OPr% =(NP/NEW) *100 

where OPr% is the overall percentage pass among learners in that year, NP is the number of learners who passed the matric examinations based on what was defined as the pass mark in that year, NEW is the number of learners that were registered for the matric examinations in that year. Learners who were not registered to write the matric examinations in a specified year should not be included in the denominator in computing the pass rate for that year. Completion or survival rate is a more complex computation because many factors could be responsible for attrition among the original size (in absolute terms) of the cohort that were registered in a particular year. Computing a completion or survival rate for a specific cohort registered in a particular year entails following the cohort through each year from the time they registered up to the time they wrote the examinations. Attrition in the cohort could be due to socioeconomic and demographic factors such as illness, inability to continue paying for education, pregnancy, emigration etc, thus requires a lot more data to compute than computing pass rate. One of the appropriate methods of computation is survival analysis using life table methods. If one has all the data pertaining to “loss to follow” among some of the original cohort, then one may employ multiple decrement life table approach. The focus here is not on pass mark but on computing pass rate and survival/completion rate.

Emeritus Professor Eric O Udjo (PhD, London)
Independent Consultant

Eric Udjo Consulting
www.ericudjoconsulting.co.za
Email: info@ericudjoconsulting.co.za
14 January 2026.

The post Pass Rate Vs Completion Rate in Matric Result appeared first on Eric Udjo Consulting.

Q: What exactly is the point and purpose of a census?

A: A modern census is the enumeration or count of persons as well as obtaining the characteristics of the persons counted in a defined territory. Such information is the cornerstone of planning, decision making and monitoring of policies in all sectors of the economy in a country.

A: The last census in South Africa was in 2011. Section 7 (2a) of the Statistics Act 1999 states that: “The Statistician General must cause a population census to be taking place in the year 2001 and every five years thereafter…”. However, a census was not undertaken in 2006.

Q: How will this one work when it comes to the Popi Act? How can we be sure our private information will be protected?

A: I am not able to answer this question as I cannot speak for Statistics South Africa on this issue. I would suggest that this question is posed to a senior official in the Census division in Statistics South Africa, or even to the Statistician General himself.

Q: What sort of questions are asked and how can officials be sure the answers are correct? Are

they checked or taken at face value?

A: I have not seen the final questions to be administered in the 2022 South Africa census but usually in South Africa, the questions are grouped into the following sections: Demographic (such as date of birth, marital status), Migration (this is strictly demographic), General Health, Parental survival (this is strictly demographic), Education, Employment, Fertility (this is strictly demographic), Housing, Mortality (this is strictly demographic).

The census enumerators are trained to check that the answers they receive to the questions make sense. However, strictly speaking, they cannot be sure that the answers are correct. It requires specialised analysis of the answers to check their accuracy.

Unfortunately, answers to questions in a census are often taken at face value but that should not be the case. Censuses everywhere will always have some errors. The first type of error is coverage. This relates to the fact that some may not be counted in a census and in some extreme situations, some people are overcounted in a census. In either case, the census result will not reflect the true population of the country. There are methods of checking the extent of coverage in a census one of which is the Post Enumeration Survey (PES). The results from a PES are used to adjust the actual census count. But the PES has its flaws and the adjustment carried out through the PES may not reflect the true population size of the country. The PES in South Africa’s 2011 Census was particularly controversial.

In my opinion, the actual census count was over adjusted using the PES. Thus, in my opinion based on the analysis I have done, the population size of South Africa since 2011 is not as large as the official figures would make us believe. The second type of error that is present in every census in varying degree is content. This relates to the accuracy of the characteristics of the persons that were enumerated. Take for example: one of the questions in a census questionnaire is Date of Birth or Age last birthday. Some people intentionally lie about their date of birth or their true age for various reasons. Also, some people may not know their true age or date of birth. Regarding Marital Status, some people may report they are single when in fact they are widowed but single (never married) is not the same as widowed (spouse has passed away). Some people for various reasons may respond to the questions on parental survival that their biological father is dead when in fact, he is alive. In these few examples, the census enumerator cannot be sure that the answers provided are correct.

With improvements in technology, a computer program may be used behind the scenes in the administration of the census questionnaire in a tablet or handheld computer. The program may do logical or consistency checks on the answers provided and to warn about inconsistencies in the answers. Even then, one cannot be sure that the answers provided are correct. To reiterate, it requires specialised analysis to check the accuracy of the answers to a census. It should be stressed that this is not unique to South Africa but a global phenomenon.

Q: Do you have to comply or can you say no thank you?

A: This is a tricky one. I should firstly emphasize that we should all endeavour to participate in a census because it is in our interest to do so. Regarding the question, I have posed this in a different way at Research Ethics Committee Meetings in which I was member at the Human Sciences Research Council as well as the University of South Africa. There are different views on this. The reason I have posed the question is because, although I am not a legal expert, there seems to be a conflict between a section of South Africa Statistics Act and a section of the South Africa Bill of Rights in South Africa’s Constitution. In my reasoning, South Africa’s constitution supersedes any Act in the country.

Sections 16 (1) and (2) of South Africa Statistics Act 1999 state:

16. (1) The Statistician-General, or an officer of Statistics South Africa authorised by him or her, may, in performing his or her functions in terms of this Act, put, to any person any questions which the Statistician-General or that authorised officer considers reasonably necessary for the collection of statistics.

(2) Every person, including every employee of any organ of state, must— (a) to the best of his or her or its knowledge and belief and subject to the right to dignity and privacy, answer, when so required, all questions put orally or in writing …….

It would appear therefore from the Statistics Act that everyone must comply with answering census questions and cannot say no thank you.

However, Section 16 (1b) under Freedom of Expression in Chapter 2 of South Africa Bill of Rights states:

16. (1) Everyone has the right to freedom of expression, which includes— (b) freedom to receive or impart information or ideas.

It appears therefore that the above-mentioned section in South Africa Statistics Act 1999 contradicts the above stated section of South Africa’s constitution. Reasoning that South Africa’s Constitution is supreme to any other Act, one can say no thank you to answering census questions. South Africa’s constitutional court is the ultimate authority to provide guidance on this issue.

Q: How does the census affect us as citizens?

A: Census information enable policy makers to plan better in improving the welfare of a country’s citizens if such information are properly utilised.

Q: Why is it important for us to be part of the census and make sure we register?

A: I am not sure about registering for the census, but it is crucially important that we all get counted in the census because the information from the census will be used for planning and decision making for our benefit.

Emeritus Professor Eric O Udjo (PhD, London)
Independent Consultant

Eric Udjo Consulting
www.ericudjoconsulting.co.za
Email: info@ericudjoconsulting.co.za
27 January 2022.

The post You and the Census appeared first on Eric Udjo Consulting.