing teacher-student engagement, and giving children enough food to make sure they can bear with the classes. There are other good policies, but you can readily see how these actions would help a student learn the math even without a swimming pool in the school compound.
Suffice to note, I did not ask what’s the purpose of education. Education is a broad word that encompasses many things, most of which are not taught and learned in school. Education even encompasses mannerisms and behaviors that are learned in different spheres of life. Education has been a necessity throughout history, but schools are a more recent phenomena. Also worth noting is I do not think of universities as schools. These are complex institutions that instill skills, knowledge, and experiences beyond the mandate of elementary schools and high schools.
What’s the nature of students
Historically, children were thought to be small adults. As time passed we came to realize children are different from adults physically, cognitively, and emotionally. One myth that continues to persist among adults, however, is that children have empty brains and it’s the job of schools to fill them up.
That myth is called the blank slate and it was originated by among others John Locke and Jean Jacques Rousseau. This myth is so perverse its coded in every day language.
One of my school’s guiding words back in primary school was a Bible verse from Proverbs 22: 6 that said “train up a child in the way he should go, and when he is old he will not depart from it.” The verse implies children can be molded to meet societal expectations.
You’ve also probably heard of the saying from Proverbs 13:24 that says “spare the rod and spoil the child.” The verse posits there’s a way to shape children and one of them is to spank them when they misbehave.
In Kiswahili, we also have the “samaki mkunje angali mbichi,” proverb, which means shape something when it’s still fresh.
The common theme is that children are pieces of clay and we can shape them the way we want. That’s wrong.
Children are not blank slates and they come to school already packed. children come to school with personalities. A child’s personality could ultimately determine whether they learn their math or not. A timid child will definitely get bullied. A child who is more aggressive than his peers will also give them a rough time. If children are not motivated enough, they could also fail in school. I have written in this article how a personality called conscientiousness affects students’ academic performance.
Being short or tall will also be an advantage or disadvantage in school and so is being attractive or unattractive.
Children also come to school with complex brain mechanisms which may affect the learning process. For example, a child may come to school with autism, ADHD, OCD, and other mental disorders. These conditions will ultimately affect how these children engage with others and how they learn. Attempts to shape a child with any of these conditions will face difficulties.
However, the most important thing that children come to school with is intelligence. Children with a high IQ will tend to learn and understand concepts fast, while others will take time to learn similar concepts. Some children will also come to school with learning disabilities so severe they’d rather attend special schools.
All those factors from personality, mental health, and intelligence will ultimately affect the students’ academic outcomes. These differences among children are not acquired in school. A significant degree of these issues are acquired from parents, a result of shared DNA.
Even before children step in schools, the die is cast. The smart ones will learn fast and easily, average students will take a slower approach, while those below average will struggle. This is nobody’s fault, but it is a fact that all schools have to contend with.
In essence: All children are not equal, so don’t expect education outcomes to be equal. Also worth noting is, schools don’t make a difference in the educational outcomes of a child. Schools teach but they don’t determine how a student performs. The performance of a student depends on the student and nothing else. For example, students in Alliance High School don’t pass exams because they are in Alliance. The school is made great by the quality of students it admits and not its infrastructure.
Relationship between IQ and Education
One way in which students in a classroom will differ is in their intelligence. Some students will be smart while others will be average and below average. Many people have the false idea intelligence is not important in predicting academic performance, but as you shall see, education and intelligence have a tight knit relationship that cannot be ignored.
I like to think of intelligence as a tether. A tethered goat will eat grass within the distance mediated by the length of the rope. The length of the rope is the radius, and the goat will eat grass only within the circumference of the circle. Now imagine, the length of the rope is 7 meters, it goes without saying by the end of the day the goat will have eaten grass covering an area of 154m2.
A person’s intelligence is a tether that determines how much they can learn. Students with longer tethers will tend to learn a lot while those with shorter tethers won’t learn much.
My analogy is simplistic in several ways. First, it doesn’t fully capture the speed in which student’s learn. Second, it doesn’t capture their store of relevant knowledge in both short term and long term memories. Third, my analogy does not capture reasoning abilities and problem solving skills. All these are dimensions of intelligence, that ultimately determine how much a student can learn and how fast they can do it. Ether way, students with longer tethers will have all these dimensions of intelligence at their disposal.
Considering schools are supposed to teach students some essential skills, it all depends on the specific length of each students tether. I am using tether as a metaphor for IQ. Research on intelligence has found the correlation between IQ and Education to be 0.8. One of the highest correlations ever seen in nature. The purpose of exams is to identify each student and the length of their tether. Students with longer tethers will be given much and a lot will be expected from them.
These differences in IQ ultimately determine who can learn calculus and who cannot, who can read deep philosophical works and who cannot, who can recite a poem from the top of his head and who cannot. These individual differences in IQ determine academic performance and schools have to contend with them. Teachers will do all they can to teach students these skills but if they can’t grasp it, there’s nothing else teachers can do.
Do exams measure intelligence?
Even though intelligence correlates with education there is some contention on whether exams measure intelligence. Typically, intelligence is measured using IQ tests or standardized exams. A good example of a standardized test is the SAT or ACT. Both the SAT and the ACT are heavily g-loaded to measure intelligence.
However, national exams like KCPE and KCSE are not g-loaded in any way. The result someone gets in these exams may not be fully indicative of their IQ.(g means the general factor of intelligence, the most established theory of human cognition. A test is g-loaded if it taps on this general factor).
In research, we have concepts like validity and reliability which are used to determine whether a test serves a certain purpose as expected.
Reliability indicates the consistency of a measure. For example, we know that a thermometer is working well if it gives consistent results. If a thermometer was to give me three different results within a span of a moment, I’d know it’s faulty. The same is true of an exam. An exam needs to give consistent results every time it is used. If a student gets an A today then gets a D tomorrow we know the exams are garbage.
Validity is just as important as reliability but more nuanced. Validity tells us how accurate a test is in measuring something. A thermometer that incorrectly reads my temperature is a bad thermometer. Imagine it’s 2020 and the temperature scan misreports your temperature thrice. There’s no doubt you’ll end up in quarantine. The same is true about exams. Just because a student consistently gets Ds does not mean the exam is an accurate measure of his intelligence. Another student could get all As consistently, but still doesn’t mean the student is smart as the exams insinuate. It’s possible to get all As in exams that are too easy.
For an exam to measure the intelligence of a student, it has to demonstrate validity. That is, the results of the exam and the results of IQ tests should not differ by much. The same exam should also demonstrate consistent results to indicate reliability. So far, I do not think KCPE and KCSE are valid and reliable exams when it comes to measuring cognitive performance.
What’s wrong with KCPE and KCSE?
As Kenyans, we all have our experiences with these exams. I remember I was not in school for two terms in my form four, but I still went ahead and did an exam that was supposed to determine the next course of my life. So yes, I believe these exams are faulty certainly because they are one time exams that do not take into consideration the special circumstances a student faces. However, I won’t be dealing with those kinds of reasons here.
As I’ve highlighted in the previous section, KCPE and KCSE have validity and reliability concerns. These are the questions that stand out: What do these exams attempt to measure and do they do a good job at it?
If these exams measure academic performance, the next question should be how accurately they do it. These are scientific questions. You do not just sit in an office in the Ministry of Education or the Kenya National Examinations Council (KNEC) and attempt to answer them. One has to design a study, do the research, and publish the results.
The next question is also difficult. How does one know that these exams measure academic performance and not something else. This is also a scientific question testing whether KCPE measures academic performance and not something else being confused with academic performance. Many critics argue that KCPE and KCSE measure a student’s ability to cram. In this case, is cramming the same as academic performance? No!
Another problem with KCSE is whether it accurately defines the variability between students. Remember I started by saying all students are not equal and that some are smart while others are not. A test will be unimportant if it does not tell us how the students’ performance varies.
Let’s take height as an example. Once in a while you meet some very tall guys. These guys are so tall you could go for days without meeting someone like them. Then we have people with above average tallness. These people are more common but they are few. You are likely to meet one in a day. Then we have average people. These guys are neither tall nor short and happen to be quite common. There is a high likelihood you are part of this group. But what about dwarfs? When was the last time you saw a dwarf? They are rare just like the super tall individuals. We also have short people who are more common than their dwarf counterparts. You certainly meet them once a day.
The same is true of academic performance. The purpose of a good test is to tell us who is super smart, who is above average, who is average, and finally who is below average. Also remember that these measures need to be accurate and consistent.
An exam that does not show us the variation between students will suffer floor and ceiling effects. Ceiling effects happen when all students pass an exam. What such an exam implies is that all students are clever. A good example is KCSE 2015, when many students passed the exam. All student’s are not equally clever and we do not expect them to score the same in a test.
Floor effects happen when students fail so much. It happens a lot when examiners deliberately make an exam difficult. It does not make sense for all students to fail an exam equally. KCSE 2016 is an example of floor effects. A well designed exam should follow the normal distribution, identifying all students that fall within each level of cognitive performance.
If you look at the exam results below you will notice some anomalies. Grade C+ and C are the everage grades, but two thirds of the students in 2016 scored below it. That phenomenon beats the very definition of average. You will also notice students in 2016 failed terribly compared to students in 2015. One reason for this is because the Cabinet Secretary for education Prof. Kaimenyi was replaced with Dr. Matiangi? One was tough and the other was lenient. These results tell us more about the Cabinet Secretaries of education than they do the academic performance of students. Exam results should not only be consistent within groups but also consistent between groups. KCSE 2015 should not give us different results when compared to KCSE 2016. This does not mean either of these exams was better than the other, it means KCSE as a national exam is flawed.
There’s a lot to say about these figures, but I’ll leave that for a different article.
What can we do about it?
There are several ways to deal with the problems that KCPE and KCSE present. The first is to make sure KCSE is not the only exam that determines a student’s trajectory in life. The trick is to make sure all exams that students do between form one and four four contribute to their final grade. Kenyan universities do this and it should be easy for high schools to adopt it.
The Grade Point Average (GPA) is a good measure of that nature. Schools need to record all exams that students do then calculate the GPA. This can be done within schools and there is no need for state supervision. The only thing teachers need to know is overinflated GPAs are not acceptable. It’s very easy to detect an overinflated GPA.
For example, if a student comes with an overinflated GPA for admissions to a university, the admissions officer would need to compare the student’s performance with that of his classmates and the subjects the students took. If all the students in the class have high GPA scores, there’s a high likelihood teachers were too lenient when grading or cheating was involved. The student should be rejected. (Floor and ceiling effects are signs of cheating, lack of rigor, leniency, and defective exams). One also needs to check the consistency of the student’s performance to determine whether the student was serious.
For this to happen, teachers will have to be trained on how to administer exams, grade, and record them. Teachers also need to be made aware of the repercussions that come with leniency, allowing cheating, over-inflating scores, or making exams deliberately difficult. The work of KNEC should be auditing each individual schools’ results and advising on best practices.
Using GPAs completely removes the needs for national exams. However, this system needs honesty and dedication towards monitoring performance on a school level. A school that understands the lives of its students depend on how honest their results are will go a long way in helping the students secure a bright future.
GPA, however, is also problematic. Just like KCPE and KCSE, exams done on a school level won’t tell us much about the cognitive performance of students. Problems of validity and reliability continue to come up. A student’s GPA score still does not equal their IQ, and neither does it accurately reflect Academic performance. The other solution is standardized testing.
Standardized tests are designed specifically to measure cognitive performance. The validity of these tests are high and so are their reliability. As I pointed out, the SAT administered in the US is an example of a standardized test. There’s also the ASVAB used in recruiting soldiers for the military. Countries worldwide use different types of standardized tests, all of them designed with the specific intent of measuring cognitive performance.
Standardized testing can be used together with GPA. The GPA tells you a student’s academic performance in the past while a standardized test tells you how a student will perform in future. Both are important in deciding which students should be admitted to a high level university and which ones should go to second and third tire universities. Both of them also help in deciding which students should go to which courses.
Just like the GPA, standardized tests do not need the supervision of the government. A standardized test is convenient and can be done at any time. That means no student will do it when they are sick or when they do not feel comfortable. In the US, the SAT is administered by College Board at the student’s convenience.
In cases where students have overinflated GPAs, a standardized test can be requested as an alternative. A student with a high GPA who performs poorly on a standardized test is suspect. The same is true of a student with a low GPA who performs well on a standardized test. Could it be the student was in a school where exams were made deliberately difficult?
Both systems are good and should be used together. The GPA works well for the student with a clean academic record. The standardized test works well for the clever student who got poor grades due to some unavoidable circumstances.The excerpt below comes from a memoir by Jose Antonio Vergas called “Dear America.” You can see how GPAs and Standardized tests are leveraged by universities to admit students. You also notice both systems leave students with enough choices.