Find Us on Facebook
11 Jul

Using the Cloze Test in our Masikhulisane Reading Programme

Added by : Lesley Foster


Part of any curriculum initiative developed by the GM South Africa Foundation includes an evaluation of the impact of the programme. The aim of the Masikhulisane Reading Programme is to increase learners’ oral reading fluency and comprehension. It was decided to use the cloze test procedure to measure the oral reading fluency and comprehension reading levels of the learners at the start of the Masikhulisane Reading Programme and again at the end of the programme. It was hoped that the results of the baseline cloze test would also assist teachers in pairing their learners.

The Cloze Test Procedure

The cloze procedure is a technique in which words are deleted from a passage to create blanks. Learners must then fill in these blanks with appropriate word choices, sometimes provided in a word bank. The Cloze procedure is a useful tool for reading assessment and can be used with both Home Language and First Additional language learners.

The Department of Education recommends the use of Cloze tests in a handbook entitled Teaching Reading in the Early Grades, January 2008. In this handbook, a ‘Cloze’ test is described as a test which can ‘help you assess learner’s reading comprehension’ and ‘ will help you organize learners into groups of similar language ability’.

The process used to develop non standardized ‘Cloze’ tests for Grades 4, 5, 6 and 7:

The advantages of using a ‘Cloze’ test:

Possible problems arising out of having teachers test their learners:

It is important to point out that the development and implementation of the cloze tests were in the nature of an experiment. Before administering the tests to the learners in the 2017 programme the test was trialled with the all the learners from grade 4 to 7 in a school which has participated in the programme since 2015. The results indicated that the test was reasonably reliable and could provide an overview of learners reading competencies.

Before administering the test it was accepted that a number of factors could impact on the scores obtained by learners, for example:

Evidence that some of these factors played a role in the scores obtained by learners was revealed by the scores obtained by certain classes.

Each school was provided with an analysis of the data submitted by the teachers and recorded on the Excel spreadsheet in the following format:

The number of learners in each category is recorded followed by the % range for each number.


Name of school and Grade

and below


and above

No of learners
In class


No of learners and %

No of learners and %

No of learners and %

Grade 4 - Class 1

9 -23.5%

11 – 28.5%

18 – 47%


Grade 4 - Class 2

23 – 57.5%

6 - 15%

11 - 27.5%


Grade 4 - Class 3

11 – 29%

16 – 42%

11 – 29%


Grade 5

10 – 27%

11 – 30%

16 – 43%


Grade 6

6 – 15%

9 – 22.5%

25 – 62.5%


Response from teachers to the administration of the cloze test

  • It did not take a long time to administer or score.
  • The text could be used as part of CAPS language assessment requirements.
  • They felt that they had gained insight into the reading ability of their learners which they could compare with the scores learners had been obtaining for other CAPS language assessment tools.
  • Perhaps, most importantly, the learners seemed to enjoy the exercise.

The value of using the cloze test procedure for the Masikhulisane Reading programme and recommendations for the future

An analysis of the results of the cloze test provided facilitators with an overall picture of the reading ability of learners in the schools participating in the programme and teachers with the information which they could use to pair learners. Despite the possible manipulation of learners’ scores the information gained also provided an entry point for a conversation with the teachers in each school during school visits. In addition, the data provides support for anecdotal feedback on the impact of the programme. It is, however, acknowledged that the test is informal and non-standardized.

The cloze tests developed for Grades 4, 5 and 6 have provided acceptable data which can be used to measure the impact of the programme. However, the results from the Grade 7 close test were, in some cases, so high that there is little room for improvement. It is recommended that a new version of the Grade 7 test be prepared for use in the future.

The administration of a baseline and a summative test measuring the oral reading fluency and comprehension of learners places the focus of the success of the programme on how effectively teachers implement the bilingual, paired reading for enjoyment strategies presented during the programme in their classrooms and gives purpose to our school visits and mentoring.

Leave Your Comment

Your Name:
Your Email Address:
Your Comment:
Security check: If you mix black and white you get...
© 2006-2017 GM South Africa Foundation | Home | Contact us | Terms & Conditions | Sitemap
Website development Online Innovations