How NFER Online Assessments are standardised

Modified on Tue, 20 Aug at 3:31 PM

NFER Online Assessments have been trialled and standardised against a nationally representative sample, so all reporting information is derived from the analysis of data from the nationally representative sample.

Because these are standardised assessments all the learners should take the tests in the same way to answer the same questions which should then be marked in a consistent manner. For the open response questions which are teacher marked, this means marking is in accordance with the guidance in the mark scheme.

To produce our standardised assessments, NFER carried out extensive trialling of all the questions on a considerable number of pupils in standardisation trials.

The standardisation trials provide a reliable and trustworthy source of information for the following reasons:

The sample for each NFER trial is nationally representative. This means pupils are from various regions across the country and are representative of the population for school attainment and FSM.
Each trial is taken with pupils of the appropriate age for the assessment and at the point of the school year where the test is likely to be used. For example, the standardisation trials for the Year 3 summer assessments were taken by Year 3 pupils in June.
Each child trials a test that reflects the final test in method of delivery, number of questions, papers in the test and the time allowed to complete it. However, for each test, we trial double the total number of questions. This means that if a question does not perform as we would like, it can be dropped, and only the best questions are chosen to make up the live assessments. Using the same method, we can ensure that the test assesses across the curriculum; this further contributes to ensuring the final assessments are an accurate measure of learner performance.
Each NFER trial is undertaken with between 1200 and 2000 pupils for each subject and year group. This provides a large amount of data to ensure that any findings are reliable and valid and not a result of chance.
The trials enable NFER to refine the mark schemes to ensure that every question can be marked consistently and in the same way for every learner.

Statistical analysis is carried out on all the data from the standardisation trials. The evidence from the trials is used in the reports so your school can compare how well your learners are doing against a nationally representative sample.

For the demo account, the assessments are for example only and have not been trialled and standardised. All standardisation data provided is mocked up and for illustrative purposes only.

The sample assessments are representative of the content, layout, and question types you can expect to find in our NFER Tests, but to retain the integrity of the standardised papers, they are not from the actual published materials.