Reflections from Northland Maths Assoc day
Over the weekend I attended the Northland Maths Association day where a National Assessment Moderator from NZQA was a guest speaker. This post includes my understanding of some of the conversations on the day and may not necessarily be the views of the moderator or NZQA. If you have queries please check.
The role of NZQA and the external moderation process is outlined in the NCEA annual report published in May each year. See page 88 at this link
"Each year NZQA moderates approximately 100,000 individual pieces of student work that are submitted by schools as part of their external moderation process. The standards moderated are from the range of standards for internally-assessed standards that schools can select to use in their teaching programmes"
From this work our moderators build a national picture where we as teachers see only our school picture.
"The focus of the moderation process is to ensure that student evidence for specific standards meets the criteria of the standard on a national basis"
They do not moderate your task unless they disagree with your judgements and the task is the reason for the grade disagreement.
If you have sent in a piece of student work that you feel is a grade boundary - err on the side of the student and put a note with the work that explains your thinking.
Sometimes you find yourself submitting work from the year prior. NZQA suggest you date the work being submitted and make it clear that the work is from a previous year.
Two standards highlighted were Physics (1.1) and Bivariate data (91036).
On reading the detail of each standard I noticed in Physics the students may conduct their investigation "with direction" and in 91036 students must identify and communicate how they will manage their variation.
In Physics students have a requirement for only 4 pieces of data and in 91036 they must collect enough to see a relationship and so it goes on.
The message I took from this is, if you are going to collect evidence for multiple standards from a single task it is important that all stakeholders sit down and develop a clear picture of what is required. The students need to be given the opportunity to meet the requirements of each standard at all levels.
Always use the NZQA subject page to find these documents. The links on the subject page take you to the most up to date versions of documents. (A google search sometimes takes us to out of date documents)
Students can achieve at different levels for each standard.
Tasks must give the students the opportunity for achieving at the excellence level in all standards the task is collecting evidence against. In my experience this is the most difficult part.
We came to the conclusion that it was very difficult to do this well and meet our obligations to the standards and process.
The conversation turned to the inference standards and in particular 91035.
Why is understanding of this standard is still problematic? From our general discussion we came to the conclusion that maybe not enough time is being given to stats in the junior programmes.
The role of NZQA and the external moderation process is outlined in the NCEA annual report published in May each year. See page 88 at this link
"Each year NZQA moderates approximately 100,000 individual pieces of student work that are submitted by schools as part of their external moderation process. The standards moderated are from the range of standards for internally-assessed standards that schools can select to use in their teaching programmes"
From this work our moderators build a national picture where we as teachers see only our school picture.
"The focus of the moderation process is to ensure that student evidence for specific standards meets the criteria of the standard on a national basis"
What moderators do & not do?
"The moderator will consider if the students’ evidence meets the requirements of the standard. The assessment tasks or activities are only moderated if the moderator has cause to consider that the assessment tasks or activities were the cause of the inappropriate awarding of the grade" (NCEA 2018, p 88)
They moderate assessor judgments in the student work you submit against the standard not your mark schedule.They do not moderate your task unless they disagree with your judgements and the task is the reason for the grade disagreement.
If you have sent in a piece of student work that you feel is a grade boundary - err on the side of the student and put a note with the work that explains your thinking.
Sometimes you find yourself submitting work from the year prior. NZQA suggest you date the work being submitted and make it clear that the work is from a previous year.
Unsure about your feedback?
You have an opportunity to query your moderation report rather than go through the appeal process. The purpose is to ensure you are clear about comments in your moderation reports. NZQA will respond to these queries as quickly as possible. Do this through your Principal's Nominee.
A question that was asked of us "if we are completing our internals as assignments/open book assessments should we need to be doing this?". What do you think?
We were also reminded that we cannot direct students when doing a resubmission. "we should not be prompting them for a message they did not deliver"
Read the annotations carefully.
The best exemplars you can use for students are from your own benchmarking folder. (folder of examples from your own students, in particular the student work that has been externally moderated). You should get students permission before using their work as exemplars for other students.
Resubmission & Reassessment
There is a lot of misinformation or what I call "bush law" sitting around resubmissions and reassessments. NZQA have published a series of Myth sheets which highlight some of the main issues surrounding Myth 4; Resubmissions & Myth 5; ReassessmentsA question that was asked of us "if we are completing our internals as assignments/open book assessments should we need to be doing this?". What do you think?
We were also reminded that we cannot direct students when doing a resubmission. "we should not be prompting them for a message they did not deliver"
About Exemplars
The exemplars on the NZQA website are designed to help teachers identify the grade boundary. They are not intended to be "models" for students.Read the annotations carefully.
The best exemplars you can use for students are from your own benchmarking folder. (folder of examples from your own students, in particular the student work that has been externally moderated). You should get students permission before using their work as exemplars for other students.
Collecting evidence from one task
As always in a workshop we deviated from the plan as questions came up. One such question related to collecting evidence against more than one standard from a single task.Two standards highlighted were Physics (1.1) and Bivariate data (91036).
On reading the detail of each standard I noticed in Physics the students may conduct their investigation "with direction" and in 91036 students must identify and communicate how they will manage their variation.
In Physics students have a requirement for only 4 pieces of data and in 91036 they must collect enough to see a relationship and so it goes on.
The message I took from this is, if you are going to collect evidence for multiple standards from a single task it is important that all stakeholders sit down and develop a clear picture of what is required. The students need to be given the opportunity to meet the requirements of each standard at all levels.
Always use the NZQA subject page to find these documents. The links on the subject page take you to the most up to date versions of documents. (A google search sometimes takes us to out of date documents)
Students can achieve at different levels for each standard.
Tasks must give the students the opportunity for achieving at the excellence level in all standards the task is collecting evidence against. In my experience this is the most difficult part.
We came to the conclusion that it was very difficult to do this well and meet our obligations to the standards and process.
The conversation turned to the inference standards and in particular 91035.
Why is understanding of this standard is still problematic? From our general discussion we came to the conclusion that maybe not enough time is being given to stats in the junior programmes.
91035 - Multivariate data.
Common misunderstandings raised in the discussion- Data sets do not have enough variables for students to make their own decisions around the variables they investigate. A multivariate data set should have a minimum of 2 categorical variables
- The students are taking a sample from the data set given. Students are not required to take a sample, they should be provided with the sample data
- Awarding a "holistic" grade if only one step of the statistical inquiry cycle is missing. All steps of the statistical inquiry cycle (bullet points in explanatory notes of the standard) must be evident in a students report.
- Assuming a generic discussion of sampling variation is enough for an E grade. Discussion of sample variation is not a requirement for 91035
- Awarding 91035 from 91264. It is difficult to integrate these standards using one task. There are some differences in the statistical enquiry cycle (See explanatory notes of each standard) and the type of question that needs to be posed.
- Analysis of the sample is descriptive of each group. Analysis should be comparative
- The question (& inference) is about the sample. The question and inference must be about the population and contain a world like "tends to".
- Quantity does not equal quality
- Generic statements like "back in the population". Students should be using the context of the investigation.
The question posed is about the population which should be included in the question ----->
When checking your students question before they do their analysis it is important to check that they have identified the population accurately
To achieve at Excellence student’s personal contextual knowledge (PCK) should be integrated throughout their report. It is particularly important that this is evident when students are posing a comparative question, discussing features of the distributions and when communicating findings such as their informal inference and supporting evidence in a conclusion.
When checking your students question before they do their analysis it is important to check that they have identified the population accurately
Contextual knowledge
It is important for us to have the conversations around the context in class so students know what the population is. Understanding the context and the population will help them better write an appropriate investigative question . It will also give them the opportunity to better integrate their personal contextual knowledge.To achieve at Excellence student’s personal contextual knowledge (PCK) should be integrated throughout their report. It is particularly important that this is evident when students are posing a comparative question, discussing features of the distributions and when communicating findings such as their informal inference and supporting evidence in a conclusion.
Students should not be seeing unfamiliar contexts in their assessment task
By relevant we mean relevant to the question and variables rather than simply context.
2. The complexity of the analysis: which broadly is...
This document written by Michelle Dalrymple on Census at school is essential reading for inference progressions. You should also read through Pip Arnolds document on What makes a good investigative question.
As usual we needed more time. I am looking forward to Episode Two.
Thank you Northland Maths association and NZQA these sessions are really valuable for the sector.
Read about hosting an NZQA "on-request" workshop in your region or on-line here
What are the expectations around contextual knowledge?
TKI Link
- At level 6 students should be working with contexts that are familiar to them.
- At level 7 students should be provided with relevant contextual knowledge about the situation under investigation.
- At level 8 students should be sourcing relevant contextual knowledge about the situation under investigation from places such as the internet, the school or local library, newspapers and magazines.
By relevant we mean relevant to the question and variables rather than simply context.
What is the step up across the inference standards
1. Contextual knowledge (see above)2. The complexity of the analysis: which broadly is...
- Level 5; visual comparison to see if one group tends to be bigger than another
- Level 6; visual comparison with a rule to see if one group tends to be bigger than another
- Level 7; visual analysis with an informal confidence interval used to see if the median of one group is likely to be bigger than the median of the other
- Level 8; quantifies the difference between the medians of two groups using Bootstrapping
This document written by Michelle Dalrymple on Census at school is essential reading for inference progressions. You should also read through Pip Arnolds document on What makes a good investigative question.
 |
| https://www.stat.auckland.ac.nz/~wild/StatThink/ |
As usual we needed more time. I am looking forward to Episode Two.
Thank you Northland Maths association and NZQA these sessions are really valuable for the sector.
Read about hosting an NZQA "on-request" workshop in your region or on-line here
Comments
Post a Comment