In the standardisation, classes were randomly allocated to sit the reading and maths tests on either paper or computer. The content of the tests was the same; only the mode of delivery was different i.e. paper or computer. After the standardisation, we looked at average performance on both modes, across all class levels and test forms. On average, pupils sometimes performed slightly better on the paper version of the test, and sometimes slightly better on the computer-based version. In general, these differences were small. The exception to this was Level 2 (usually Second class), where the computer-based tests were more difficult, on average, than the paper tests. For this reason, we decided not to release the Level 2 tests on computer for now.
For Levels 3 to 6, the paper and computer-based tests were normed separately. This means, for example, that if a pupil takes Form 3A of the New Drumcondra Primary Reading Test (DPRT) on computer, their standard score is based on a comparison with the pupils in the standardisation sample who also took Form 3A on computer. This makes it a like-with-like comparison.
Yes, standard scores, STen scores and percentile ranks can be compared across test modes. Raw scores (number of correct answers) should not be compared directly. The reasons for this can be explained by using an example.
Example: Lena and Matthew
Lena and Matthew are friends who go to different schools. At the end of the last school year, Lena’s class sat the New DPRT on paper and Matthew’s class took it on computer. They both sat the same form of the test – Level 4, Form A.
Both Lena and Matthew achieved a standard score of 110. This corresponds to a STen score of 7 and percentile rank of 75. Lena’s standard score of 110 is based on comparing her performance on the test to those pupils in the standardisation sample who also took Form 4A of the New DPRT on paper. Matthew’s score of 110 is based on a comparison with the pupils in the standardisation who took Form 4A on computer. The percentile rank of 75 means that both Lena and Matthew performed as well as, or better than, 75% of pupils who took the same form of the test as them on the same mode.
To achieve a standard score of 110, Lena answered 61 question correctly out of 80 on the paper test. However, to achieve the same standard score, Matthew answered 60 questions correctly on the computer-based test. This shows why raw scores should not be compared directly. The same raw score can lead to slightly different standard scores and vice versa. This is due to small differences in difficulty (as explained above) and also to differences in how scores were distributed among the samples of pupils in the standardisation.
The New Drumcondra Primary Tests of English reading and mathematics were normed in spring/summer 2018 only, and, as a result, there are no autumn norms. This follows on from Circular 0056/2011, which implemented changes to standardised testing requirements as part of the Literacy and Numeracy Strategy. This circular states that standardised tests should be completed by all pupils in 2nd, 4th and 6th class during May/June (with the usual allowances for exemptions).
Circular 0056/2011 can be accessed here: https://www.education.ie/en/Circulars-and-Forms/Active-Circulars/cl0056_2011.pdf
Pupils in 2nd, 4th and 6th class typically sit standardised tests in May/June, with aggregated results subsequently reported to the Department of Education and Skills (DES). However, schools may wish also to administer the tests near the start of the school year (to these and/or other class levels). Where this occurs, the class level in question should generally sit the test one level below e.g. pupils at the start of 3rd class should take the Level 2 test. For mathematics in particular, this will ensure that pupils have covered all of the content of the assessment.
The previous versions of the Drumcondra Primary Tests (the DPMT-R and DPRT-R) were standardised in 2005 and 2006, when both spring and autumn norms were generated. The standardisations of these tests indicate that the differences between spring and autumn are reasonably small. Pupils taking the tests in autumn tended to perform slightly better than those taking the same level and form in the spring. This was to be expected, given the differences in age and curriculum coverage.
In practice, this meant that when spring and autumn norms were available, the same raw score (total number of questions correct) resulted in different standard scores, STen scores and percentile ranks depending on when the test was taken. A child who took the test in autumn, with autumn norms applied, received a slightly lower standard score than they would have with spring norms applied, because more of their peers in the autumn standardisation scored within the higher range on the test. Results on the overall scales between time points are usually within a few standard score points (10 or fewer), and within one to two STen scores of each other. For autumn administration of the new (2018) tests, it should be noted on the report of the results for each class that the tests were taken in autumn rather than spring, and that pupils took a test one Level below their grade. Results should be interpreted with this in mind.
If pupils take the same Level of the test within the same academic year (i.e. autumn followed by spring), or the same calendar year (i.e. spring followed by autumn), it is advisable for pupils to receive a different form of the test on each occasion to reduce practice effects.
A general principle is that some three to four weeks should elapse between when children return to school and when a standardised test is administered in autumn. This is intended to allow pupils to get used to school again, and to overcome any possible summer learning loss.
Special note regarding autumn testing in 2020
Even if the New Drumcondra Primary Tests had been normed in autumn as well as in spring, those norms would have been generated following a typical school year – i.e., instruction within schools until the end of June. There is no precedent for the school closures that were implemented in March 2020, and hence, it is not possible to predict the impact of such closures on performance. Schools should bear this in mind if interpreting performance in autumn 2020 based on spring test norms.
Some schools may find that their needs are better met in this instance if they interpret the tests with respect to the particular skills (represented by individual questions, or groups of questions) that children succeed on, and those they find challenging, and use this as a basis for planning instruction.
Occasionally, a school may prefer to administer the same level of a test at the start and end of a school year to track progress over that year – e.g. 3rd class would sit Level 3 in autumn, and again in May/June. If this is done, it should be borne in mind that the test assesses content that may not yet have been covered at the start of the year.
In line with the 1999 Primary School Mathematics Curriculum (PSMC) pupils can use calculators for some questions in the New DPMT for Levels 4, 5 and 6 (usually 4th to 6th class).
For these Levels, pupils complete Part A of the test without a calculator. This part of the test includes any questions that directly assess pupils’ ability to carry out operations and procedures relevant to their class level. There are also a number of questions in this part of the test that are ‘calculator neutral’ e.g. reasoning questions without computation. Pupils can use calculators to complete Part B of the test. In this part, the focus is not on the ability to carry out computations by hand, but rather on integrating, reasoning and problem-solving skills.
While pupils are allowed to use calculators for Part B, it is not a requirement. That is, if pupils at certain class levels in your school are not accustomed to using calculators, they can still complete the test without them. A small number of schools took this approach in the standardisation. However, please note that the test must still be completed within the time limits outlined in the administration manual.
Yes, the New DPMT is available in Irish for all class levels. Following standardisation, the tests were professionally translated and were made available for sale to Irish-medium schools at the same time as the English language versions. The administration manual is also available in Irish.
A review of the translations is currently underway to ensure the language used is as child-friendly as possible. Resulting changes, if any, will not be in place for testing in 2020 but will be implemented for testing in 2021.
It depends on the level of the test that is being administered.
Level 1 of the NEW DPMT is read aloud in full and there is minimal text displayed on the pages of the booklets. There are two parallel forms available. Both have identical instructions for each question, but there are small differences in the numbers and images used, so the correct answers are usually different. This means that both forms can be administered to the same class group simultaneously. The booklets can be distributed in such a way that pupils beside each other have different forms.
Part A of Level 2 uses the same 'parallel form' approach as Level 1 and is read aloud. In Part B, pupils work through the test independently. Any context-free computation questions are in this part of the test. To further reduce reading load in Part B, images are used to help pupils interpret questions and identify key information. In addition, if a pupil has difficulty reading a word in Part B, they may ask their teacher to read it for them. This is noted in the manual and forms part of the script for Part B, so pupils are notified before they start. Please note that, while any word can be read, mathematical terms must not be defined. This includes, but is not limited to, names of 2-D and 3-D shapes, as well as terms such as 'edge, 'face' and 'even number'. Knowledge and understanding of mathematical vocabulary is part of what the New DPMT assesses.
Levels 3 to 6
At Levels 3 to 6, pupils work through the full test independently. However, as for Part B of Level 2, teachers may read any words for pupils in any part of the test. Again, pupils are notified of this as part of the administration script. The restriction on defining mathematical terms also applies here. Examples of relevant terms that feature at these levels include 'parallel', 'area', 'average' and 'prime number'.
The test administration manuals indicate that you may exempt pupils who have a physical or intellectual/learning disability that would prevent them from engaging with the test in a meaningful way, or who have insufficient experience (generally, less than one year) of instruction through English or Irish.
In general, however, we advise that pupils be included whenever possible. You may decide that it is possible to include a pupil in testing only if special accommodations are made. This can be done provided that:
- You note the accommodation next to the pupil’s result;
- You interpret results cautiously, as the test has been taken outside of standardisation conditions;
- If the pupil is in 2nd, 4th or 6th class, you mark them as exempted on returns to Esinet (i.e., you do not include their result in this report).
Accommodations could include, but are not limited to:
- a child taking the test at a lower level than their class level;
- a reader for the mathematics test;
- taking the test in a quiet room;
- answering a paper test in the Level 3 – 6 range using the test booklet rather than an answer sheet;
- extra time for comfort breaks;
- taking the test with the support of a Special Needs Assistant.
Additional guidance about exemptions and special accommodations is available from the DES here.
This document states that the decision to exempt a particular pupil lies with the school principal, taking into account the pupil’s needs, abilities and best interests.